Methods And Materials For Classification Of Tissue Of Origin Of Tumor Samples

Abstract

The present invention provides a process for classification of cancers and tissues of origin through the analysis of the expression patterns of specific microRNAs and nucleic acid molecules relating thereto. Classification according to a microRNA tree-based expression framework allows optimization of treatment, and determination of specific therapy.

Description

FIELD OF THE INVENTION

[0001] The present invention relates to methods and materials for classification of cancers and the identification of their tissue of origin. Specifically the invention relates to microRNA molecules associated with specific cancers, as well as various nucleic acid molecules relating thereto or derived therefrom.

BACKGROUND OF THE INVENTION

[0002] microRNAs (miRs, miRNAs) are a novel class of non-coding, regulatory RNA genes.sup.1-3 which are involved in oncogenesis.sup.4 and show remarkable tissue-specificity.sup.5-7. They have emerged as highly tissue-specific biomarkers.sup.2,5,6 postulated to play important roles in encoding developmental decisions of differentiation. Various studies have tied microRNAs to the development of specific malignancies.sup.4. MicroRNAs are also stable in tissue, stored frozen or as formalin-fixed, paraffin-embedded (FFPE) samples, and in serum.

[0003] Hundreds of thousands of patients in the U.S. are diagnosed each year with a cancer that has already metastasized, without a clearly identified primary site. Oncologists and pathologists are constantly faced with a diagnostic dilemma when trying to identify the primary origin of a patient's metastasis. As metastases need to be treated according to their primary origin, accurate identification of the metastases' primary origin can be critical for determining appropriate treatment.

[0004] Once a metastatic tumor is found, the patient may undergo a wide range of costly, time consuming, and at times inefficient tests, including physical examination of the patient, histopathology analysis of the biopsy, imaging methods such as chest X-ray, CT and PET scans, in order to identify the primary origin of the metastasis.

[0005] Metastatic cancer of unknown primary (CUP) accounts for 3-5% of all new cancer cases, and as a group is usually a very aggressive disease with a poor prognosis.sup.10. The concept of CUP comes from the limitation of present methods to identify cancer origin, despite an often complicated and costly process which can significantly delay proper treatment of such patients. Recent studies revealed a high degree of variation in clinical management, in the absence of evidence based treatment for CUP.sup.11. Many protocols were evaluated.sup.12 but have shown relatively small benefit.sup.13. Determining tumor tissue of origin is thus an important clinical application of molecular diagnostics.sup.9.

[0006] Molecular classification studies for tumor tissue origin.sup.14-17 have generally used classification algorithms that did not utilize domain-specific knowledge: tissues were treated as a-priori equivalents, ignoring underlying similarities between tissue types with a common developmental origin in embryogenesis. An exception of note is the study by Shedden and co-workers.sup.18, that was based on a pathology classification tree. These studies used machine-learning methods that average effects of biological features (e.g., mRNA expression levels), an approach which is more amenable to automated processing but does not use or generate mechanistic insights.

[0007] Various markers have been proposed to indicate specific types of cancers and tumor tissue of origin. However, the diagnostic accuracy of tumor markers has not yet been defined. There is thus a need for a more efficient and effective method for diagnosing and classifying specific types of cancers.

SUMMARY OF THE INVENTION

[0008] The present invention provides specific nucleic acid sequences for use in the identification, classification and diagnosis of specific cancers and tumor tissue of origin. The nucleic acid sequences can also be used as prognostic markers for prognostic evaluation and determination of appropriate treatment of a subject based on the abundance of the nucleic acid sequences in a biological sample. The present invention provides a method for accurate identification of tumor tissue origin.

[0009] The invention is based in part on the development of a microRNA-based classifier for tumor classification. microRNA expression levels were measured in 1300 primary and metastatic tumor paraffin-embedded samples. microRNAs were profiled using a custom array platform. Using the custom array platform, a set of over 300 microRNAs was identified for the normalization of the array data and 65 microRNAs were used for the accurate classification of over 40 different tumor types. The accuracy of the assay exceeds 85%.

[0010] The findings demonstrate the utility of microRNA as novel biomarkers for the tissue of origin of a metastatic tumor. The classifier has wide biological as well as diagnostic applications.

[0011] According to a first aspect, the present invention provides a method of identifying a tissue of origin of a cancer, the method comprising obtaining a biological sample from a subject, measuring the relative abundance in said sample of nucleic acid sequences selected from the group consisting of SEQ ID NOS: 1-390, any combinations thereof, or a sequence having at least about 80% identity thereto; and comparing the measurement to a reference abundance of the nucleic acid by using a classifier algorithm, wherein the relative abundance of said nucleic acid sequences allows for the identification of the tissue of origin of said sample.

[0012] According to one aspect, the classifier algorithm is selected from the group consisting of decision tree classifier, K-nearest neighbor classifier (KNN), logistic regression classifier, nearest neighbor classifier, neural network classifier, Gaussian mixture model (GMM), Support Vector Machine (SVM) classifier, nearest centroid classifier, linear regression classifier and random forest classifier. According to one aspect, the sample is obtained from a subject with cancer of unknown primary (CUP), with a primary cancer or with a metastatic cancer.

[0013] According to certain embodiments, the cancer is selected from the group consisting of adrenocortical carcinoma; anus or skin squamous cell carcinoma; biliary tract adenocarcinoma; Ewing sarcoma; gastrointestinal stromal tumor (GIST); gastrointestinal tract carcinoid; renal cell carcinoma: chromophobe, clear cell and papillary; pancreatic islet cell tumor; pheochromocytoma; urothelial cell carcinoma (TCC); lung, head & neck, or esophagus squamous cell carcinoma (SCC); brain: astrocytic tumor, oligodendroglioma; breast adenocarcinoma; uterine cervix squamous cell carcinoma; chondrosarcoma; germ cell cancer; sarcoma; colorectal adenocarcinoma; liposarcoma; hepatocellular carcinoma (HCC); lung large cell or adenocarcinoma; lung carcinoid; pleural mesothelioma; lung small cell carcinoma; B-cell lymphoma; T-cell lymphoma; melanoma; malignant fibrous histiocytoma (MFH) or fibrosarcoma; osteosarcoma; ovarian primitive germ cell tumor; ovarian carcinoma; pancreatic adenocarcinoma; prostate adenocarcinoma; rhabdomyosarcoma; gastric or esophageal adenocarcinoma; synovial sarcoma; non-seminomatous testicular germ cell tumor; seminomatous testicular germ cell tumor; thymoma/thymic carcinoma; follicular thyroid carcinoma; medullary thyroid carcinoma; and papillary thyroid carcinoma.

[0014] The invention further provides a method for identifying a cancer of germ cell origin, comprising measuring the relative abundance of SEQ ID NO: 55 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of germ cell origin. According to some embodiments the germ cell is selected from the group consisting of an ovarian primitive cell and a testis cell. According to some embodiments the group of nucleic acid furthers consists of SEQ ID NOS: 29, 62 or a sequence having at least about 80% identity thereto, and the abundance of said nucleic acid sequence is indicative of a testis cell cancer origin selected from the group consisting of seminomatous testicular germ cell and non-seminomatous testicular germ cell.

[0015] The invention further provides a method for identifying a cancer origin selected from the group consisting of biliary tract adenocarcinoma and hepatocellular carcinoma, comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 9, 29 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer origin selected from the group consisting of biliary tract adenocarcinoma and hepatocellular carcinoma.

[0016] The invention further provides a method for identifying a cancer of brain origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 156, 66, 68 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of brain origin.

[0017] According to some embodiments the group of nucleic acid furthers consists of SEQ ID NOS: 40, 60 or a sequence having at least about 80% identity thereto, and wherein the abundance of said nucleic acid sequence is indicative of a brain cancer origin selected from the group consisting of oligodendroglioma and astrocytoma.

[0018] The invention further provides a method for identifying a cancer of prostate adenocarcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of prostate adenocarcinoma origin.

[0019] The invention further provides a method for identifying a cancer of breast adenocarcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 27, 35, 14, 21, 32, 51, 7, 25, 50, 11, 148, 4, 49, 67 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of breast adenocarcinoma origin.

[0020] The invention further provides a method for identifying a cancer of ovarian carcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 27, 35, 14, 21, 32, 51, 7, 25, 4, 39, 50, 11, 148, 49, 67, 57, 34 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of an ovarian carcinoma origin.

[0021] The invention further provides a method for identifying a cancer of thyroid carcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 11, 148, 4 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of thyroid carcinoma origin.

[0022] According to some embodiments the group of nucleic acid furthers consists of SEQ ID NOS: 17, 34 or a sequence having at least about 80% identity thereto, and wherein said thyroid carcinoma origin is selected from the group consisting of follicular and papillary.

[0023] The invention further provides a method for identifying a cancer origin selected from the group consisting of lung large cell and lung adenocarcinoma, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 11, 148, 4, 49, 67, 57, 34 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer origin selected from the group consisting of lung large cell and lung adenocarcinoma.

[0024] The invention further provides a method for identifying a cancer origin selected from the group consisting of lung large cell and lung adenocarcinoma, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 11, 148, 4, 49, 67, 57, 34 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer origin selected from the group consisting of lung large cell and lung adenocarcinoma.

[0025] The invention further provides a method for identifying a cancer of thymic carcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 50, 4, 39, 3, 34 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of a thymic carcinoma origin.

[0026] The invention further provides a method for identifying a cancer origin selected from the group consisting of a urothelial cell carcinoma and squamous cell carcinoma, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 50, 4, 39, 3, 34, 69, 24, 44 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of is indicative of a cancer origin selected from the group consisting of urothelial cell carcinoma and squamous cell carcinoma.

[0027] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 1, 5, 54 or a sequence having at least about 80% identity thereto, and wherein the abundance of said nucleic acid sequence is indicative of squamous-cell-carcinoma origin selected from the group consisting of uterine cervix squamous-cell-carcinoma and non uterine cervix squamous cell carcinoma.

[0028] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 11, 23 or a sequence having at least about 80% identity thereto in said sample, and wherein the abundance of said nucleic acid sequence is indicative of a non-uterine cervix squamous cell carcinoma origin selected from the group consisting of anus or skin squamous cell carcinoma; and lung, head & neck, and esophagus squamous cell carcinoma.

[0029] The invention further provides a method for identifying a cancer origin selected from melanoma and lymphoma, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 47, 50 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer origin selected from the group consisting of melanoma and lymphoma.

[0030] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 35, 48 or a sequence having at least about 80% identity thereto, and wherein the abundance of said nucleic acid sequence is indicative of a lymphoma cancer origin selected from the group consisting of B-cell lymphoma and T-cell lymphoma.

[0031] The invention further provides a method for identifying a cancer of lung small cell carcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20, 45 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of lung small cell carcinoma origin.

[0032] The invention further provides a method for identifying a cancer of medullary thyroid carcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20, 45, 40, 67, 68 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of medullary thyroid carcinoma origin.

[0033] The invention further provides a method for identifying a cancer of lung carcinoid origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20, 45, 40, 67, 68, 64, 53, 37 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of lung carcinoid origin.

[0034] The invention further provides a method for identifying a cancer origin selected from the group consisting of gastrointestinal tract carcinoid and pancreatic islet cell tumor, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20, 45, 40, 67, 68, 64, 53, 37, 34, 18 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer origin selected from the group consisting of gastrointestinal tract carcinoid and pancreatic islet cell tumor.

[0035] The invention further provides a method for identifying a cancer origin selected from the group consisting of gastric and esophageal adenocarcinoma, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 42, 36, 146 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer origin elected from the group consisting of gastric and esophageal adenocarcinoma.

[0036] The invention further provides a method for identifying a cancer of colorectal adenocarcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 42, 36, 146, 20, 43 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of colorectal adenocarcinoma origin.

[0037] The invention further provides a method for identifying a cancer origin selected from the group consisting of pancreatic adenocarcinoma and biliary tract adenocarcinoma, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 42, 36, 146, 20, 4351, 49, 16, or a sequence having at least about 80% identity thereto, and wherein the abundance of said nucleic acid sequence is indicative of a cancer origin selected from the group consisting of pancreatic adenocarcinoma or biliary tract adenocarcinoma.

[0038] The invention further provides a method for identifying a cancer of renal cell carcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of renal cell carcinoma origin.

[0039] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 36, 147 or a sequence having at least about 80% identity thereto, and wherein the abundance of said nucleic acid sequence is indicative of a chromophobe renal cell carcinoma origin.

[0040] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 49, 9 or a sequence having at least about 80% identity thereto, and wherein the abundance of said nucleic acid sequence is indicative of a renal cell carcinoma origin selected from the group consisting of clear cell and papillary.

[0041] The invention further provides a method for identifying a cancer of pheochromocytoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65, 56 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of pheochromocytoma origin.

[0042] The invention further provides a method for identifying a cancer of adrenocortical origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65, 56, 31, 38, 61 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of adrenocortical origin.

[0043] The invention further provides a method for identifying a cancer of gastrointestinal stromal tumor origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65, 56, 31, 38, 61, 14, 45 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of gastrointestinal stromal tumor origin.

[0044] The invention further provides a method for identifying a cancer origin selected from the group consisting of pleural mesothelioma and sarcoma, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65, 56, 31, 38, 61, 14, 45, 35, 10, 5 or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer origin selected from the group consisting of pleural mesothelioma and sarcoma.

[0045] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 3, 40, 15 or a sequence having at least about 80% identity thereto, and wherein said sarcoma is synovial sarcoma.

[0046] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 3, 40, 15, 12, 58 or a sequence having at least about 80% identity thereto, and wherein said sarcoma is chondrosarcoma.

[0047] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 3, 40, 15, 12, 58, 36, 26 or a sequence having at least about 80% identity thereto, and wherein said sarcoma is liposarcoma.

[0048] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 3, 40, 15, 12, 58, 36, 26, 21, 25, 49 or a sequence having at least about 80% identity thereto and wherein said sarcoma is selected from the group consisting of Ewing sarcoma and osteosarcoma.

[0049] According to some embodiments the group of nucleic acid further consists of SEQ ID NOS: 3, 40, 15, 12, 58, 36, 26, 21, 59, 39, 33 or a sequence having at least about 80% identity thereto and wherein said sarcoma is selected from the group consisting of rhabdomyosarcoma; and malignant fibrous histiocytoma and fibrosarcoma.

[0050] According to another aspect, the present invention provides a method of distinguishing between cancers of different origins, said method comprising:

[0051] (a) obtaining a biological sample from a subject;

[0052] (b) measuring the relative abundance in said sample of nucleic acid sequences selected from the group consisting of SEQ ID NOS: 1-390 or a sequence having at least about 80% identity thereto; and

[0053] (c) comparing said measurement to a reference abundance of said nucleic acid by using a classifier algorithm;

[0054] wherein the relative abundance of said nucleic acid sequence in said sample allows for distinguishing between cancers of different origins.

[0055] According to some embodiments the measurement of the relative abundance of SEQ ID NOS: 372, 233, 55, 200, 201 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from a germ-cell tumor and a cancer originating from the group consisting of non-germ-cell tumors.

[0056] According to some embodiments the measurement of the relative abundance of SEQ ID NOS: 6, 30, 13 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from hepatobiliary tumors and a cancer originating from the group consisting of non-germ-cell non-hepatobiliary tumors.

[0057] According to some embodiments the measurement of the relative abundance of SEQ ID NOS: 28, 29, 231, 9 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from liver tumors and a cancer originating from biliary-tract carcinomas.

[0058] According to some embodiments the measurement of the relative abundance of SEQ ID NOS: 46, 5, 12, 30, 29, 28, 32, 13, 152, 49 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from the group consisting of tumors from an epithelial origin and a cancer originating from the group consisting of tumors from a non-epithelial origin.

[0059] According to some embodiments the measurement of the relative abundance of SEQ ID NOS: 164, 168, 170, 16, 198, 50, 176, 186, 11, 158, 20, 155, 231, 4, 8, 46, 3, 2, 7 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from the group consisting of melanoma and lymphoma and a cancer originating from the group consisting of all other non-epithelial tumors.

[0060] According to some embodiments the measurement of the relative abundance of SEQ ID NOS: 159, 66, 225, 187, 162, 161, 68, 232, 173, 11, 8, 174, 155, 231, 4, 182, 181, 37 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from brain tumors and a cancer originating from the group consisting of all non-brain, non-epithelial tumors.

[0061] According to some embodiments the measurement of the relative abundance of SEQ ID NOS: 40, 208, 60, 153, 230, 228, 147, 34, 206, 35, 52, 25, 229, 161, 187, 179 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from astrocytoma and a cancer originating from oligodendroglioma.

[0062] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 56, 65, 25, 175, 152, 155, 32, 49, 35, 181, or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from the group consisting of neuroendocrine tumors and a cancer originating from the group consisting of all non-neuroendocrine, epithelial tumors.

[0063] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 27, 177, 4, 32, 35 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from the group consisting of gastrointestinal epithelial tumors and a cancer originating from the group consisting of non-gastrointestinal epithelial tumors.

[0064] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 56, 199, 14, 15, 165, 231, 36, 154, 21, 49 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from prostate tumors and a cancer originating from the group consisting of all other non-gastrointestinal epithelial tumors.

[0065] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 222, 62, 29, 28, 211, 214, 227, 215, 218, 152, 216, 212, 224, 13, 194, 192, 221, 217, 205, 219, 32, 193, 223, 220, 210, 209, 213, 163, 30 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from seminoma and a cancer originating from the group consisting of non-seminoma testis-tumors.

[0066] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 42, 32, 36, 178, 243, 242, 49, 240, 57, 11, 46, 17, 47, 51, 7, 8, 154, 190, 157, 196, 197, or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from the group consisting of squamous cell carcinoma, transitional cell carcinoma and thymoma, and a cancer originating from the group consisting of non gastrointestinal adenocarcinoma tumors.

[0067] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 56, 46, 25, 152, 50, 45, 191, 181, 179, 49, 32, 42, 184, 40, 147, 236, 57, 203, 36, or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from breast adenocarcinoma, and a cancer originating from the group consisting of squamous cell carcinoma, transitional cell carcinoma, thymomas and ovarian carcinoma.

[0068] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 253, 32, 4, 39, 10, 46, 5, 226, 2, 195, 32, 185, 11, 168, 184, 16, 242, 12, 237, 243, 250, 49, 246, 167 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from ovarian carcinoma, and a cancer originating from the group consisting of squamous cell carcinoma, transitional cell carcinoma and thymomas.

[0069] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 11, 147, 17, 157, 40, 8, 49, 9, 191, 205, 207, 195, 51, 46, 45, 52, 234, 231, 21, 169, 43, 3, 196, 154, 390, 171, 255, 197, 190, 189, 39, 7, 48, 47, 32, 36, 4, 178, 37, 181, 25, 183, 182, 35, 240, 57, 242, 204, 236, 176, 158, 148, 206, 50, 20, 34, 186, 239, 251, 244, 24, 188, 172, 238 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from thyroid carcinoma, and a cancer originating from the group consisting of breast adenocarcinoma, lung large cell carcinoma, lung adenocarcinoma and ovarian carcinoma.

[0070] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 249, 180, 65, 235, 241, 248, 254, 247, 160, 243, 245, 252, 17, 49, 166, 225, 168, 34 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from follicular thyroid carcinoma and a cancer originating from papillary thyroid carcinoma.

[0071] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 32, 56, 50, 45, 25, 253, 152, 9, 46, 191, 178, 49, 40, 10, 147, 4, 36, 228, 236, 230, 189, 240, 67, 202, 17 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from breast adenocarcinoma and a cancer originating from the group consisting of lung adenocarcinoma and ovarian carcinoma.

[0072] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 56, 11, 168, 16, 237, 21, 52, 12, 154, 279, 9, 39, 47, 23, 50, 167, 383, 34, 35, 388, 5, 359, 245, 254, 10, 240, 236, 202, 4, 25, 203, 231, 20, 158, 186, 258, 244, 172, 2, 235, 256, 28, 277, 296, 374, 153, 181 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from lung adenocarcinoma and a cancer originating from ovarian carcinoma.

[0073] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 161, 164, 22, 53, 285, 3, 152, 191, 154, 21, 206, 174, 19, 45, 171, 179, 8, 296, 284, 18, 51, 258, 49, 184, 35, 34, 37, 42, 228, 15, 14, 242, 230, 253, 36, 182, 293, 292, 4, 294, 297, 354, 377, 189, 30, 386, 249, 5, 274 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from thymic carcinoma and a cancer originating from the group consisting of transitional cell carcinoma and squamous cell carcinoma.

[0074] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 69, 28, 280, 13, 191, 152, 29, 175, 30, 204, 4, 24, 5, 329, 273, 170, 184, 26, 231, 368, 37, 16, 169, 155, 35, 40, 17 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from transitional cell carcinoma and a cancer originating from the group consisting of squamous cell carcinoma.

[0075] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 164, 5, 231, 54, 1, 242, 372, 249, 167, 254, 354, 381, 380, 245, 358, 364, 240, 11, 378 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between squamous cell carcinoma cancers originating from the uterine cervix, and squamous cell carcinoma cancers originating from the group consisting of anus and skin, lung, head & neck and esophagus.

[0076] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 305, 184, 41, 183, 49, 382, 235, 291, 181, 5, 296, 289, 206, 338, 334, 25, 11, 19, 198, 23 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between squamous cell carcinoma cancers originating from the group consisting of anus and skin, and between squamous cell carcinoma cancers originating from the group consisting of lung, head & neck and esophagus.

[0077] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 4, 11, 46, 8, 274, 169, 36, 47, 363, 231, 303, 349, 10, 7, 3, 16, 164, 170, 168, 198, 50, 245, 365, 45, 382, 259, 296, 364, 314, 12 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from melanoma and a cancer originating from lymphoma.

[0078] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 11, 191, 48, 35, 228 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from B-cell lymphoma and a cancer originating from T-cell lymphoma.

[0079] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 158, 20, 176, 186, 148, 36, 51, 172, 260, 265, 67, 188, 277, 284, 302, 68, 168, 242, 204, 162, 177, 27, 65, 263, 155, 191, 190, 45, 59, 43, 56, 266, 14, 15, 8, 7, 39, 189, 249, 231, 293, 2 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from lung small cell carcinoma and a cancer originating from the group consisting of lung carcinoid, medullary thyroid carcinoma, gastrointestinal tract carcinoid and pancreatic islet cell tumor.

[0080] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 159, 40, 147, 11, 311, 4, 8, 231, 301, 297, 68, 67, 265, 36 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from medullary thyroid carcinoma and a cancer originating from other neuroendocrine tumors selected from the group consisting of lung carcinoid, gastrointestinal tract carcinoid and pancreatic islet cell tumor.

[0081] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 331, 162, 59, 326, 306, 350, 317, 155, 325, 318, 339, 264, 332, 262, 336, 324, 322, 330, 321, 263, 309, 53, 320, 275, 352, 312, 355, 367, 269, 64, 308, 175, 190, 54, 302, 152, 301, 266, 47, 313, 359, 65, 307, 191, 242, 4, 147, 40, 372, 168, 16, 182, 167, 356, 148, 382, 37, 364, 35 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from lung carcinoid tumors, and a cancer originating from gastrointestinal neuroendocrine tumors selected from the group consisting of gastrointestinal tract carcinoid and pancreatic islet cell tumor.

[0082] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 263, 288, 18, 286, 162, 225, 287, 206, 205, 296, 258, 313, 377, 373, 256, 153, 259, 265, 303, 268, 267, 165, 15, 272, 14, 202, 236, 203, 4, 168, 310, 298, 27, 29, 34, 228, 3, 349, 35, 26 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from pancreatic islet cell tumors and a Gastrointestinal neuroendocrine carcinoid cancer originating from the group consisting of small intestine and duodenum; appendicitis, stomach and pancreas.

[0083] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 36, 267, 268, 165, 15, 14, 356, 167, 372, 272, 370, 42, 41, 146 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between adenocarcinoma tumors of the gastrointestinal system originating from:

[0084] the group consisting of gastric and esophageal adenocarcinoma, and

[0085] the group consisting of cholangiocarcinoma or adenocarcinoma of the extrahepatic biliary tract, pancreatic adenocarcinoma and colorectal adenocarcinoma.

[0086] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 42, 184, 67, 158, 20, 186, 284, 389, 203, 240, 236, 146, 204, 43, 176, 202, 49, 46, 38, 363 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from colorectal adenocarcinoma and a cancer originating from the group consisting of adenocarcinoma of biliary tract or pancreas.

[0087] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 49, 11, 13, 373, 154, 5, 30, 45, 178, 147, 274, 16, 40, 21, 43, 253, 245, 256, 12, 374, 379, 180, 153, 51, 52, 1, 295, 257, 385, 293, 294 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from pancreatic adenocarcinoma, and a cancer originating from the group consisting of cholangiocarcinoma or adenocarcinoma of the extrahepatic biliary tract.

[0088] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 29, 28, 30, 46, 49, 195, 152, 175, 47, 4, 387, 196, 177, 375, 27, 304, 40, 191, 147, 35, 16, 34, 5, 155, 181, 312, 183, 182, 320, 59, 38, 324, 323, 37, 322, 325, 19, 42, 334, 265, 22 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from:

[0089] renal cell tumors selected from the group consisting of chromophobe renal cell carcinoma, clear cell renal cell carcinoma and papillary renal cell carcinoma, and

[0090] the group consisting of sarcomas, adrenal tumors and pleural mesothelioma.

[0091] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 65, 56, 11, 162, 59, 331, 350, 155, 335, 159, 336, 332, 263, 306, 339, 337, 275, 301, 276, 330, 317, 309, 45, 318, 324, 352, 191, 262, 269, 313, 19, 367, 326, 325, 322, 327, 190, 261, 321, 360, 353, 312, 371, 5, 328, 205, 183, 38, 181, 37, 40, 182, 147, 17, 42, 382, 34, 18, 3 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from pheochromocytoma, and a cancer originating from the group consisting of all sarcoma, adrenal carcinoma and mesothelioma tumors.

[0092] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 61, 333, 31, 347, 346, 344, 345, 387, 334, 351, 324, 326, 269, 155, 320, 322, 59, 318, 325, 245, 254, 331, 275, 180, 355, 370, 323, 312, 178, 249, 183, 181, 38, 182, 37, 3, 25 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from adrenal carcinoma and a cancer originating from the group consisting of mesothelioma and sarcoma tumors.

[0093] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 165, 14, 15, 333, 272, 270, 45, 301, 191, 46, 195, 266, 190, 19, 334, 155, 25, 147, 40, 34 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from a gastrointestinal stromal tumor and a cancer originating from the group consisting of mesothelioma and sarcoma tumors.

[0094] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 13, 30, 361, 280, 362, 147, 40, 291, 387, 290, 299, 152, 178, 303, 242, 49, 11, 35, 34, 36, 206, 16, 170, 177, 17 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from a chromophobe renal cell carcinoma tumor and a cancer originating from the group consisting of clear cell and papillary renal cell carcinoma tumors.

[0095] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 344, 382, 9, 338, 29, 49, 28, 195, 46, 4, 11, 254 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a renal carcinoma cancer originating from a clear cell tumor and a cancer originating from a papillary tumor.

[0096] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 49, 35, 17, 34, 25, 36, 168, 170, 26, 4, 190, 46, 10, 240, 43, 39, 385, 63, 202, 181, 37, 5, 183, 182, 38, 206, 296, 1 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from pleural mesothelioma and a cancer originating from the group consisting of sarcoma tumors.

[0097] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 152, 29, 159, 28, 339, 275, 352, 19, 320, 155, 262, 38, 37, 182, 331, 317, 323, 355, 3, 282, 312, 181, 269, 318, 59, 266, 322, 8, 324, 10, 40, 147, 169, 205, 34, 168, 14, 15, 12, 46, 255, 39, 23, 190, 236, 386, 379, 202 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from a synovial sarcoma and a cancer originating from the group consisting of other sarcoma tumors.

[0098] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 12, 271, 206, 333, 11, 58, 36, 18, 178, 293, 189, 382, 381, 240, 249, 5, 377, 235, 17, 20, 385, 384, 46, 283 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from chondrosarcoma and a cancer originating from the group consisting of other non-synovial sarcoma tumors.

[0099] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 295, 205, 25, 26, 231, 183, 42, 254, 168, 64, 14, 178, 15, 39, 36, 154, 265, 174, 384, 67 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from liposarcoma and a cancer originating from the group consisting of other non chondrosarcoma and non synovial sarcoma tumors.

[0100] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 22, 154, 21, 174, 205, 158, 186, 148, 20, 59, 8, 183, 231 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from:

[0101] the group consisting of Ewing sarcoma and osteosarcoma, and

[0102] the group consisting of rhabdomyosarcoma, malignant fibrous histiocytoma and fibrosarcoma.

[0103] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 155, 179, 43, 208, 278, 17, 385, 174, 5, 52, 257, 366, 48, 49, 12, 25, 169, 34, 35, 23, 384, 189, 377, 265, 294, 293, 292 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from Ewing sarcoma and a cancer originating from osteosarcoma.

[0104] According to some embodiments, measurement of the relative abundance of SEQ ID NOS: 33, 268, 267, 333, 276, 319, 306, 320, 334, 323, 300, 281, 59, 339, 316, 176, 348, 352, 349, 67, 357, 315, 343, 342, 355, 340, 344, 10, 341, 331, 20, 277, 318, 158, 265, 284, 36, 183, 40, 63, 147, 43, 289, 52, 190, 4, 5, 39, 169, 208 or a sequence having at least about 80% identity thereto in said sample allows for distinguishing between a cancer originating from rhabdomyosarcoma and a cancer originating from the group consisting of malignant fibrous histiocytoma and fibrosarcoma.

[0105] According to some aspects of the invention the biological sample is selected from the group consisting of bodily fluid, a cell line, a tissue sample, a biopsy sample, a needle biopsy sample, a fine needle biopsy (FNA) sample, a surgically removed sample, and a sample obtained by tissue-sampling procedures such as endoscopy, bronchoscopy, or laparoscopic methods. According to some embodiments, the tissue is a fresh, frozen, fixed, wax-embedded or formalin-fixed paraffin-embedded (FFPE) tissue.

[0106] According to additional aspects of the invention the nucleic acid sequence relative abundance is determined by a method selected from the group consisting of nucleic acid hybridization and nucleic acid amplification. According to some embodiments, the nucleic acid hybridization is performed using a solid-phase nucleic acid biochip array or in situ hybridization. According to some embodiments, the nucleic acid amplification method is real-time PCR. According to some embodiments, the real-time PCR comprises forward and reverse primers. According to additional embodiments, the real-time PCR method further comprises a probe. According to additional embodiments, the probe comprises a sequence selected from the group consisting of a sequence that is complementary to a sequence selected from SEQ ID NOS: 1-390; a fragment thereof and a sequence having at least about 80% identity thereto.

[0107] According to another aspect, the present invention provides a kit for cancer origin identification, the kit comprising a probe comprising a sequence selected from the group consisting of a sequence that is complementary to a sequence selected from SEQ ID NOS: 1-390; a fragment thereof and a sequence having at least about 80% identity thereto.

[0108] These and other embodiments of the present invention will become apparent in conjunction with the figures, description and claims that follow.

BRIEF DESCRIPTION OF THE DRAWINGS

[0109] FIGS. 1A-1F demonstrate the structure of the binary decision-tree classifier, with 45 nodes and 46 leaves. Each node is a binary decision between two sets of samples, those to the left and right of the node. A series of binary decisions, starting at node #1 and moving downwards, lead to one of the possible tumor types, which are the "leaves" of the tree. A sample which is classified to the right branch at node #1 continues to node #2, otherwise it continues to node #11. A sample which is classified to the right branch at node #2 continues to node #4, otherwise it continues to node #3. A sample that reaches node #3, is further classified to either the left branch at node #3, and is assigned to the "hepatocellular carcinoma" class, or to the right branch at node #3, and is assigned to the "biliary tract adenocarcinoma" class.

[0110] Decisions are made at consecutive nodes using microRNA expression levels, until an end-point ("leaf" of the tree) is reached, indicating the predicted class for this sample. In specifying the tree structure, clinico-pathological considerations were combined with properties observed in the training set data.

[0111] FIGS. 2A-2D demonstrate binary decisions at node #1 of the decision-tree. When training a decision algorithm for a given node, only samples from classes which are possible outcomes of this node are used for training. The "non germ cell" classes (right branch at node #1); are easily distinguished from tumors of the "germ cell" classes (left branch at node #1) using the expression levels of hsa-miR-373 (SEQ ID NO: 233, 2A), hsa-miR-372 (SEQ ID NO: 55, 2B), hsa-miR-371-3p (SEQ ID NO: 200, 2C), and hsa-miR-371-5p (SEQ ID NO: 201, 2D). The boxplot presentations comparing distribution of the expression of the statistically significant miRs in tumor samples from the "germ cell" classes (left box) and "non germ cell" classes (right box). The line in the box indicates the median value. The box contains 50% of the data and the horizontal lines and crosses (outliers) show the full range of signals in this group.

[0112] FIG. 3 demonstrates binary decisions at node #3 of the decision-tree. Tumors of hepatocellular carcinoma (HCC) origin (left branch at node #3, marked by squares) are easily distinguished from tumors of biliary tract adenocarcinoma origin (right branch at node #3, marked by diamonds) using the expression levels of hsa-miR-200b (SEQ ID NO: 29, y-axis) and hsa-miR-126 (SEQ ID NO: 9, x-axis).

[0113] FIG. 4 demonstrates binary decisions at node #4 of the decision-tree. Tumors originating in epithelial (diamonds) are easily distinguished from tumors of non-epithelial origin (squares) using the expression levels of hsa-miR-30a (SEQ ID NO: 46, y-axis) and hsa-miR-200c (SEQ ID NO: 30, x-axis).

[0114] FIG. 5 demonstrates binary decisions at node #5 of the decision-tree. Tumors originating in the lymphoma or melanoma (diamonds) are easily distinguished from tumors of non epithelial, non lymphoma/melanoma origin (squares) using the expression levels of hsa-miR-146a (SEQ ID NO: 16, y-axis), hsa-miR-30a (SEQ ID NO: 46, x-axis) and hsa-let-7e (SEQ ID NO: 2, z-axis).

[0115] FIG. 6 demonstrates binary decisions at node #6 of the decision-tree. Tumors originating in the brain (left branch at node #6, marked by diamonds) are easily distinguished from tumors of non epithelial, non brain (right branch at node #6, marked by squares) using the expression levels of hsa-miR-9* (SEQ ID NO: 66, y-axis) and hsa-miR-92b (SEQ ID NO: 68, x-axis).

[0116] FIG. 7 demonstrates binary decisions at node #7 of the decision-tree. Tumors originating in astrocytoma (right branch at node #7, marked by diamonds) are easily distinguished from tumors of oligodendroglioma origins (left branch at node #7, marked by squares) using the expression levels of hsa-miR-497 (SEQ ID NO: 60, y-axis) and hsa-miR-222 (SEQ ID NO: 40, x-axis).

[0117] FIG. 8 demonstrates binary decisions at node #8 of the decision-tree. Tumors originating in the neuroendocrine (diamonds) are easily distinguished from tumors of epithelial, origin (squares) using the expression levels of hsa-miR-193a-3p (SEQ ID NO: 181, y-axis), hsa-miR-7 (SEQ ID NO: 65, x-axis) and hsa-miR-375 (SEQ ID NO: 56, z-axis).

[0118] FIG. 9 demonstrates binary decisions at node #9 of the decision-tree. Tumors originating in gastro-intestinal (GI) (left branch at node #9, marked by diamonds) are easily distinguished from tumors of non GI origins (right branch at node #9, marked by squares) using the expression levels of hsa-miR-21* (SEQ ID NO: 35, y-axis) and hsa-miR-194 (SEQ ID NO: 27, x-axis).

[0119] FIG. 10 demonstrates binary decisions at node #10 of the decision-tree. Tumors originating in prostate adenocarcinoma (left branch at node #10, marked by diamonds) are easily distinguished from tumors of non prostate origins (right branch at node #10, marked by squares) using the expression levels of hsa-miR-181a (SEQ ID NO: 21, y-axis) and hsa-miR-143 (SEQ ID NO: 14, x-axis).

[0120] FIG. 11 demonstrates binary decisions at node #12 of the decision-tree. Tumors originating in seminomatous testicular germ cell (left branch at node #12, marked by diamonds) are easily distinguished from tumors of non seminomatous origins (right branch at node #12, marked by squares) using the expression levels of hsa-miR-516a-5p (SEQ ID NO: 62, y-axis) and hsa-miR-200b (SEQ ID NO: 29, x-axis).

[0121] FIG. 12 demonstrates binary decisions at node #16 of the decision-tree. Tumors originating in thyroid carcinoma (diamonds) are easily distinguished from tumors of adenocarcinoma of the lung, breast and ovarian origin (squares) using the expression levels of hsa-miR-93 (SEQ ID NO: 148, y-axis), hsa-miR-138 (SEQ ID NO: 11, x-axis) and hsa-miR-10a (SEQ ID NO: 4, z-axis).

[0122] FIG. 13 demonstrates binary decisions at node #17 of the decision-tree. Tumors originating in follicular thyroid carcinoma (left branch at node #17, marked by diamonds) are easily distinguished from tumors of papillary thyroid carcinoma origins (right branch at node #17, marked by squares) using the expression levels of hsa-miR-21 (SEQ ID NO: 34, y-axis) and hsa-miR-146b-5p (SEQ ID NO: 17, x-axis).

[0123] FIG. 14 demonstrates binary decisions at node #18 of the decision-tree. Tumors originating in breast (diamonds) are easily distinguished from tumors of lung and ovarian origin (squares) using the expression levels of hsa-miR-92a (SEQ ID NO: 67, y-axis), hsa-miR-193a-3p (SEQ ID NO: 25, x-axis) and hsa-miR-31 (SEQ ID NO: 49, z-axis).

[0124] FIG. 15 demonstrates binary decisions at node #19 of the decision-tree. Tumors originating in lung adenocarcinoma (diamonds) are easily distinguished from tumors of ovarian carcinoma origin (squares) using the expression levels of hsa-miR-21 (SEQ ID NO: 34, y-axis), hsa-miR-378 (SEQ ID NO: 57, x-axis) and hsa-miR-138 (SEQ ID NO: 11, z-axis).

[0125] FIG. 16 demonstrates binary decisions at node #20 of the decision-tree. Tumors originating in thymic carcinoma (left branch at node #20, marked by diamonds) are easily distinguished from tumors of urothelial carcinoma, transitional cell carcinoma (TCC) carcinoma and squamous cell carcinoma (SCC) origins (right branch at node #20, marked by squares) using the expression levels of hsa-miR-21 (SEQ ID NO: 34, y-axis) and hsa-miR-100 (SEQ ID NO: 3, x-axis).

[0126] FIG. 17 demonstrates binary decisions at node #22 of the decision-tree. Tumors originating in SCC of the uterine cervix (diamonds) are easily distinguished from tumors of other SCC origin (squares) using the expression levels of hsa-miR-361-5p (SEQ ID NO: 54, y-axis), hsa-let-7c (SEQ ID NO: 1, x-axis) and hsa-miR-10b (SEQ ID NO: 5, z-axis).

[0127] FIG. 18 demonstrates binary decisions at node #24 of the decision-tree. Tumors originating in melanoma (diamonds) are easily distinguished from tumors of lymphoma origin (squares) using the expression levels of hsa-miR-342-3p (SEQ ID NO: 50, y-axis) and hsa-miR-30d (SEQ ID NO: 47, x-axis).

[0128] FIG. 19 demonstrates binary decisions at node #27 of the decision-tree. Tumors originating in thyroid carcinoma, medullary (diamonds) are easily distinguished from tumors of other neuroendocrine origin (squares) using the expression levels of hsa-miR-92b (SEQ ID NO: 68, y-axis), hsa-miR-222 (SEQ ID NO: 40, x-axis) and hsa-miR-92a (SEQ ID NO: 67, z-axis).

[0129] FIG. 20 demonstrates binary decisions at node #30 of the decision-tree. Tumors originating in gastric or esophageal adenocarcinoma (diamonds) are easily distinguished from tumors of other GI adenocarcinoma origin (squares) using the expression levels of hsa-miR-1201 (SEQ ID NO: 146, y-axis), hsa-miR-224 (SEQ ID NO: 42, x-axis) and hsa-miR-210 (SEQ ID NO: 36, z-axis).

[0130] FIG. 21 demonstrates binary decisions at node #31 of the decision-tree. Tumors originating in colorectal adenocarcinoma (diamonds) are easily distinguished from tumors of adenocarcinoma of biliary tract or pancreas origin (squares) using the expression levels of hsa-miR-30a (SEQ ID NO: 46, y-axis), hsa-miR-17 (SEQ ID NO: 20, x-axis) and hsa-miR-29a (SEQ ID NO: 43, z-axis).

[0131] FIG. 22 demonstrates binary decisions at node #33 of the decision-tree. Tumors originating in kidney (diamonds) are easily distinguished from tumors of adrenal, mesothelioma and sarcoma origin (squares) using the expression levels of hsa-miR-200b (SEQ ID NO: 29, y-axis), hsa-miR-30a (SEQ ID NO: 46, x-axis) and hsa-miR-149 (SEQ ID NO: 19, z-axis).

[0132] FIG. 23 demonstrates binary decisions at node #34 of the decision-tree. Tumors originating in pheochromocytoma (diamonds) are easily distinguished from tumors of adrenal, mesothelioma and sarcoma origin (squares) using the expression levels of hsa-miR-375 (SEQ ID NO: 56, y-axis) and hsa-miR-7 (SEQ ID NO: 65, x-axis).

[0133] FIG. 24 demonstrates binary decisions at node #44 of the decision-tree. Tumors originating in Ewing sarcoma (diamonds) are easily distinguished from tumors of osteosarcoma origin (squares) using the expression levels of hsa-miR-31 (SEQ ID NO: 49, y-axis) and hsa-miR-193a-3p (SEQ ID NO: 25, x-axis).

[0134] FIG. 25 demonstrates binary decisions at node #45 of the decision-tree. Tumors originating in Rhabdomyosarcoma (diamonds) are easily distinguished from tumors of malignant fibrous histiocytoma (MFH) or fibrosarcoma origin (squares) using the expression levels of hsa-miR-206 (SEQ ID NO: 33, y-axis), hsa-miR-22 (SEQ ID NO: 39, x-axis) and hsa-miR-487b (SEQ ID NO: 59, z-axis).

DETAILED DESCRIPTION OF THE INVENTION

[0135] Identification of the tissue-of-origin of a tumor is vital to its management. The present invention is based in part on the discovery that specific nucleic acid sequences can be used for the identification of the tissue-of-origin of a tumor. The present invention provides a sensitive, specific and accurate method which can be used to distinguish between different tumor origins. A new microRNA-based classifier was developed for determining tissue origin of tumors based on 65 microRNAs markers. The classifier uses a specific algorithm and allows a clear interpretation of the specific biomarkers.

[0136] According to the present invention each node in the classification tree may be used as an independent differential diagnosis tool, for example in the identification of different types of lung cancers. The possibility to distinguish between different tumor origins facilitates providing the patient with the best and most suitable treatment.

[0137] The present invention provides diagnostic assays and methods, both quantitative and qualitative for detecting, diagnosing, monitoring, staging and prognosticating cancers by comparing the levels of the specific microRNA molecules of the invention. Such levels are preferably measured in at least one of biopsies, tumor samples, fine-needle aspiration (FNA), cells, tissues and/or bodily fluids. The methods provided in the present invention are particularly useful for discriminating between different cancers.

[0138] All the methods of the present invention may optionally further include measuring levels of additional cancer markers. The cancer markers measured in addition to said microRNA molecules depend on the cancer being tested and are known to those of skill in the art.

[0139] Assay techniques can be used to determine levels of gene expression, such as genes encoding the nucleic acids of the present invention in a sample derived from a patient. Such assay methods, which are well known to those of skill in the art, include, but are not limited to, nucleic acid microarrays and biochip analysis, reverse transcriptase PCR (RT-PCR) assays, immunohistochemistry assays, in situ hybridization assays, competitive-binding assays, northern blot analyses and ELISA assays.

[0140] According to one embodiment, the assay is based on expression level of 65 microRNAs in RNA extracted from FFPE metastatic tumor tissue.

[0141] The expression levels are used to infer the sample origin using analysis techniques such as, but not limited to, decision-tree classifier, K nearest neighbors classifier, logistic regression classifier, linear regression classifier, nearest neighbor classifier, neural network classifier and nearest centroid classifier.

[0142] In use of the decision tree classifier the expression levels are used to make binary decisions (at each relevant node) following the pre-defined structure of the binary decision-tree (defined using a training set).

[0143] At each node, the expressions of one or several microRNAs are combined together using a function of the form P=exp (.beta.0+.beta.1*miR1+.beta.2*miR2+.beta.3*miR3 . . . )/(1-exp (.beta.0+.beta.1*miR1+.beta.2*miR2+.beta.3*miR3 . . . )), where the values of .beta.0, .beta.1, .beta.2 . . . and the identities of the microRNAs have been pre-determined (using a training set). The resulting P is compared to a probability threshold level (P.sub.TH, which was also determined using the training set), and the classification continues to the left or right branch according to whether P is larger or smaller than the P.sub.TH for that node. This continues until an end-point ("leaf") of the tree is reached. According to some embodiments, P.sub.TH=0.5 for all nodes, and the value of .beta.0 is adjusted accordingly. According to further embodiments, .beta.0, .beta.1, .beta.2, . . . are adjusted so that the slope of the log of the odds ratio function is limited.

[0144] Training the tree algorithm means determining the tree structure--which nodes there are and what is on each side, and, for each node: which miRs are used, the values of .beta.0, .beta.1, .beta.2 . . . and the P.sub.TH. These are determined by a combination of machine learning, optimization algorithm, and trial and error by experts in machine learning and diagnostic algorithms.

[0145] An arbitrary threshold of the expression level of one or more nucleic acid sequences can be set for assigning a sample to one of two groups. Alternatively, in a preferred embodiment, expression levels of one or more nucleic acid sequences of the invention are combined by a method such as logistic regression to define a metric which is then compared to previously measured samples or to a threshold. The threshold is treated as a parameter that can be used to quantify the confidence with which samples are assigned to each class. The threshold can be scaled to favor sensitivity or specificity, depending on the clinical scenario. The correlation value to the reference data generates a continuous score that can be scaled and provides diagnostic information on the likelihood that a sample belongs to a certain class of cancer origin or type. In multivariate analysis the microRNA signature provides a high level of prognostic information.

[0146] In another preferred embodiment, expression level of nucleic acids is used to classify a test sample by comparison to a training set of samples. In this embodiment, the test sample is compared in turn to each one of the training set samples. The comparison is performed by comparing the expression levels of one or multiple nucleic acids between the test sample and the specific training sample. Each such pairwise comparison generates a combined metric for the multiple nucleic acids, which can be calculated by various numeric methods such as correlation, cosine, Euclidian distance, mean square distance, or other methods known to those skilled in the art. The training samples are then ranked according to this metric, and the samples with the highest values of the metric (or lowest values, according to the type of metric) are identified, indicating those samples that are most similar to the test sample. By choosing a parameter K, this generates a list that includes the K training samples that are most similar to the test sample. Various methods can then be applied to identify from this list the predicted class of the test sample. In a favored embodiment, the test sample is predicted to belong to the class that has the highest number of representative in the list of K most-similar training samples (this method is known as the K Nearest Neighbors method). Other embodiments may provide a list of predictions including all or part of the classes represented in the list, those classes that are represented more than a given minimum number of times, or other voting schemes whereby classes are grouped together.

1. DEFINITIONS

[0147] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. It must be noted that, as used in the specification and the appended claims, the singular forms "a," "an" and "the" include plural referents unless the context clearly dictates otherwise.

[0148] For the recitation of numeric ranges herein, each intervening number there between with the same degree of precision is explicitly contemplated. For example, for the range of 6-9, the numbers 7 and 8 are contemplated in addition to 6 and 9, and for the range 6.0-7.0, the number 6.0, 6.1, 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9 and 7.0 are explicitly contemplated.

[0149] About

[0150] As used herein, the term "about" refers to +/-10%.

[0151] Attached

[0152] "Attached" or "immobilized", as used herein, to refer to a probe and a solid support means that the binding between the probe and the solid support is sufficient to be stable under conditions of binding, washing, analysis, and removal. The binding may be covalent or non-covalent. Covalent bonds may be formed directly between the probe and the solid support or may be formed by a cross linker or by inclusion of a specific reactive group on either the solid support or the probe or both molecules. Non-covalent binding may be one or more of electrostatic, hydrophilic, and hydrophobic interactions. Included in non-covalent binding is the covalent attachment of a molecule, such as streptavidin, to the support and the non-covalent binding of a biotinylated probe to the streptavidin Immobilization may also involve a combination of covalent and non-covalent interactions.

[0153] Baseline

[0154] "Baseline", as used herein, means the initial cycles of PCR, in which there is little change in fluorescence signal.

[0155] Biological Sample

[0156] "Biological sample", as used herein, means a sample of biological tissue or fluid that comprises nucleic acids. Such samples include, but are not limited to, tissue or fluid isolated from subjects. Biological samples may also include sections of tissues such as biopsy and autopsy samples, FFPE samples, frozen sections taken for histological purposes, blood, blood fraction, plasma, serum, sputum, stool, tears, mucus, hair, skin, urine, effusions, ascitic fluid, amniotic fluid, saliva, cerebrospinal fluid, cervical secretions, vaginal secretions, endometrial secretions, gastrointestinal secretions, bronchial secretions, cell line, tissue sample, or secretions from the breast. A biological sample may be provided by fine-needle aspiration (FNA), pleural effusion or bronchial brushing. A biological sample may be provided by removing a sample of cells from a subject but can also be accomplished by using previously isolated cells (e. g., isolated by another person, at another time, and/or for another purpose), or by performing the methods described herein in vivo. Archival tissues, such as those having treatment or outcome history, may also be used. Biological samples also include explants and primary and/or transformed cell cultures derived from animal or human tissues.

[0157] Cancer

[0158] The term "cancer" is meant to include all types of cancerous growths or oncogenic processes, metastatic tissues or malignantly transformed cells, tissues, or organs, irrespective of histopathologic type or stage of invasiveness. Examples of cancers include, but are not limited, to solid tumors and leukemias, including: apudoma, choristoma, branchioma, malignant carcinoid syndrome, carcinoid heart disease, carcinoma (e.g., Walker, basal cell, basosquamous, Brown-Pearce, ductal, Ehrlich tumor, non-small cell lung (e.g., lung squamous cell carcinoma, lung adenocarcinoma and lung undifferentiated large cell carcinoma), oat cell, papillary, bronchiolar, bronchogenic, squamous cell, and transitional cell), histiocytic disorders, leukemia (e.g., B cell, mixed cell, null cell, T cell, T-cell chronic, HTLV-II-associated, lymphocytic acute, lymphocytic chronic, mast cell, and myeloid), histiocytosis malignant, Hodgkin disease, immunoproliferative small, non-Hodgkin lymphoma, plasmacytoma, reticuloendotheliosis, melanoma, chondroblastoma, chondroma, chondrosarcoma, fibroma, fibrosarcoma, giant cell tumors, histiocytoma, lipoma, liposarcoma, mesothelioma, myxoma, myxosarcoma, osteoma, osteosarcoma, Ewing sarcoma, synovioma, adenofibroma, adenolymphoma, carcinosarcoma, chordoma, craniopharyngioma, dysgerminoma, hamartoma, mesenchymoma, mesonephroma, myosarcoma, ameloblastoma, cementoma, odontoma, teratoma, thymoma, trophoblastic tumor, adeno-carcinoma, adenoma, cholangioma, cholesteatoma, cylindroma, cystadenocarcinoma, cystadenoma, granulosa cell tumor, gynandroblastoma, hepatoma, hidradenoma, islet cell tumor, Leydig cell tumor, papilloma, Sertoli cell tumor, theca cell tumor, leiomyoma, leiomyosarcoma, myoblastoma, myosarcoma, rhabdomyoma, rhabdomyosarcoma, ependymoma, ganglioneuroma, glioma, medulloblastoma, meningioma, neurilemmoma, neuroblastoma, neuroepithelioma, neurofibroma, neuroma, paraganglioma, paraganglioma nonchromaffin, angiokeratoma, angiolymphoid hyperplasia with eosinophilia, angioma sclerosing, angiomatosis, glomangioma, hemangioendothelioma, hemangioma, hemangiopericytoma, hemangiosarcoma, lymphangioma, lymphangiomyoma, lymphangiosarcoma, pinealoma, carcinosarcoma, chondrosarcoma, cystosarcoma, phyllodes, fibrosarcoma, hemangiosarcoma, leimyosarcoma, leukosarcoma, liposarcoma, lymphangiosarcoma, myosarcoma, myxosarcoma, ovarian carcinoma, rhabdomyosarcoma, sarcoma (e.g., Ewing, experimental, Kaposi, and mast cell), neurofibromatosis, and cervical dysplasia, and other conditions in which cells have become immortalized or transformed.

[0159] Classification

[0160] The term classification refers to a procedure and/or algorithm in which individual items are placed into groups or classes based on quantitative information on one or more characteristics inherent in the items (referred to as traits, variables, characters, features, etc.) and based on a statistical model and/or a training set of previously labeled items. A "classification tree" places categorical variables into classes.

[0161] Complement

[0162] "Complement" or "complementary" is used herein to refer to a nucleic acid may mean Watson-Crick (e.g., A-T/U and C-G) or Hoogsteen base pairing between nucleotides or nucleotide analogs of nucleic acid molecules. A full complement or fully complementary means 100% complementary base pairing between nucleotides or nucleotide analogs of nucleic acid molecules. In some embodiments, the complementary sequence has a reverse orientation (5'-3').

[0163] Ct

[0164] Ct signals represent the first cycle of PCR where amplification crosses a threshold (cycle threshold) of fluorescence. Accordingly, low values of Ct represent high abundance or expression levels of the microRNA.

[0165] In some embodiments the PCR Ct signal is normalized such that the normalized Ct remains inversed from the expression level. In other embodiments the PCR Ct signal may be normalized and then inverted such that low normalized-inverted Ct represents low abundance or low expression levels of the microRNA.

[0166] Data Processing Routine

[0167] As used herein, a "data processing routine" refers to a process that can be embodied in software that determines the biological significance of acquired data (i.e., the ultimate results of an assay or analysis). For example, the data processing routine can determine a tissue of origin based upon the data collected. In the systems and methods herein, the data processing routine can also control the data collection routine based upon the results determined. The data processing routine and the data collection routines can be integrated and provide feedback to operate the data acquisition, and hence provide assay-based judging methods.

[0168] Data Set

[0169] As use herein, the term "data set" refers to numerical values obtained from the analysis. These numerical values associated with analysis may be values such as peak height and area under the curve.

[0170] Data Structure

[0171] As used herein, the term "data structure" refers to a combination of two or more data sets, an application of one or more mathematical manipulation to one or more data sets to obtain one or more new data sets, or a manipulation of two or more data sets into a form that provides a visual illustration of the data in a new way. An example of a data structure prepared from manipulation of two or more data sets would be a hierarchical cluster.

[0172] Detection

[0173] "Detection" means detecting the presence of a component in a sample. Detection also means detecting the absence of a component. Detection also means determining the level of a component, either quantitatively or qualitatively.

[0174] Differential Expression

[0175] "Differential expression" means qualitative or quantitative differences in the temporal and/or spatial gene expression patterns within and among cells and tissue. Thus, a differentially expressed gene may qualitatively have its expression altered, including an activation or inactivation, in, e.g., normal versus diseased tissue. Genes may be turned on or turned off in a particular state, relative to another state, thus permitting comparison of two or more states. A qualitatively regulated gene may exhibit an expression pattern within a state or cell type which may be detectable by standard techniques. Some genes may be expressed in one state or cell type, but not in both. Alternatively, the difference in expression may be quantitative, e.g., in that expression is modulated: up-regulated, resulting in an increased amount of transcript, or down-regulated, resulting in a decreased amount of transcript. The degree to which expression differs needs only to be large enough to quantify via standard characterization techniques such as expression arrays, quantitative reverse transcriptase PCR, northern blot analysis, real-time PCR, in situ hybridization and RNase protection.

[0176] Epithelial Tumors

[0177] "Epithelial tumors" is meant to include all types of tumors from epithelial origin. Examples of epithelial tumors include, but are not limited to cholangioca or adenoca of extrahepatic biliary tract, urothelial carcinoma, adenocarcinoma of the breast, lung large cell or adenocarcinoma, lung small cell carcinoma, carcinoid, lung, ovarian carcinoma, pancreatic adenocarcinoma, prostatic adenocarcinoma, gastric or esophageal adenocarcinoma, thymoma/thymic carcinoma, follicular thyroid carcinoma, papillary thyroid carcinoma, medullary thyroid carcinoma, anus or skin squamous cell carcinoma, lung, head&neck, or esophagus squamous cell carcinoma, uterine cervix squamous cell carcinoma, gastrointestinal tract carcinoid, pancreatic islet cell tumor and colorectal adenocarcinoma.

[0178] Non Epithelial Tumors

[0179] "Non epithelial tumors" is meant to include all types of tumors from non epithelial origin. Examples of non epithelial tumors include, but are not limited to adrenocortical carcinoma, chromophobe renal cell carcinoma, clear cell renal cell carcinoma, papillary renal cell carcinoma, pleural mesothelioma, astrocytic tumor, oligodendroglioma, pheochromocytoma, B-cell lymphoma, T-cell lymphoma, melanoma, gastrointestinal stromal tumor (GIST), Ewing Sarcoma, chondrosarcoma, malignant fibrous histiocytoma (MFH) or fibrosarcoma, osteosarcoma, rhabdomyosarcoma, synovial sarcoma and liposarcoma.

[0180] Expression Profile

[0181] The term "expression profile" is used broadly to include a genomic expression profile, e.g., an expression profile of microRNAs. Profiles may be generated by any convenient means for determining a level of a nucleic acid sequence, e.g., quantitative hybridization of microRNA, labeled microRNA, amplified microRNA, cDNA, etc., quantitative PCR, ELISA for quantitation, and the like, and allow the analysis of differential gene expression between two samples. A subject or patient tumor sample, e.g., cells or collections thereof, e.g., tissues, is assayed. Samples are collected by any convenient method, as known in the art. Nucleic acid sequences of interest are nucleic acid sequences that are found to be predictive, including the nucleic acid sequences provided above, where the expression profile may include expression data for 5, 10, 20, 25, 50, 100 or more of the nucleic acid sequences, including all of the listed nucleic acid sequences. According to some embodiments, the term "expression profile" means measuring the relative abundance of the nucleic acid sequences in the measured samples.

[0182] Expression Ratio

[0183] "Expression ratio", as used herein, refers to relative expression levels of two or more nucleic acids as determined by detecting the relative expression levels of the corresponding nucleic acids in a biological sample.

[0184] FDR (False Discovery Rate)

[0185] When performing multiple statistical tests, for example in comparing between the signal of two groups in multiple data features, there is an increasingly high probability of obtaining false positive results, by random differences between the groups that can reach levels that would otherwise be considered statistically significant. In order to limit the proportion of such false discoveries, statistical significance is defined only for data features in which the differences reached a p-value (by two-sided t-test) below a threshold, which is dependent on the number of tests performed and the distribution of p-values obtained in these tests.

[0186] Fragment

[0187] "Fragment" is used herein to indicate a non-full-length part of a nucleic acid. Thus, a fragment is itself also a nucleic acid.

[0188] Gastrointestinal Tumors

[0189] "gastrointestinal tumors" is meant to include all types of tumors from gastrointestinal origin. Examples of gastrointestinal tumors include, but are not limited to cholangioca. or adenoca of extrahepatic biliary tract, pancreatic adenocarcinoma, gastric or esophageal adenocarcinoma, and colorectal adenocarcinoma.

[0190] Gene

[0191] "Gene", as used herein, may be a natural (e.g., genomic) or synthetic gene comprising transcriptional and/or translational regulatory sequences and/or a coding region and/or non-translated sequences (e.g., introns, 5'- and 3'-untranslated sequences). The coding region of a gene may be a nucleotide sequence coding for an amino acid sequence or a functional RNA, such as tRNA, rRNA, catalytic RNA, siRNA, miRNA or antisense RNA. A gene may also be an mRNA or cDNA corresponding to the coding regions (e.g., exons and miRNA) optionally comprising 5'- or 3'-untranslated sequences linked thereto. A gene may also be an amplified nucleic acid molecule produced in vitro, comprising all or a part of the coding region and/or 5'- or 3'-untranslated sequences linked thereto.

[0192] Germ Cell Tumors

[0193] "Germ cell tumors" as used herein, include, but are not limited, to non-seminomatous testicular germ cell tumors, seminomatous testicular germ cell tumors and ovarian primitive germ cell tumors.

[0194] Groove Binder/Minor Groove Binder (MGB)

[0195] "Groove binder" and/or "minor groove binder" may be used interchangeably and refer to small molecules that fit into the minor groove of double-stranded DNA, typically in a sequence-specific manner. Minor groove binders may be long, flat molecules that can adopt a crescent-like shape and thus fit snugly into the minor groove of a double helix, often displacing water. Minor groove binding molecules may typically comprise several aromatic rings connected by bonds with torsional freedom such as furan, benzene, or pyrrole rings. Minor groove binders may be antibiotics such as netropsin, distamycin, berenil, pentamidine and other aromatic diamidines, Hoechst 33258, SN 6999, aureolic anti-tumor drugs such as chromomycin and mithramycin, CC-1065, dihydrocyclopyrroloindole tripeptide (DPI.sub.3), 1,2-dihydro-(3H)-pyrrolo[3,2-e]indole-7-carboxylate (CDPI.sub.3), and related compounds and analogues, including those described in Nucleic Acids in Chemistry and Biology, 2nd ed., Blackburn and Gait, eds., Oxford University Press, 1996, and PCT Published Application No. WO 03/078450, the contents of which are incorporated herein by reference. A minor groove binder may be a component of a primer, a probe, a hybridization tag complement, or combinations thereof. Minor groove binders may increase the T.sub.m of the primer or a probe to which they are attached, allowing such primers or probes to effectively hybridize at higher temperatures.

[0196] High Expression miR-205 Tumors

[0197] "High expression miR-205 tumors" as used herein include, but are not limited, to urothelial carcinoma (TCC), thymoma/thymic carcinoma, anus or skin squamous cell carcinoma, lung, head&neck, or esophagus squamous cell carcinoma and uterine cervix squamous cell carcinoma.

[0198] Low Expression 205 Tumors

[0199] "Low expression miR-205 tumors" as used herein include, but are not limited, to lung, large cell or adenocarcinoma, follicular thyroid carcinoma and papillary thyroid carcinoma.

[0200] Host Cell

[0201] "Host cell", as used herein, may be a naturally occurring cell or a transformed cell that may contain a vector and may support replication of the vector. Host cells may be cultured cells, explants, cells in vivo, and the like. Host cells may be prokaryotic cells, such as E. coli, or eukaryotic cells, such as yeast, insect, amphibian, or mammalian cells, such as CHO and HeLa cells.

[0202] Identity

[0203] "Identical" or "identity", as used herein, in the context of two or more nucleic acids or polypeptide sequences mean that the sequences have a specified percentage of residues that are the same over a specified region. The percentage may be calculated by optimally aligning the two sequences, comparing the two sequences over the specified region, determining the number of positions at which the identical residue occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the specified region, and multiplying the result by 100 to yield the percentage of sequence identity. In cases where the two sequences are of different lengths or the alignment produces one or more staggered ends and the specified region of comparison includes only a single sequence, the residues of single sequence are included in the denominator but not the numerator of the calculation. When comparing DNA and RNA sequences, thymine (T) and uracil (U) may be considered equivalent. Identity may be performed manually or by using a computer sequence algorithm such as BLAST or BLAST 2.0.

[0204] In Situ Detection

[0205] "In situ detection", as used herein, means the detection of expression or expression levels in the original site, hereby meaning in a tissue sample such as biopsy.

[0206] K-Nearest Neighbor

[0207] The phrase "K-nearest neighbor" refers to a classification method that classifies a point by calculating the distances between it and points in the training data set. It then assigns the point to the class that is most common among its K-nearest neighbors (where K is an integer).

[0208] Leaf

[0209] A leaf, as used herein, is the terminal group in a classification or decision tree.

[0210] Label

[0211] "Label", as used herein, means a composition detectable by spectroscopic, photochemical, biochemical, immunochemical, chemical, or other physical means. For example, useful labels include .sup.32P, fluorescent dyes, electron-dense reagents, enzymes (e.g., as commonly used in an ELISA), biotin, digoxigenin, or haptens and other entities which can be made detectable. A label may be incorporated into nucleic acids and proteins at any position.

[0212] Logistic Regression

[0213] Logistic regression is part of a category of statistical models called generalized linear models. Logistic regression can allow one to predict a discrete outcome, such as group membership, from a set of variables that may be continuous, discrete, dichotomous, or a mix of any of these. The dependent or response variable can be dichotomous, for example, one of two possible types of cancer. Logistic regression models the natural log of the odds ratio, i.e., the ratio of the probability of belonging to the first group (P) over the probability of belonging to the second group (1-P), as a linear combination of the different expression levels (in log-space). The logistic regression output can be used as a classifier by prescribing that a case or sample will be classified into the first type if P is greater than 0.5 or 50%. Alternatively, the calculated probability P can be used as a variable in other contexts, such as a 1D or 2D threshold classifier.

[0214] Metastasis

[0215] "Metastasis" means the process by which cancer spreads from the place at which it first arose as a primary tumor to other locations in the body. The metastatic progression of a primary tumor reflects multiple stages, including dissociation from neighboring primary tumor cells, survival in the circulation, and growth in a secondary location.

[0216] Neuroendocrine Tumors

[0217] "Neuroendocrine tumors" is meant to include all types of tumors from neuroendocrine origin. Examples of neuroendocrine tumors include, but are not limited to lung small cell carcinoma, lung carcinoid, gastrointestinal tract carcinoid, pancreatic islet cell tumor and medullary thyroid carcinoma.

[0218] Node

[0219] A "node" is a decision point in a classification (i.e., decision) tree. Also, a point in a neural net that combines input from other nodes and produces an output through application of an activation function.

[0220] Nucleic Acid

[0221] "Nucleic acid" or "oligonucleotide" or "polynucleotide", as used herein, mean at least two nucleotides covalently linked together. The depiction of a single strand also defines the sequence of the complementary strand. Thus, a nucleic acid also encompasses the complementary strand of a depicted single strand. Many variants of a nucleic acid may be used for the same purpose as a given nucleic acid. Thus, a nucleic acid also encompasses substantially identical nucleic acids and complements thereof. A single strand provides a probe that may hybridize to a target sequence under stringent hybridization conditions. Thus, a nucleic acid also encompasses a probe that hybridizes under stringent hybridization conditions.

[0222] Nucleic acids may be single-stranded or double-stranded, or may contain portions of both double-stranded and single-stranded sequences. The nucleic acid may be DNA, both genomic and cDNA, RNA, or a hybrid, where the nucleic acid may contain combinations of deoxyribo- and ribo-nucleotides, and combinations of bases including uracil, adenine, thymine, cytosine, guanine, inosine, xanthine hypoxanthine, isocytosine and isoguanine. Nucleic acids may be obtained by chemical synthesis methods or by recombinant methods.

[0223] A nucleic acid will generally contain phosphodiester bonds, although nucleic acid analogs may be included that may have at least one different linkage, e.g., phosphoramidate, phosphorothioate, phosphorodithioate, or O-methylphosphoroamidite linkages and peptide nucleic acid backbones and linkages. Other analog nucleic acids include those with positive backbones, non-ionic backbones and non-ribose backbones, including those described in U.S. Pat. Nos. 5,235,033 and 5,034,506, which are incorporated herein by reference. Nucleic acids containing one or more non-naturally occurring or modified nucleotides are also included within one definition of nucleic acids. The modified nucleotide analog may be located for example at the 5'-end and/or the 3'-end of the nucleic acid molecule. Representative examples of nucleotide analogs may be selected from sugar- or backbone-modified ribonucleotides. It should be noted, however, that also nucleobase-modified ribonucleotides, i.e., ribonucleotides, containing a non-naturally occurring nucleobase instead of a naturally occurring nucleobase such as uridine or cytidine modified at the 5-position, e.g., 5-(2-amino) propyl uridine, 5-bromo uridine; adenosine and guanosine modified at the 8-position, e.g., 8-bromo guanosine; deaza nucleotides, e.g., 7-deaza-adenosine; 0- and N-alkylated nucleotides, e.g., N6-methyl adenosine are suitable. The 2'-OH-group may be replaced by a group selected from H, OR, R, halo, SH, SR, NH.sub.2, NHR, NR.sub.2 or CN, wherein R is C1-C6 alkyl, alkenyl or alkynyl and halo is F, Cl, Br or I. Modified nucleotides also include nucleotides conjugated with cholesterol through, e.g., a hydroxyprolinol linkage as described in Krutzfeldt et al., Nature 2005; 438:685-689, Soutschek et al., Nature 2004; 432:173-178, and U.S. Patent Publication No. 20050107325, which are incorporated herein by reference. Additional modified nucleotides and nucleic acids are described in U.S. Patent Publication No. 20050182005, which is incorporated herein by reference. Modifications of the ribose-phosphate backbone may be done for a variety of reasons, e.g., to increase the stability and half-life of such molecules in physiological environments, to enhance diffusion across cell membranes, or as probes on a biochip. The backbone modification may also enhance resistance to degradation, such as in the harsh endocytic environment of cells. The backbone modification may also reduce nucleic acid clearance by hepatocytes, such as in the liver and kidney. Mixtures of naturally occurring nucleic acids and analogs may be made; alternatively, mixtures of different nucleic acid analogs, and mixtures of naturally occurring nucleic acids and analogs may be made.

[0224] Probe

[0225] "Probe", as used herein, means an oligonucleotide capable of binding to a target nucleic acid of complementary sequence through one or more types of chemical bonds, usually through complementary base pairing, usually through hydrogen bond formation. Probes may bind target sequences lacking complete complementarity with the probe sequence depending upon the stringency of the hybridization conditions. There may be any number of base pair mismatches which will interfere with hybridization between the target sequence and the single-stranded nucleic acids described herein. However, if the number of mutations is so great that no hybridization can occur under even the least stringent of hybridization conditions, the sequence is not a complementary target sequence. A probe may be single-stranded or partially single- and partially double-stranded. The strandedness of the probe is dictated by the structure, composition, and properties of the target sequence. Probes may be directly labeled or indirectly labeled such as with biotin to which a streptavidin complex may later bind.

[0226] Reference Value

[0227] As used herein, the term "reference value" or "reference expression profile" refers to a criterion expression value to which measured values are compared in order to identify a specific cancer. The reference value may be based on the abundance of the nucleic acids, or may be based on a combined metric score thereof.

[0228] In preferred embodiments the reference value is determined from statistical analysis of studies that compare microRNA expression with known clinical outcomes.

[0229] Sarcoma

[0230] Sarcoma is meant to include all types of tumors from sarcoma origin. Examples of sarcoma tumors include, but are not limited to gastrointestinal stromal tumor (GIST), Ewing sarcoma, chondrosarcoma, malignant fibrous histiocytoma (MFH) or fibrosarcoma, osteosarcoma, rhabdomyosarcoma, synovial sarcoma and liposarcoma.

[0231] Sensitivity

[0232] "Sensitivity", as used herein, may mean a statistical measure of how well a binary classification test correctly identifies a condition, for example, how frequently it correctly classifies a cancer into the correct class out of two possible classes. The sensitivity for class A is the proportion of cases that are determined to belong to class "A" by the test out of the cases that are in class "A", as determined by some absolute or gold standard.

[0233] Specificity

[0234] "Specificity", as used herein, may mean a statistical measure of how well a binary classification test correctly identifies a condition, for example, how frequently it correctly classifies a cancer into the correct class out of two possible classes. The specificity for class A is the proportion of cases that are determined to belong to class "not A" by the test out of the cases that are in class "not A", as determined by some absolute or gold standard.

[0235] Stringent Hybridization Conditions

[0236] "Stringent hybridization conditions", as used herein, mean conditions under which a first nucleic acid sequence (e.g., probe) will hybridize to a second nucleic acid sequence (e.g., target), such as in a complex mixture of nucleic acids. Stringent conditions are sequence-dependent and will be different in different circumstances. Stringent conditions may be selected to be about 5-10.degree. C. lower than the thermal melting point (T.sub.m) for the specific sequence at a defined ionic strength pH. The T.sub.m may be the temperature (under defined ionic strength, pH, and nucleic concentration) at which 50% of the probes complementary to the target hybridize to the target sequence at equilibrium (as the target sequences are present in excess, at T.sub.m, 50% of the probes are occupied at equilibrium). Stringent conditions may be those in which the salt concentration is less than about 1.0 M sodium ion, such as about 0.01-1.0 M sodium ion concentration (or other salts) at pH 7.0 to 8.3 and the temperature is at least about 30.degree. C. for short probes (e.g., about 10-50 nucleotides) and at least about 60.degree. C. for long probes (e.g., greater than about 50 nucleotides). Stringent conditions may also be achieved with the addition of destabilizing agents such as formamide. For selective or specific hybridization, a positive signal may be at least 2 to 10 times background hybridization. Exemplary stringent hybridization conditions include the following: 50% formamide, 5.times.SSC, and 1% SDS, incubating at 42.degree. C., or, 5.times.SSC, 1% SDS, incubating at 65.degree. C., with wash in 0.2.times.SSC, and 0.1% SDS at 65.degree. C.

[0237] Substantially Complementary

[0238] "Substantially complementary", as used herein, means that a first sequence is at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 97%, 98% or 99% identical to the complement of a second sequence over a region of 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100 or more nucleotides, or that the two sequences hybridize under stringent hybridization conditions.

[0239] Substantially Identical

[0240] "Substantially identical", as used herein, means that a first and a second sequence are at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 97%, 98% or 99% identical over a region of 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100 or more nucleotides or amino acids, or with respect to nucleic acids, if the first sequence is substantially complementary to the complement of the second sequence.

[0241] Subject

[0242] As used herein, the term "subject" refers to a mammal, including both human and other mammals. The methods of the present invention are preferably applied to human subjects.

[0243] Target Nucleic Acid

[0244] "Target nucleic acid", as used herein, means a nucleic acid or variant thereof that may be bound by another nucleic acid. A target nucleic acid may be a DNA sequence. The target nucleic acid may be RNA. The target nucleic acid may comprise a mRNA, tRNA, shRNA, siRNA or Piwi-interacting RNA, or a pri-miRNA, pre-miRNA, miRNA, or anti-miRNA.

[0245] The target nucleic acid may comprise a target miRNA binding site or a variant thereof. One or more probes may bind the target nucleic acid. The target binding site may comprise 5-100 or 10-60 nucleotides. The target binding site may comprise a total of 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30-40, 40-50, 50-60, 61, 62 or 63 nucleotides. The target site sequence may comprise at least 5 nucleotides of the sequence of a target miRNA binding site disclosed in U.S. patent application Ser. Nos. 11/384,049, 11/418,870 or 11/429,720, the contents of which are incorporated herein.

[0246] 1D/2D Threshold Classifier

[0247] "1D/2D threshold classifier", as used herein, may mean an algorithm for classifying a case or sample such as a cancer sample into one of two possible types such as two types of cancer. For a 1D threshold classifier, the decision is based on one variable and one predetermined threshold value; the sample is assigned to one class if the variable exceeds the threshold and to the other class if the variable is less than the threshold. A 2D threshold classifier is an algorithm for classifying into one of two types based on the values of two variables. A threshold may be calculated as a function (usually a continuous or even a monotonic function) of the first variable; the decision is then reached by comparing the second variable to the calculated threshold, similar to the 1D threshold classifier.

[0248] Tissue Sample

[0249] As used herein, a tissue sample is tissue obtained from a tissue biopsy using methods well known to those of ordinary skill in the related medical arts. The phrase "suspected of being cancerous", as used herein, means a cancer tissue sample believed by one of ordinary skill in the medical arts to contain cancerous cells. Methods for obtaining the sample from the biopsy include gross apportioning of a mass, microdissection, laser-based microdissection, or other art-known cell-separation methods.

[0250] Tumor

[0251] "Tumor", as used herein, refers to all neoplastic cell growth and proliferation, whether malignant or benign, and all pre-cancerous and cancerous cells and tissues.

[0252] Variant

[0253] "Variant", as used herein, referring to a nucleic acid means (i) a portion of a referenced nucleotide sequence; (ii) the complement of a referenced nucleotide sequence or portion thereof; (iii) a nucleic acid that is substantially identical to a referenced nucleic acid or the complement thereof; or (iv) a nucleic acid that hybridizes under stringent conditions to the referenced nucleic acid, complement thereof, or a sequence substantially identical thereto.

[0254] Wild Type

[0255] As used herein, the term "wild-type" sequence refers to a coding, a non-coding or an interface sequence which is an allelic form of sequence that performs the natural or normal function for that sequence. Wild-type sequences include multiple allelic forms of a cognate sequence, for example, multiple alleles of a wild type sequence may encode silent or conservative changes to the protein sequence that a coding sequence encodes.

[0256] The present invention employs miRNAs for the identification, classification and diagnosis of specific cancers and the identification of their tissues of origin.

[0257] 1. microRNA Processing

[0258] A gene coding for microRNA (miRNA) may be transcribed leading to production of a miRNA primary transcript known as the pri-miRNA. The pri-miRNA may comprise a hairpin with a stem and loop structure. The stem of the hairpin may comprise mismatched bases. The pri-miRNA may comprise several hairpins in a polycistronic structure.

[0259] The hairpin structure of the pri-miRNA may be recognized by Drosha, which is an RNase III endonuclease. Drosha may recognize terminal loops in the pri-miRNA and cleave approximately two helical turns into the stem to produce a 60-70 nt precursor known as the pre-miRNA. Drosha may cleave the pri-miRNA with a staggered cut typical of RNase III endonucleases yielding a pre-miRNA stem loop with a 5' phosphate and .about.2 nucleotide 3' overhang. Approximately one helical turn of stem (.about.10 nucleotides) extending beyond the Drosha cleavage site may be essential for efficient processing. The pre-miRNA may then be actively transported from the nucleus to the cytoplasm by Ran-GTP and the export receptor Ex-portin-5.

[0260] The pre-miRNA may be recognized by Dicer, which is also an RNase III endonuclease. Dicer may recognize the double-stranded stem of the pre-miRNA. Dicer may also cut off the terminal loop two helical turns away from the base of the stem loop, leaving an additional 5' phosphate and a .about.2 nucleotide 3' overhang. The resulting siRNA-like duplex, which may comprise mismatches, comprises the mature miRNA and a similar-sized fragment known as the miRNA*. The miRNA and miRNA* may be derived from opposing arms of the pri-miRNA and pre-miRNA. MiRNA* sequences may be found in libraries of cloned miRNAs, but typically at lower frequency than the miRNAs.

[0261] Although initially present as a double-stranded species with miRNA*, the miRNA may eventually become incorporated as a single-stranded RNA into a ribonucleoprotein complex known as the RNA-induced silencing complex (RISC). Various proteins can form the RISC, which can lead to variability in specificity for miRNA/miRNA* duplexes, binding site of the target gene, activity of miRNA (repress or activate), and which strand of the miRNA/miRNA* duplex is loaded in to the RISC.

[0262] When the miRNA strand of the miRNA:miRNA* duplex is loaded into the RISC, the miRNA* may be removed and degraded. The strand of the miRNA:miRNA* duplex that is loaded into the RISC may be the strand whose 5' end is less tightly paired. In cases where both ends of the miRNA:miRNA* have roughly equivalent 5' pairing, both miRNA and miRNA* may have gene silencing activity.

[0263] The RISC may identify target nucleic acids based on high levels of complementarity between the miRNA and the mRNA, especially by nucleotides 2-7 of the miRNA. Only one case has been reported in animals where the interaction between the miRNA and its target was along the entire length of the miRNA. This was shown for miR-196 and Hox B8 and it was further shown that miR-196 mediates the cleavage of the Hox B8 mRNA (Yekta et al. Science 2004; 304:594-596). Otherwise, such interactions are known only in plants (Bartel & Bartel 2003; 132:709-717).

[0264] A number of studies have looked at the base-pairing requirement between miRNA and its mRNA target for achieving efficient inhibition of translation (reviewed by Bartel 2004; 116:281-297). In mammalian cells, the first 8 nucleotides of the miRNA may be important (Doench & Sharp Genes Dev 2004; 18:504-511). However, other parts of the microRNA may also participate in mRNA binding. Moreover, sufficient base pairing at the 3' can compensate for insufficient pairing at the 5' (Brennecke et al., PloS Biol 2005; 3:e85). Computation studies, analyzing miRNA binding on whole genomes have suggested a specific role for bases 2-7 at the 5' of the miRNA in target binding but the role of the first nucleotide, found usually to be "A" was also recognized (Lewis et al. Cell 2005; 120:15-20) Similarly, nucleotides 1-7 or 2-8 were used to identify and validate targets by Krek et al. (Nat Genet 2005; 37:495-500).

[0265] The target sites in the mRNA may be in the 5' UTR, the 3' UTR or in the coding region. Interestingly, multiple miRNAs may regulate the same mRNA target by recognizing the same or multiple sites. The presence of multiple miRNA binding sites in most genetically identified targets may indicate that the cooperative action of multiple RISCs provides the most efficient translational inhibition.

[0266] miRNAs may direct the RISC to down-regulate gene expression by either of two mechanisms: mRNA cleavage or translational repression. The miRNA may specify cleavage of the mRNA if the mRNA has a certain degree of complementarity to the miRNA. When a miRNA guides cleavage, the cut may be between the nucleotides pairing to residues 10 and 11 of the miRNA. Alternatively, the miRNA may repress translation if the miRNA does not have the requisite degree of complementarity to the miRNA. Translational repression may be more prevalent in animals since animals may have a lower degree of complementarity between the miRNA and binding site.

[0267] It should be noted that there may be variability in the 5' and 3' ends of any pair of miRNA and miRNA*. This variability may be due to variability in the enzymatic processing of Drosha and Dicer with respect to the site of cleavage. Variability at the 5' and 3' ends of miRNA and miRNA* may also be due to mismatches in the stem structures of the pri-miRNA and pre-miRNA. The mismatches of the stem strands may lead to a population of different hairpin structures. Variability in the stem structures may also lead to variability in the products of cleavage by Drosha and Dicer.

[0268] 2. Nucleic Acids

[0269] Nucleic acids are provided herein. The nucleic acids comprise the sequences of SEQ ID NOS: 1-390 or variants thereof. The variant may be a complement of the referenced nucleotide sequence. The variant may also be a nucleotide sequence that is substantially identical to the referenced nucleotide sequence or the complement thereof. The variant may also be a nucleotide sequence which hybridizes under stringent conditions to the referenced nucleotide sequence, complements thereof, or nucleotide sequences substantially identical thereto.

[0270] The nucleic acid may have a length of from about 10 to about 250 nucleotides. The nucleic acid may have a length of at least 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, 125, 150, 175, 200 or 250 nucleotides. The nucleic acid may be synthesized or expressed in a cell (in vitro or in vivo) using a synthetic gene described herein. The nucleic acid may be synthesized as a single-strand molecule and hybridized to a substantially complementary nucleic acid to form a duplex. The nucleic acid may be introduced to a cell, tissue or organ in a single- or double-stranded form or capable of being expressed by a synthetic gene using methods well known to those skilled in the art, including as described in U.S. Pat. No. 6,506,559, which is incorporated herein by reference.

[0271] SEQ ID NOs 1-34 are in accordance with Sanger database version 10; SEQ ID NOs 35-390 are in accordance with Sanger database version 11;

TABLE-US-00001 TABLE 1 SEQ ID NOS of sequences used in the invention miR name miR SEQ ID NO hairpin SEQ ID NO hsa-let-7c 1 70 hsa-let-7e 2, 156 71 hsa-miR-100 3 72 hsa-miR-10a 4 73 hsa-miR-10b 5 74 hsa-miR-122 6 75 hsa-miR-125a-5p 7 76 hsa-miR-125b 8 77, 78 hsa-miR-126 9 79 hsa-miR-130a 10 80 hsa-miR-138 11 81, 82 hsa-miR-140-3p 12 83 hsa-miR-141 13 84 hsa-miR-143 14 85 hsa-miR-145 15 86 hsa-miR-146a 16 87 hsa-miR-146b-5p 17 88 hsa-miR-148a 18 89 hsa-miR-149 19 90 hsa-miR-17 20 91 hsa-miR-181a 21 92, 93 hsa-miR-181a* 22 92, 93 hsa-miR-185 23 94 hsa-miR-191 24 95 hsa-miR-193a-3p 25 96 hsa-miR-193a-5p 26 96 hsa-miR-194 27 97, 98 hsa-miR-200a 28 99 hsa-miR-200b 29 100 hsa-miR-200c 30 101 hsa-miR-202 31 102 hsa-miR-205 32 103 hsa-miR-206 33 104 hsa-miR-21 34 105 hsa-miR-21* 35 105 hsa-miR-210 36 106 hsa-miR-214 37 107 hsa-miR-214* 38 107 hsa-miR-22 39 108 hsa-miR-222 40 109 hsa-miR-223 41 110 hsa-miR-224 42 111 hsa-miR-29a 43 112 hsa-miR-29c 44, 191 113 hsa-miR-29c* 45 113 hsa-miR-30a 46 114 hsa-miR-30d 47 115 hsa-miR-30e 48 116 hsa-miR-31 49 117 hsa-miR-342-3p 50 118 hsa-miR-345 51 119 hsa-miR-34a 52 120 hsa-miR-34c-5p 53 121 hsa-miR-361-5p 54 122 hsa-miR-372 55 123 hsa-miR-375 56 124 hsa-miR-378 57, 202 125 hsa-miR-455-5p 58 126 hsa-miR-487b 59 127 hsa-miR-497 60, 208 128 hsa-miR-509-3p 61 129, 130, 131 hsa-miR-516a-5p 62, 211 132, 133 hsa-miR-574-5p 63 134 hsa-miR-652 64 135 hsa-miR-7 65 136, 137, 138 hsa-miR-9* 66 139, 140, 141 hsa-miR-92a 67 142, 143 hsa-miR-92b 68 144 hsa-miR-934 69 145 hsa-miR-1201 146 149 hsa-miR-221 147 150 hsa-miR-93 148 151 hsa-miR-182 152 hsa-let-7d 153 hsa-miR-181b 154 hsa-miR-127-3p 155 hsa-let-7i 157 hsa-miR-106a 158 hsa-miR-124 159 hsa-miR-1248 160 hsa-miR-128 161 hsa-miR-129-3p 162 hsa-miR-1323 163 hsa-miR-142-5p 164 hsa-miR-143* 165 hsa-miR-146b-3p 166 hsa-miR-149* 167 hsa-miR-150 168 hsa-miR-152 169 hsa-miR-155 170 hsa-miR-15a 171 hsa-miR-15b 172 hsa-miR-181c 173 hsa-miR-181d 174 hsa-miR-183 175 hsa-miR-18a 176 hsa-miR-192 177 hsa-miR-193b 178 hsa-miR-195 179 hsa-miR-1973 180 hsa-miR-199a-3p 181 hsa-miR-199a-5p 182 hsa-miR-199b-5p 183 hsa-miR-203 184 hsa-miR-205* 185 hsa-miR-20a 186 hsa-miR-219-2-3p 187 hsa-miR-25 188 hsa-miR-27b 189 hsa-miR-29b 190 hsa-miR-302a 192 hsa-miR-302a* 193 hsa-miR-302d 194 hsa-miR-30a* 195 hsa-miR-30c 196 hsa-miR-331-3p 197 hsa-miR-342-5p 198 hsa-miR-363 199 hsa-miR-371-3p 200 hsa-miR-371-5p 201 hsa-miR-422a 203 hsa-miR-425 204 hsa-miR-451 205 hsa-miR-455-3p 206 hsa-miR-486-5p 207 hsa-miR-498 209 hsa-miR-512-5p 210 hsa-miR-516b 212 hsa-miR-517a 213 hsa-miR-517c 214 hsa-miR-518a-3p 215 hsa-miR-518e 216 hsa-miR-518f* 217 hsa-miR-519a 218 hsa-miR-519d 219 hsa-miR-520a-5p 220 hsa-miR-520c-3p 221 hsa-miR-520d-5p 222 hsa-miR-524-5p 223 hsa-miR-527 224 hsa-miR-551b 225 hsa-miR-625 226 hsa-miR-767-5p 227 hsa-miR-886-3p 228 hsa-miR-9 229 hsa-miR-886-5p 230 hsa-miR-99a 231 hsa-miR-99a* 232 hsa-miR-373 233 hsa-miR-1977 234 hsa-miR-1978 235 MID-00689 236 MID-15684 237, 369 MID-15867 238 MID-15907 239 MID-15965 240 MID-16318 241 MID-16489 242 MID-16869 243 MID-17144 244 MID-18336 245 MID-18422 246 MID-19340 247 MID-19533 248 MID-20524 249 MID-20703 250 MID-21271 251 MID-22664 252 MID-23256 253 MID-23291 254 MID-23794 255 MID-00405 390 hsa-let-7a 256 hsa-let-7b 257 hsa-let-7f 258 hsa-let-7g 259 hsa-miR-106b 260 hsa-miR-1180 261 hsa-miR-127-5p 262 hsa-miR-129* 263 hsa-miR-129-5p 264 hsa-miR-130b 265 hsa-miR-132 266 hsa-miR-133a 267 hsa-miR-133b 268 hsa-miR-134 269 hsa-miR-139-5p 270 hsa-miR-140-5p 271 hsa-miR-145* 272 hsa-miR-148b 273 hsa-miR-151-3p 274 hsa-miR-154 275 hsa-miR-154* 276 hsa-miR-17* 277 hsa-miR-181a-2* 278 hsa-miR-1826 279 hsa-miR-187 280 hsa-miR-188-5p 281 hsa-miR-196a 282 hsa-miR-1979 283 hsa-miR-19b 284 hsa-miR-20b 285 hsa-miR-216a 286 hsa-miR-216b 287 hsa-miR-217 288 hsa-miR-22* 289 hsa-miR-221* 290 hsa-miR-222* 291 hsa-miR-23a 292 hsa-miR-23b 293 hsa-miR-24 294 hsa-miR-26a 295 hsa-miR-26b 296 hsa-miR-27a 297 hsa-miR-28-3p 298 hsa-miR-296-5p 299 hsa-miR-299-3p 300 hsa-miR-29b-2* 301 hsa-miR-301a 302 hsa-miR-30b 303 hsa-miR-30e* 304 hsa-miR-31* 305 hsa-miR-323-3p 306 hsa-miR-324-5p 307 hsa-miR-328 308 hsa-miR-329 309 hsa-miR-330-3p 310 hsa-miR-335 311 hsa-miR-337-5p 312 hsa-miR-338-3p 313 hsa-miR-361-3p 314 hsa-miR-362-3p 315 hsa-miR-362-5p 316 hsa-miR-369-5p 317 hsa-miR-370 318 hsa-miR-376a 319 hsa-miR-376c 320 hsa-miR-377* 321 hsa-miR-379 322 hsa-miR-381 323 hsa-miR-382 324 hsa-miR-409-3p 325 hsa-miR-409-5p 326 hsa-miR-410 327 hsa-miR-411 328

hsa-miR-425* 329 hsa-miR-431* 330 hsa-miR-432 331 hsa-miR-433 332 hsa-miR-483-3p 333 hsa-miR-483-5p 334 hsa-miR-485-3p 335 hsa-miR-485-5p 336 hsa-miR-487a 337 hsa-miR-494 338 hsa-miR-495 339 hsa-miR-500 340 hsa-miR-500* 341 hsa-miR-501-3p 342 hsa-miR-502-3p 343 hsa-miR-503 344 hsa-miR-506 345 hsa-miR-509-3-5p 346 hsa-miR-513a-5p 347 hsa-miR-532-3p 348 hsa-miR-532-5p 349 hsa-miR-539 350 hsa-miR-542-5p 351 hsa-miR-543 352 hsa-miR-598 353 hsa-miR-612 354 hsa-miR-654-3p 355 hsa-miR-658 356 hsa-miR-660 357 hsa-miR-665 358 hsa-miR-708 359 hsa-miR-873 360 hsa-miR-874 361 hsa-miR-891a 362 hsa-miR-99b 363 MID-00064 364 MID-00078 365 MID-00144 366 MID-00465 367 MID-00672 368 MID-15986 370 MID-16270 371 MID-16469 372 MID-16582 373 MID-16748 374 MID-17356 (3651) 389 MID-17375 375 MID-17576 376 MID-17866 377 MID-18307 378 MID-18395 379 MID-19898 380 MID-19962 381 MID-22331 382 MID-22912 383 MID-23017 384 MID-23168 385 MID-23178 386 MID-23751 387 hsa-miR-423-5p 388

[0272] 3. Nucleic Acid Complexes

[0273] The nucleic acid may further comprise one or more of the following: a peptide, a protein, a RNA-DNA hybrid, an antibody, an antibody fragment, a Fab fragment, and an aptamer.

[0274] 4. Pri-miRNA

[0275] The nucleic acid may comprise a sequence of a pri-miRNA or a variant thereof. The pri-miRNA sequence may comprise from 45-30,000, 50-25,000, 100-20,000, 1,000-1,500 or 80-100 nucleotides. The sequence of the pri-miRNA may comprise a pre-miRNA, miRNA and miRNA*, as set forth herein, and variants thereof. The sequence of the pri-miRNA may comprise any of the sequences of SEQ ID NOS: 1-390 or variants thereof.

[0276] The pri-miRNA may comprise a hairpin structure. The hairpin may comprise a first and a second nucleic acid sequence that are substantially complimentary. The first and second nucleic acid sequence may be from 37-50 nucleotides. The first and second nucleic acid sequence may be separated by a third sequence of from 8-12 nucleotides. The hairpin structure may have a free energy of less than -25 Kcal/mole, as calculated by the Vienna algorithm with default parameters, as described in Hofacker et al. (Monatshefte f. Chemie 1994; 125:167-188), the contents of which are incorporated herein by reference. The hairpin may comprise a terminal loop of 4-20, 8-12 or 10 nucleotides. The pri-miRNA may comprise at least 19% adenosine nucleotides, at least 16% cytosine nucleotides, at least 23% thymine nucleotides and at least 19% guanine nucleotides.

[0277] 5. Pre-miRNA

[0278] The nucleic acid may also comprise a sequence of a pre-miRNA or a variant thereof. The pre-miRNA sequence may comprise from 45-90, 60-80 or 60-70 nucleotides. The sequence of the pre-miRNA may comprise a miRNA and a miRNA* as set forth herein. The sequence of the pre-miRNA may also be that of a pri-miRNA excluding from 0-160 nucleotides from the 5' and 3' ends of the pri-miRNA. The sequence of the pre-miRNA may comprise the sequence of SEQ ID NOS: 1-390 or variants thereof.

[0279] 6. miRNA

[0280] The nucleic acid may also comprise a sequence of a miRNA (including miRNA*) or a variant thereof. The miRNA sequence may comprise from 13-33, 18-24 or 21-23 nucleotides. The miRNA may also comprise a total of at least 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39 or 40 nucleotides. The sequence of the miRNA may be the first 13-33 nucleotides of the pre-miRNA. The sequence of the miRNA may also be the last 13-33 nucleotides of the pre-miRNA. The sequence of the miRNA may comprise the sequence of SEQ ID NOS: 1-69, 146-148, 152-390 or variants thereof.

[0281] 7. Probes

[0282] A probe comprising a nucleic acid described herein is also provided. Probes may be used for screening and diagnostic methods, as outlined below. The probe may be attached or immobilized to a solid substrate, such as a biochip.

[0283] The probe may have a length of from 8 to 500, 10 to 100 or 20 to 60 nucleotides. The probe may also have a length of at least 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280 or 300 nucleotides. The probe may further comprise a linker sequence of from 10-60 nucleotides. The probe may comprise a nucleic acid that is complementary to a sequence selected from the group consisting of SEQ ID NOS: 1-390 or variants thereof.

[0284] 8. Biochip

[0285] A biochip is also provided. The biochip may comprise a solid substrate comprising an attached probe or plurality of probes described herein. The probes may be capable of hybridizing to a target sequence under stringent hybridization conditions. The probes may be attached at spatially defined addresses on the substrate. More than one probe per target sequence may be used, with either overlapping probes or probes to different sections of a particular target sequence. The probes may be capable of hybridizing to target sequences associated with a single disorder appreciated by those in the art. The probes may either be synthesized first, with subsequent attachment to the biochip, or may be directly synthesized on the biochip.

[0286] The solid substrate may be a material that may be modified to contain discrete individual sites appropriate for the attachment or association of the probes and is amenable to at least one detection method. Representative examples of substrates include glass and modified or functionalized glass, plastics (including acrylics, polystyrene and copolymers of styrene and other materials, polypropylene, polyethylene, polybutylene, polyurethanes, TeflonJ, etc.), polysaccharides, nylon or nitrocellulose, resins, silica or silica-based materials including silicon and modified silicon, carbon, metals, inorganic glasses and plastics. The substrates may allow optical detection without appreciably fluorescing.

[0287] The substrate may be planar, although other configurations of substrates may be used as well. For example, probes may be placed on the inside surface of a tube, for flow-through sample analysis to minimize sample volume. Similarly, the substrate may be flexible, such as flexible foam, including closed cell foams made of particular plastics.

[0288] The biochip and the probe may be derivatized with chemical functional groups for subsequent attachment of the two. For example, the biochip may be derivatized with a chemical functional group including, but not limited to, amino groups, carboxyl groups, oxo groups or thiol groups. Using these functional groups, the probes may be attached using functional groups on the probes either directly or indirectly using a linker. The probes may be attached to the solid support by either the 5' terminus, 3' terminus, or via an internal nucleotide.

[0289] The probe may also be attached to the solid support non-covalently. For example, biotinylated oligonucleotides can be made, which may bind to surfaces covalently coated with streptavidin, resulting in attachment. Alternatively, probes may be synthesized on the surface using techniques such as photopolymerization and photolithography.

[0290] 9. Diagnostics

[0291] As used herein, the term "diagnosing" refers to classifying pathology, or a symptom, determining a severity of the pathology (e.g., grade or stage), monitoring pathology progression, forecasting an outcome of pathology and/or prospects of recovery.

[0292] As used herein, the phrase "subject in need thereof" refers to an animal or human subject who is known to have cancer, at risk of having cancer (e.g., a genetically predisposed subject, a subject with medical and/or family history of cancer, a subject who has been exposed to carcinogens, occupational hazard, environmental hazard) and/or a subject who exhibits suspicious clinical signs of cancer (e.g., blood in the stool or melena, unexplained pain, sweating, unexplained fever, unexplained loss of weight up to anorexia, changes in bowel habits (constipation and/or diarrhea), tenesmus (sense of incomplete defecation, for rectal cancer specifically), anemia and/or general weakness). Additionally or alternatively, the subject in need thereof can be a healthy human subject undergoing a routine well-being check up.

[0293] Analyzing presence of malignant or pre-malignant cells can be effected in vivo or ex vivo, whereby a biological sample (e.g., biopsy, blood) is retrieved. Such biopsy samples comprise cells and may be an incisional or excisional biopsy. Alternatively, the cells may be retrieved from a complete resection.

[0294] While employing the present teachings, additional information may be gleaned pertaining to the determination of treatment regimen, treatment course and/or to the measurement of the severity of the disease.

[0295] As used herein the phrase "treatment regimen" refers to a treatment plan that specifies the type of treatment, dosage, follow-up plans, schedule and/or duration of a treatment provided to a subject in need thereof (e.g., a subject diagnosed with a pathology). The selected treatment regimen can be an aggressive one which is expected to result in the best clinical outcome (e.g., complete cure of the pathology) or a more moderate one which may relieve symptoms of the pathology yet results in incomplete cure of the pathology. It will be appreciated that in certain cases the treatment regimen may be associated with some discomfort to the subject or adverse side effects (e.g., damage to healthy cells or tissue). The type of treatment can include a surgical intervention (e.g., removal of lesion, diseased cells, tissue, or organ), a cell replacement therapy, an administration of a therapeutic drug (e.g., receptor agonists, antagonists, hormones, chemotherapy agents) in a local or a systemic mode, an exposure to radiation therapy using an external source (e.g., external beam) and/or an internal source (e.g., brachytherapy) and/or any combination thereof. The dosage, schedule and duration of treatment can vary, depending on the severity of pathology and the selected type of treatment, and those of skill in the art are capable of adjusting the type of treatment with the dosage, schedule and duration of treatment.

[0296] A method of diagnosis is also provided. The method comprises detecting an expression level of a specific cancer-associated nucleic acid in a biological sample. The sample may be derived from a patient. Diagnosis of a specific cancer state in a patient may allow for prognosis and selection of therapeutic strategy. Further, the developmental stage of cells may be classified by determining temporarily expressed specific cancer-associated nucleic acids.

[0297] In situ hybridization of labeled probes to tissue arrays may be performed. When comparing the fingerprints between individual samples the skilled artisan can make a diagnosis, a prognosis, or a prediction based on the findings. It is further understood that the nucleic acid sequences which indicate the diagnosis may differ from those which indicate the prognosis and molecular profiling of the condition of the cells or exosomes may lead to distinctions between responsive or refractory conditions or may be predictive of outcomes.

[0298] 10. Kits

[0299] A kit is also provided and may comprise a nucleic acid described herein together with any or all of the following: assay reagents, buffers, probes and/or primers, and sterile saline or another pharmaceutically acceptable emulsion and suspension base. In addition, the kits may include instructional materials containing directions (e.g., protocols) for the practice of the methods described herein. The kit may further comprise a software package for data analysis of expression profiles.

[0300] For example, the kit may be a kit for the amplification, detection, identification or quantification of a target nucleic acid sequence. The kit may comprise a poly (T) primer, a forward primer, a reverse primer, and a probe.

[0301] Any of the compositions described herein may be comprised in a kit. In a non-limiting example, reagents for isolating miRNA, labeling miRNA, and/or evaluating a miRNA population using an array are included in a kit. The kit may further include reagents for creating or synthesizing miRNA probes. The kits will thus comprise, in suitable container means, an enzyme for labeling the miRNA by incorporating labeled nucleotide or unlabeled nucleotides that are subsequently labeled. It may also include one or more buffers, such as reaction buffer, labeling buffer, washing buffer, or a hybridization buffer, compounds for preparing the miRNA probes, components for in situ hybridization and components for isolating miRNA. Other kits of the invention may include components for making a nucleic acid array comprising miRNA, and thus may include, for example, a solid support.

[0302] The following examples are presented in order to more fully illustrate some embodiments of the invention. They should, in no way be construed, however, as limiting the broad scope of the invention.

EXAMPLES

Methods

1. Tumor Samples

[0303] 1300 primary and metastatic tumor FFPE were used in the study. Tumor samples were obtained from several sources. Institutional review approvals were obtained for all samples in accordance with each institute's institutional review board or IRB equivalent guidelines. Samples included primary tumors and metastases of defined origins, according to clinical records. Tumor content was at least 50% for >95% of samples, as determined by a pathologist based on hematoxylin-eosin (H&E) stained slides.

2. RNA Extraction

[0304] For FFPE samples, total RNA was isolated from seven to ten 10-.mu.m-thick tissue sections using the miR extraction protocol developed at Rosetta Genomics. Briefly, the sample was incubated a few times in xylene at 57.degree. C. to remove paraffin excess, followed by ethanol washes. Proteins were degraded by proteinase K solution at 45.degree. C. for a few hours. The RNA was extracted with acid phenol:chloroform followed by ethanol precipitation and DNAse digestion. Total RNA quantity and quality was checked by spectrophotometer (Nanodrop ND-1000).

3. miR Array Platform

[0305] Custom microarrays (Agilent Technologies, Santa Clara, Calif.) were produced by printing DNA oligonucleotide probes to: 982 miRs sequences, 17 negative controls, 23 spikes, and 10 positive controls (total of 1032 probes). Each probe, printed in triplicate, carried up to 28-nucleotide (nt) linker at the 3' end of the microRNA's complement sequence. 17 negative control probes were designed using as sequences which do not match the genome. Two groups of positive control probes were designed to hybridize to miR array: (i) synthetic small RNAs were spiked to the RNA before labeling to verify the labeling efficiency; and (ii) probes for abundant small RNA (e.g., small nuclear RNAs (U43, U24, Z30, U6, U48, U44), 5.8s and 5s ribosomal RNA are spotted on the array to verify RNA quality.

4. Cy-Dye Labeling of miRNA for miR Array

[0306] One .mu.g of total RNA were labeled by ligation (Thomson et al. Nature Methods 2004; 1:47-53) of an RNA-linker, p-rCrU-Cy/dye (Eurogentec or equivalent), to the 3' end with Cy3 or Cy5. The labeling reaction contained total RNA, spikes (0.1-100 fmoles), 400 ng RNA-linker-dye, 15% DMSO, 1.times. ligase buffer and 20 units of T4 RNA ligase (NEB), and proceeded at 4.degree. C. for 1 h, followed by 1 h at 37.degree. C., followed by 4.degree. C. up to 40 min.

[0307] The labeled RNA was mixed with 30 .mu.l hybridization mixture (mixture of 45 .mu.L of the 10.times.GE Agilent Blocking Agent and 246 .mu.L of 2.times. Hi-RPM Hybridization). The labeling mixture was incubated at 100.degree. C. for 5 minutes followed by ice incubation in water bath for 5 minutes. Slides were Hybridize at 54.degree. C. for 16-20 hours, followed by two washes. The first wash was conducted at room temperature with Agilent GE Wash Buffer 1 for 5 min followed by a second wash with Agilent GE Wash Buffer 2 at 37.degree. C. for 5 min

[0308] Arrays were scanned using an Agilent Microarray Scanner Bundle G2565BA (resolution of 5 .mu.m at XDR Hi 100%, XDR Lo 5%). Array images were analyzed using Feature Extraction 10.7 software (Agilent).

5. Array Signal Calculation and Normalization

[0309] Triplicate spots were combined to produce one signal for each probe by taking the logarithmic mean of reliable spots. All data were log 2-transformed and the analysis was performed in log 2-space. A reference data vector for normalization R was calculated by taking the median expression level for each probe across all samples. For each sample data vector S, a 2nd degree polynomial F was found so as to provide the best fit between the sample data and the reference data, such that R.apprxeq.F(S). Remote data points ("outliers") were not used for fitting the polynomial F. For each probe in the sample (element Si in the vector S), the normalized value (in log-space) Mi was calculated from the initial value Si by transforming it with the polynomial function F, so that Mi=F(Si).

6. Logistic Regression

[0310] The aim of a logistic regression model is to use several features, such as expression levels of several microRNAs, to assign a probability of belonging to one of two possible groups, such as two branches of a node in a binary decision-tree. Logistic regression models the natural log of the odds ratio, i.e., the ratio of the probability of belonging to the first group, for example, the left branch in a node of a binary decision-tree (P) over the probability of belonging to the second group, for example, the right branch in such a node (1-P), as a linear combination of the different expression levels (in log-space). The logistic regression assumes that:

ln ( P 1 - P ) = .beta. 0 + i = 1 N .beta. i M i = .beta. 0 + .beta. 1 M 1 + .beta. 2 M 2 + , ##EQU00001##

[0311] where .beta..sub.0 is the bias, M.sub.i is the expression level (normalized, in log 2-space) of the i-th microRNA used in the decision node, and .beta..sub.i is its corresponding coefficient. .beta.i>0 indicates that the probability to take the left branch (P) increases when the expression level of this microRNA (Mi) increases, and the opposite for .beta.i<0. If a node uses only a single microRNA (M), then solving for P results in:

P = .beta. 0 + .beta. i M 1 + .beta. 0 + .beta. 1 M . ##EQU00002##

[0312] The regression error on each sample is the difference between the assigned probability P and the true "probability" of this sample, i.e., 1 if this sample is in the left branch group and 0 otherwise. The training and optimization of the logistic regression model calculates the parameters .beta. and the p-values [for each microRNA by the Wald statistic and for the overall model by the .chi.2 (chi-square) difference], maximizing the likelihood of the data given the model and minimizing the total regression error

Samples in first group ( 1 - P j ) + Samples in second group P j . ##EQU00003##

[0313] The probability output of the logistic model is here converted to a binary decision by comparing P to a threshold, denoted by P.sub.TH, i.e., if P.gtoreq.P.sub.TH then the sample belongs to the left branch ("first group") and vice versa. Choosing at each node the branch which has a probability >0.5, i.e., using a probability threshold of 0.5, leads to a minimization of the sum of the regression errors. However, as the goal was the minimization of the overall number of misclassifications (and not of their probability), a modification which adjusts the probability threshold (P.sub.TH) was used in order to minimize the overall number of mistakes at each node (Table 2). For each node the threshold to a new probability threshold P.sub.TH was optimized such that the number of classification errors is minimized. This change of probability threshold is equivalent (in terms of classifications) to a modification of the bias .beta..sub.0, which may reflect a change in the prior frequencies of the classes. Once the threshold was chosen .beta..sub.0 was modified such that the threshold will be shifted back to 0.5. In addition, .beta.0, .beta.1, .beta.2, . . . were adjusted so that the slope of the log of the odds ratio function is limited.

7. Stepwise Logistic Regression and Feature Selection

[0314] The original data contain the expression levels of multiple microRNAs for each sample, i.e., multiple of data features. In training the classifier for each node, only a small subset of these features was selected and used for optimizing a logistic regression model. In the initial training this was done using a forward stepwise scheme. The features were sorted in order of decreasing log-likelihoods, and the logistic model was started off and optimized with the first feature. The second feature was then added, and the model re-optimized. The regression error of the two models was compared: if the addition of the feature did not provide a significant advantage (a .chi.2 difference less than 7.88, p-value of 0.005), the new feature was discarded. Otherwise, the added feature was kept. Adding a new feature may make a previous feature redundant (e.g., if they are very highly correlated). To check for this, the process iteratively checks if the feature with lowest likelihood can be discarded (without losing .chi.2 difference as above). After ensuring that the current set of features is compact in this sense, the process continues to test the next feature in the sorted list, until features are exhausted. No limitation on the number of features was inserted into the algorithm.

[0315] The stepwise logistic regression method was used on subsets of the training set samples by re-sampling the training set with repetition ("bootstrap"), so that each of the 20 runs contained somewhat different training set. All the features that took part in one of the 20 models were collected. A robust set of 1-3 features per each node was selected by comparing features that were repeatedly chosen in the bootstrap sets to previous evidence, and considering their signal strengths and reliability. When using these selected features to construct the classifier, the stepwise process was not used and the training optimized the logistic regression model parameters only.

8. K-Nearest-Neighbors (KNN) Classification Algorithm

[0316] The KNN algorithm (see e.g., Ma et al., Arch Pathol Lab Med 2006; 130:465-73) calculates the distance (Pearson correlation) of any sample to all samples in the training set, and classifies the sample by the majority vote of the k samples which are most similar (k being a parameter of the classifier). The correlation is calculated on the pre-defined set of microRNAs (the microRNAs that were used by the decision-tree). KNN algorithms with k=1; 10 were compared, and the optimal performer was selected, using k=5. The KNN was based on comparing the expression of all 65 microRNAs in each sample to all other samples in the training database.

9. Reporting a Final Answer (Prediction):

[0317] The decision-tree and KNN each return a predicted tissue of origin and histological type where applicable. The tissue of origin and histological type may be one of the exact origins and types in the training or a variant thereof. For example, whereas the training includes brain oligodendroglioma and brain astrocytoma, the answer may simply be brain carcinoma. In addition to the tissue of origin and histological type, the KNN and decision-tree each return a confidence measure. The KNN returns the number of samples within the K nearest neighbors that agreed with the answer reported by the KNN (denoted by V), and the decision-tree returns the probability of the result (P), which is the multiplication of the probabilities at each branch point made on the way to that answer. The classifier returns the two different predictions or a single prediction in case the predictions concur, can be unified into a single answer (for example into the prediction brain if the KNN returned brain oligodendroglioma and the decision-tree brain astrocytoma), or if based on V and P, one answer is chosen to override the other.

Example 1

Decision-Tree Classification Algorithm

[0318] A tumor classifier was built using the microRNA expression levels by applying a binary tree classification scheme (FIGS. 1A-F). This framework is set up to utilize the specificity of microRNAs in tissue differentiation and embryogenesis: different microRNAs are involved in various stages of tissue specification, and are used by the algorithm at different decision points or "nodes". The tree breaks up the complex multi-tissue classification problem into a set of simpler binary decisions. At each node, classes which branch out earlier in the tree are not considered, reducing interference from irrelevant samples and further simplifying the decision. The decision at each node can then be accomplished using only a small number of microRNA biomarkers, which have well-defined roles in the classification (Table 2). The structure of the binary tree was based on a hierarchy of tissue development and morphological similarity.sup.18, which was modified by prominent features of the microRNA expression patterns. For example, the expression patterns of microRNAs indicated a significant difference between germ cell tumors and tumors of non-germ cell origin, and these are therefore distinguished at node #1 (FIG. 2) into separate branches (FIG. 1A).

[0319] For each of the individual nodes logistic regression models were used, a robust family of classifiers which are frequently used in epidemiological and clinical studies to combine continuous data features into a binary decision (FIGS. 2-25 and Methods). Since gene expression classifiers have an inherent redundancy in selecting the gene features, bootstrapping was used on the training sample set as a method to select a stable microRNA set for each node (Methods). This resulted in a small number (usually 2-3) of microRNA features per node, totaling 65 microRNAs for the full classifier (Table 2). This approach provides a systematic process for identifying new biomarkers for differential expression.

TABLE-US-00002 TABLE 2 microRNAs used per class in the tree classifier miR List: Class hsa-miR-372 (SEQ ID NO: 55) Germ cell cancer hsa-miR-372, hsa-miR-122 (SEQ ID NO: 6), hsa-miR-126 (SEQ ID Biliary tract NO: 9), hsa-miR-200b (SEQ ID NO: 29) adenocarcinoma hsa-miR-372, hsa-miR-122, hsa-miR-126, hsa-miR-200b Hepatocellular carcinoma (HCC) hsa-miR-372, hsa-miR-122, hsa-miR-200c (SEQ ID NO: 30), hsa-miR- Brain tumor 30a (SEQ ID NO: 46), hsa-miR-146a (SEQ ID NO: 16), hsa-let-7e (SEQ ID NO: 156), hsa-miR-9* (SEQ ID NO: 66), hsa-miR-92b (SEQ ID NO: 68) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Brain - 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-222 (SEQ ID oligodendroglioma NO: 40), hsa-miR-497 (SEQ ID NO: 60) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Brain - astrocytoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-222, hsa-miR-497 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375 Prostate (SEQ ID NO: 56), hsa-miR-7 (SEQ ID NO: 65), hsa-miR-193a-3p (SEQ Adenocarcinoma ID NO: 25), hsa-miR-194 (SEQ ID NO: 27), hsa-miR-21* (SEQ ID NO: 35), hsa-miR-143 (SEQ ID NO: 14), hsa-miR-181a (SEQ ID NO: 21) hsa-miR-372 Ovarian primitive germ cell tumor hsa-miR-372 Testis hsa-miR-372, hsa-miR-200b, hsa-miR-516a-5p (SEQ ID NO: 62) Seminomatous testicular germ cell tumor hsa-miR-372, hsa-miR-200b, hsa-miR-516a-5p Non seminomatous testicular germ cell tumor hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-7, Breast hsa-miR-194, hsa-miR-21*, hsa-miR-143, hsa-miR-181a, hsa-miR-205 adenocarcinoma (SEQ ID NO: 32), hsa-miR-345 (SEQ ID NO: 51), hsa-miR-125a-5p (SEQ ID NO: 7), hsa-miR-193a-3p (SEQ ID NO: 25), hsa-miR-375, hsa- miR-342-3p (SEQ ID NO: 50) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-7, Ovarian carcinoma hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR-143, hsa-miR- 181a, hsa-miR-345, hsa-miR-125a-5p, hsa-miR-193a-3p, hsa-miR-375, hsa-miR-342-3p, hsa-miR-205 (SEQ ID NO: 32), hsa-miR-10a (SEQ ID NO: 4), hsa-miR-22 (SEQ ID NO: 39) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Thyroid carcinoma hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-138 (SEQ ID NO: 11), hsa-miR-93 (SEQ ID NO: 148), hsa-miR- 10a (SEQ ID NO: 4) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Thyroid carcinoma hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- follicular 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-138, hsa-miR-93, hsa-miR-10a, hsa-miR-146b-5p (SEQ ID NO: 17), hsa-miR-21 (SEQ ID NO: 34) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Thyroid carcinoma hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- papillary 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-138, hsa-miR-93, hsa-miR-10a, hsa-miR-146b-5p, hsa-miR-21 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Breast hsa-miR-7, hsa-miR-194, hsa-miR-21*, hsa-miR-143, hsa-miR-181a, adenocarcinoma hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa-miR-138, hsa-miR- 93, hsa-miR-10a, hsa-miR-193a-3p (SEQ ID NO: 25), hsa-miR-31 (SEQ ID NO: 49), hsa-miR-92a (SEQ ID NO: 67) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Lung large cell or hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- adenocarcinoma 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-93, hsa-miR-10a, hsa-miR-193a-3p, hsa-miR-31, hsa-miR-92a, hsa-miR-138 (SEQ ID NO: 11), hsa-miR-378 (SEQ ID NO: 57), hsa- miR-21 (SEQ ID NO: 34) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Ovarian carcinoma hsa-miR-7, hsa-miR-194, hsa-miR-21*, hsa-miR-143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa-miR-93, hsa-miR- 10a, hsa-miR-193a-3p, hsa-miR-31, hsa-miR-92a, hsa-miR-138, hsa- miR-378, hsa-miR-21 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Thymoma hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-342-3p, hsa-miR-10a, hsa-miR-22, hsa-miR-100, hsa-miR-21 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Urothelial carcinoma hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- (TCC) 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-342-3p, hsa-miR-205, hsa-miR-10a, hsa-miR-22, hsa-miR-100, hsa-miR-21, hsa-miR-934 (SEQ ID NO: 69), hsa-miR-191 (SEQ ID NO: 24), hsa-miR-29c (SEQ ID NO: 44) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Squamous cell hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- carcinoma (SCC) 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-342-3p, hsa-miR-10a, hsa-miR-22, hsa-miR-100, hsa-miR-21, hsa- miR-934, hsa-miR-191, hsa-miR-29c hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Uterine cervix SCC hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-342-3p, hsa-miR-10a, hsa-miR-22, hsa-miR-100, hsa-miR-21, hsa- miR-934, hsa-miR-191, hsa-miR-29c, hsa-miR-10b (SEQ ID NO: 5), hsa-let-7c (SEQ ID NO: 1), hsa-miR-361-5p (SEQ ID NO: 54) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Anus or Skin SCC hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-193a-3p, hsa-miR-375, hsa-miR-342-3p, hsa-miR-205, hsa-miR- 10a, hsa-miR-22, hsa-miR-100, hsa-miR-21, hsa-miR-934, hsa-miR- 191, hsa-miR-29c, hsa-miR-10b, hsa-let-7c, hsa-miR-361-5p, hsa-miR- 138, hsa-miR-185 (SEQ ID NO: 23) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Lung, Head& Neck or hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- Esophagus SCC 143, hsa-miR-181a, hsa-miR-205, hsa-miR-345, hsa-miR-125a-5p, hsa- miR-342-3p, hsa-miR-10a, hsa-miR-22, hsa-miR-100, hsa-miR-21, hsa- miR-934, hsa-miR-191, hsa-miR-29c, hsa-let-7c, hsa-miR-361-5p, hsa- miR-10b, hsa-miR-138, hsa-miR-185 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-146a Melanoma (SEQ ID NO: 16), hsa-let-7e (SEQ ID NO: 2), hsa-miR-30d (SEQ ID NO: 47), hsa-miR-342-3p hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Lymphoma 146a, hsa-let-7e, hsa-miR-30d, hsa-miR-342-3p hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- B cell lymphoma 146a, hsa-let-7e, hsa-miR-30d, hsa-miR-342-3p, hsa-miR-21*, hsa- miR-30e (SEQ ID NO: 48) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- T cell lymphoma 146a, hsa-let-7e, hsa-miR-30a, hsa-miR-30d, hsa-miR-342-3p, hsa- miR-21*, hsa-miR-30e hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Lung small cell hsa-miR-7, hsa-miR-193a-3p, hsa-miR-17 (SEQ ID NO: 20), hsa-miR- carcinoma 29c* (SEQ ID NO: 45) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Medullary thyroid hsa-miR-7, hsa-miR-193a-3p, hsa-miR-17, hsa-miR-29c*, hsa-miR-222 carcinoma (SEQ ID NO: 40), hsa-miR-92a (SEQ ID NO: 67), hsa-miR-92b (SEQ ID NO: 68) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Lung carcinoid hsa-miR-7, hsa-miR-193a-3p, hsa-miR-17, hsa-miR-29c*, hsa-miR- 222, hsa-miR-92a, hsa-miR-92b, hsa-miR-652 (SEQ ID NO: 64), hsa- miR-34c-5p (SEQ ID NO: 53), hsa-miR-214 (SEQ ID NO: 37) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Gastrointestinal (GI) hsa-miR-7, hsa-miR-193a-3p, hsa-miR-17, hsa-miR-29c*, hsa-miR- tract carcinoid 222, hsa-miR-92a, hsa-miR-92b, hsa-miR-652, hsa-miR-34c-5p, hsa- miR-214, hsa-miR-21 (SEQ ID NO: 34), hsa-miR-148a (SEQ ID NO: 18) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Pancreas islet cell hsa-miR-7, hsa-miR-193a-3p, hsa-miR-17, hsa-miR-29c*, hsa-miR- tumor 222, hsa-miR-92a, hsa-miR-92b, hsa-miR-652, hsa-miR-34c-5p, hsa- miR-214, hsa-miR-21, hsa-miR-148a hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Gastric or Esophageal hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194 (SEQ ID NO: 27), hsa-miR- Adenocarcinoma 21*(SEQ ID NO: 35), hsa-miR-224 (SEQ ID NO: 42), hsa-miR-210 (SEQ ID NO: 36), hsa-miR-1201 (SEQ ID NO: 146) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Colorectal hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- Adenocarcinoma 224, hsa-miR-210, hsa-miR-1201, hsa-miR-17 (SEQ ID NO: 20), hsa- miR-29a (SEQ ID NO: 43) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Pancreas or bile hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- 224, hsa-miR-210, hsa-miR-1201, hsa-miR-17, hsa-miR-29a hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Pancreatic hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- adenocarcinoma 224, hsa-miR-210, hsa-miR-1201, hsa-miR-17, hsa-miR-29a, hsa-miR- 345 (SEQ ID NO: 51), hsa-miR-31 (SEQ ID NO: 49), hsa-miR-146a (SEQ ID NO: 16) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-375, Biliary tract hsa-miR-7, hsa-miR-193a-3p, hsa-miR-194, hsa-miR-21*, hsa-miR- adenocarcinoma 224, hsa-miR-210, hsa-miR-1201, hsa-miR-17, hsa-miR-29a, hsa-miR- 345, hsa-miR-31, hsa-miR-146a hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR-146a Renal cell carcinoma (SEQ ID NO: 16), hsa-let-7e, hsa-miR-9* (SEQ ID NO: 66), hsa-miR- chromophobe 92b (SEQ ID NO: 68), hsa-miR-149 (SEQ ID NO: 19), hsa-miR-200b (SEQ ID NO: 29) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Pheochromocytoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-30a, hsa-miR-149, hsa-miR-200b, hsa-miR-7 (SEQ ID NO: 65), hsa-miR-375 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Adrenocortical 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202 (SEQ ID NO: 31), hsa- miR-214* (SEQ ID NO: 38), hsa-miR-509-3p (SEQ ID NO: 61) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Gastrointestinal 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- stromal tumor (GIST) 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR- 509-3p, hsa-miR-143 (SEQ ID NO: 14), hsa-miR-29c* hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Renal cell carcinoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- chromophobe 200b, hsa-miR-210 (SEQ ID NO: 36), hsa-miR-221 (SEQ ID NO: 147) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Renal cell carcinoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- clear cell 200b, hsa-miR-210, hsa-miR-221, hsa-miR-31 (SEQ ID NO: 49), hsa- miR-126 (SEQ ID NO: 9) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Renal cell carcinoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- papillary 200b, hsa-miR-210, hsa-miR-221, hsa-miR-31, hsa-miR-126 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Pleural mesothelioma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7 (SEQ ID NO: 65), hsa-miR-375, hsa-miR-202 (SEQ ID NO: 31), hsa-miR-214* (SEQ ID NO: 38), hsa-miR-509-3p (SEQ ID NO: 61), hsa-miR-143 (SEQ ID NO: 14), hsa-miR-29c*, hsa-miR-21* (SEQ ID NO: 35), hsa-miR-130a (SEQ ID NO: 10), hsa-miR-10b (SEQ ID NO: 5) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Sarcoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR- 509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR-21*, hsa-miR-130a, hsa- miR-10b hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Synovial

sarcoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR- 509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR-21*, hsa-miR-130a, hsa- miR-10b, hsa-miR-100 (SEQ ID NO: 3), hsa-miR-222 (SEQ ID NO: 40), hsa-miR-145 (SEQ ID NO: 15) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Chondrosarcoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR- 509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR-21*, hsa-miR-130a, hsa- miR-10b, hsa-miR-100, hsa-miR-222, hsa-miR-145, hsa-miR-140-3p (SEQ ID NO: 12), hsa-miR-455-5p (SEQ ID NO: 58) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Liposarcoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR- 509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR-21*, hsa-miR-130a, hsa- miR-10b, hsa-miR-100, hsa-miR-222, hsa-miR-145, hsa-miR-140-3p, hsa-miR-455-5p, hsa-miR-210 (SEQ ID NO: 36), hsa-miR-193a-5p (SEQ ID NO: 26) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Ewing sarcoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR- 509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR-21*, hsa-miR-130a, hsa- miR-10b, hsa-miR-100, hsa-miR-222, hsa-miR-145, hsa-miR-140-3p, hsa-miR-455-5p, hsa-miR-210, hsa-miR-193a-5p, hsa-miR-181a, hsa- miR-193a-3p (SEQ ID NO: 25), hsa-miR-31 (SEQ ID NO: 49) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Osteosarcoma 146a, hsa-let-7e, hsa-miR-9*, hsa-miR-92b, hsa-miR-149, hsa-miR- 200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR- 509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR-21*, hsa-miR-130a, hsa- miR-10b, hsa-miR-100, hsa-miR-222, hsa-miR-145, hsa-miR-140-3p, hsa-miR-455-5p, hsa-miR-210, hsa-miR-193a-5p, hsa-miR-181a, hsa- miR-193a-3p, hsa-miR-31 hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Rhabdomyo sarcoma 146a, hsa-let-7e, hsa-miR-30a, hsa-miR-9*, hsa-miR-92b, hsa-miR-30a, hsa-miR-149, hsa-miR-200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, hsa-miR-214*, hsa-miR-509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR- 21*, hsa-miR-130a, hsa-miR-10b, hsa-miR-100, hsa-miR-222, hsa- miR-145, hsa-miR-140-3p, hsa-miR-455-5p, hsa-miR-210, hsa-miR- 193a-5p, hsa-miR-181a, hsa-miR-487b (SEQ ID NO: 59), hsa-miR-22 (SEQ ID NO: 39), hsa-miR-206 (SEQ ID NO: 33) hsa-miR-372, hsa-miR-122, hsa-miR-200c, hsa-miR-30a, hsa-miR- Malignant fibrous 146a, hsa-let-7e, hsa-miR-30a, hsa-miR-9*, hsa-miR-92b, hsa-miR-30a, histiocytoma (MFH) hsa-miR-149, hsa-miR-200b, hsa-miR-7, hsa-miR-375, hsa-miR-202, or fibresarcoma hsa-miR-214*, hsa-miR-509-3p, hsa-miR-143, hsa-miR-29c*, hsa-miR- 21*, hsa-miR-130a, hsa-miR-10b, hsa-miR-100, hsa-miR-222, hsa- miR-145, hsa-miR-140-3p, hsa-miR-455-5p, hsa-miR-210, hsa-miR- 193a-5p, hsa-miR-181a, hsa-miR-487b, hsa-miR-22, hsa-miR-206

Example 2

Expression of miRs Provides for Distinguishing Between Tumors

TABLE-US-00003 [0320] TABLE 3 miR expression (in fluorescence units) distinguishing between the group consisting of germ-cell tumors and the group consisting of all other tumors SEQ fold- ID median values change p-value NO. miR name 2.7e+004-5.0e+001 545.73 (+) <e-240 233 hsa-miR-373 1.8e+004-5.0e+001 365.93 (+) <e-240 55 hsa-miR-372 8.6e+003-5.0e+001 171.72 (+) <e-240 200 hsa-miR-371-3p 5.9e+003-5.1e+001 115.94 (+) 7.3e-249.sup. 201 hsa-miR-371-5p (+) for all the listed miRs, the higher expression is in tumors from a germ-cell origin.

[0321] hsa-miR-372 (SEQ ID NO: 55) is used at node 1 of the binary-tree-classifier detailed in the invention to distinguish between germ-cell tumors and all other tumors.

[0322] FIGS. 2A-D are boxplot presentations comparing distribution of the expression of the statistically significant miRs in tumor samples from the "germ cell" class (left box) and "non germ cell" class (right box).

TABLE-US-00004 TABLE 4 miR expression (in fluorescence units) distinguishing between the group consisting of hepatobiliary tumors and the group consisting of non germ-cell non-hepatobiliary tumors SEQ fold- ID medianvalues change p-value NO. miR name 1.0e+005-5.0e+001 2024.31 (+) 1.1e-123 6 hsa-miR-122 7.4e+001-8.1e+003 109.63 (-) 3.6e-010 30 hsa-miR-200c 5.0e+001-1.4e+003 27.92 (-) 4.8e-010 13 hsa-miR-141 (+) the higher expression of this miR is in tumors from a hepatobiliary origin (-) the higher expression of this miR is in tumors from a non germ-cell, non-hepatobiliary origin

[0323] hsa-miR-122 (SEQ ID NO: 6) is used at node 2 of the binary-tree-classifier detailed in the invention to distinguish between hepatobiliary tumors and non germ-cell non-hepatobiliary tumors.

TABLE-US-00005 TABLE 5 miR expression (in fluorescence units) distinguishing between the group consisting of liver tumors and the group consisting of biliary- tract carcinomas (cholangiocarcinoma or gallbladder adenocarcinoma) SEQ fold- ID median values change p-value NO. miR name 6.1e+003-4.1e+002 14.74 (+) 5.5e-005 28 hsa-miR-200a 9.7e+003-9.0e+002 10.74 (+) 2.4e-004 29 hsa-miR-200b 1.9e+003-7.0e+003 3.67 (-) 8.5e-004 231 hsa-miR-99a 3.3e+003-7.5e+003 2.28 (-) 6.2e-004 9 hsa-miR-126 (+) the higher expression of this miR is in biliary tract carcinomas (-) the higher expression of this miR is in liver tumors

[0324] hsa-miR-126 (SEQ ID NO: 9) and hsa-miR-200b (SEQ ID NO: 29) are used at node 3 of the binary-tree-classifier detailed in the invention to distinguish between liver tumors and biliary-tract carcinoma.

[0325] FIG. 3 demonstrates that tumors of hepatocellular carcinoma (HCC) origin (marked by squares) are easily distinguished from tumors of biliary tract adenocarcinoma origin (marked by diamonds) using the expression levels of hsa-miR-200b (SEQ ID NO: 29, y-axis) and hsa-miR-126 (SEQ ID NO: 9, x-axis).

TABLE-US-00006 TABLE 6 miR expression (in fluorescence units) distinguishing between the group consisting of tumors from an epithelial origin and the group consisting of tumors from a non-epithelial origin SEQ fold- ID median values change p-value NO. miR name 1.5e+004-7.7e+001 196.43 (+) 1.5e-300 30 hsa-miR-200c 9.0e+003-5.0e+001 180.07 (+) 1.3e-208 29 hsa-miR-200b 3.9e+003-5.0e+001 78.09 (+) 2.2e-187 28 hsa-miR-200a 2.7e+003-5.0e+001 54.64 (+) 7.0e-078 32 hsa-miR-205 2.6e+003-5.0e+001 51.98 (+) 1.2e-265 13 hsa-miR-141 5.4e+002-9.2e+001 5.90 (+) 6.3e-048 152 hsa-miR-182 1.1e+003-2.5e+002 4.35 (+) 4.8e-022 49 hsa-miR-31 (+) for all the listed miRs, the higher expression is in tumors from epithelial origins

[0326] A combination of the expression level of any of the miRs detailed in table 6 with the expression level of any of hsa-miR-30a (SEQ ID NO: 46), hsa-miR-10b (SEQ ID NO: 5) and hsa-miR-140-3p (SEQ ID NO: 12) also provides for distinguishing between tumors from epithelial origins and tumors from non-epithelial origins. This is demonstrated at node 4 of the binary-tree-classifier detailed in the invention with hsa-miR-200c (SEQ ID NO: 30) and hsa-miR-30a (SEQ ID NO: 46) (FIG. 4). Tumors originating in epithelial (diamonds) are easily distinguished from tumors of non-epithelial origin (squares) using the expression levels of hsa-miR-30a (SEQ ID NO: 46, y-axis) and hsa-miR-200c (SEQ ID NO: 30, x-axis).

TABLE-US-00007 TABLE 7 miR expression (in fluorescence units) distinguishing between the group consisting of melanoma and lymphoma (B-cell, T-cell), and the group consisting of all other non-epithelial tumors SEQ fold- ID median values change p-value NO. miR name 2.0e+003-7.0e+001 28.25 (-) 1.9e-074 164 hsa-miR-142-5p 1.2e+004-6.3e+002 18.86 (-) 6.0e-061 168 hsa-miR-150 5.4e+003-3.1e+002 17.29 (-) 5.6e-060 170 hsa-miR-155 4.2e+003-3.5e+002 12.03 (-) 8.4e-068 16 hsa-miR-146a 5.9e+002-1.4e+002 4.25 (-) 8.2e-048 198 hsa-miR-342-5p 7.5e+003-1.9e+003 4.02 (-) 4.8e-056 50 hsa-miR-342-3p 8.9e+002-2.5e+002 3.53 (-) 6.0e-035 176 hsa-miR-18a 4.4e+003-1.4e+003 3.28 (-) 8.0e-038 186 hsa-miR-20a 7.9e+002-2.6e+002 3.03 (-) 7.3e-005 11 hsa-miR-138 6.6e+003-2.3e+003 2.82 (-) 4.0e-039 158 hsa-miR-106a 4.1e+003-1.4e+003 2.82 (-) 2.4e-037 20 hsa-miR-17 6.2e+001-5.9e+002 9.53 (+) 3.7e-027 155 hsa-miR-127-3p 1.2e+003-7.0e+003 5.71 (+) 1.5e-047 231 hsa-miR-99a 3.9e+002-1.7e+003 4.25 (+) 6.6e-022 4 hsa-miR-10a 1.0e+004-4.1e+004 3.91 (+) 3.2e-037 8 hsa-miR-125b 6.5e+002-2.2e+003 3.37 (+) 2.4e-023 46 hsa-miR-30a 1.9e+003-5.6e+003 2.98 (+) 1.0e-025 3 hsa-miR-100 2.5e+003-7.1e+003 2.89 (+) 1.8e-051 2 hsa-let-7e 2.9e+003-8.4e+003 2.86 (+) 8.1e-047 7 hsa-miR-125a-5p (+) the higher expression of this miR is in the group of non-epithelial tumors excluding melanoma and lymphoma (-) the higher expression of this miR is in the group consisting of melanoma and lymphoma

[0327] hsa-miR-146a (SEQ ID NO: 16), hsa-let-7e (SEQ ID NO: 2) and hsa-miR-30a (SEQ ID NO: 46) are used at node 5 of the binary-tree-classifier detailed in the invention to distinguish between the group consisting of melanoma and lymphoma, and the group consisting of all other non-epithelial tumors. FIG. 5 demonstrates that tumors originating in the lymphoma or melanoma (diamonds) are easily distinguished from tumors of non epithelial, non lymphoma/melanoma origin (squares) using the expression levels of hsa-miR-146a (SEQ ID NO: 16, y-axis), hsa-miR-30a (SEQ ID NO: 46, x-axis) and hsa-let-7e (SEQ ID NO: 2, z-axis).

TABLE-US-00008 TABLE 8 miR expression (in fluorescence units) distinguishing between the group consisting of brain tumors (astrocytic tumor and oligodendroglioma) and the group consisting of all non-brain, non-epithelial tumors SEQ fold- ID median values change p-value NO. miR name 9.1e+003-5.0e+001 182.94 (+) 3.8e-059 159 hsa-miR-124 4.4e+003-5.0e+001 88.33 (+) 1.1e-125 66 hsa-miR-9* 2.1e+003-6.0e+001 34.97 (+) 6.0e-035 225 hsa-miR-551b 9.9e+002-5.0e+001 19.73 (+) 3.0e-116 187 hsa-miR-219-2-3p 6.5e+002-5.0e+001 12.95 (+) 1.8e-021 162 hsa-miR-129-3p 1.1e+003-1.0e+002 10.52 (+) 2.0e-034 161 hsa-miR-128 2.3e+003-2.5e+002 9.45 (+) 2.2e-052 68 hsa-miR-92b 5.2e+002-6.8e+001 7.61 (+) 6.7e-019 232 hsa-miR-99a* 6.9e+002-9.2e+001 7.45 (+) 5.5e-023 173 hsa-miR-181c 2.2e+003-3.5e+002 6.34 (+) 7.4e-007 11 hsa-miR-138 1.2e+005-2.4e+004 4.78 (+) 1.7e-014 8 hsa-miR-125b 1.8e+003-3.9e+002 4.70 (+) 7.2e-014 174 hsa-miR-181d 8.5e+002-1.8e+002 4.64 (+) 2.2e-002 155 hsa-miR-127-3p 1.6e+004-3.5e+003 4.60 (+) 2.4e-010 231 hsa-miR-99a 8.5e+001-1.1e+003 13.55 (-) 2.4e-014 4 hsa-miR-10a 7.7e+002-6.6e+003 8.58 (-) 8.4e-017 182 hsa-miR-199a-5p 5.7e+002-4.7e+003 8.12 (-) 1.8e-013 181 hsa-miR-199a-3p 2.8e+002-1.9e+003 6.81 (-) 1.4e-012 37 hsa-miR-214 (+) the higher expression of this miR is in the group consisting of brain tumors (-) the higher expression of this miR is in the group consisting of all non-brain, non-epithelial tumors

[0328] hsa-miR-9* (SEQ ID NO: 66) and hsa-miR-92b (SEQ ID NO: 68) are used at node 6 of the binary-tree-classifier detailed in the invention to distinguish between brain tumors and the group consisting of all non-brain, non-epithelial tumors. FIG. 6 demonstrates that tumors originating in the brain (marked by diamonds) are easily distinguished from tumors of non epithelial, non brain origin (marked by squares) using the expression levels of hsa-miR-9* (SEQ ID NO: 66, y-axis) and hsa-miR-92b (SEQ ID NO: 68, x-axis).

TABLE-US-00009 TABLE 9 miR expression (in fluorescence units) distinguishing between astrocytic tumors and oligodendrogliomas SEQ fold- ID median values change p-value NO. miR name 2.5e+003-2.3e+002 11.10 (+) 5.1e-011 230 hsa-miR-886-5p 4.4e+003-4.9e+002 9.06 (+) 1.1e-009 228 hsa-miR-886-3p 1.0e+004-1.7e+003 5.99 (+) 7.7e-008 147 hsa-miR-221 1.3e+004-2.6e+003 5.03 (+) 2.6e-006 40 hsa-miR-222 3.3e+004-7.3e+003 4.54 (+) 3.9e-004 34 hsa-miR-21 8.4e+002-2.2e+002 3.78 (+) 3.7e-006 206 hsa-miR-455-3p 6.0e+002-1.8e+002 3.30 (+) 1.3e-002 35 hsa-miR-21* 5.8e+003-1.8e+003 3.15 (+) 2.4e-005 52 hsa-miR-34a 1.1e+003-3.5e+002 3.04 (+) 1.0e-003 25 hsa-miR-193a-3p 1.6e+002-8.2e+002 5.17 (-) 1.2e-004 229 hsa-miR-9 4.6e+002-2.3e+003 5.09 (-) 7.1e-003 161 hsa-miR-128 4.1e+002-1.8e+003 4.43 (-) 1.3e-002 187 hsa-miR-219-2-3p 3.8e+003-1.3e+004 3.31 (-) 1.9e-002 179 hsa-miR-195 (+) the higher expression of this miR is in astrocytic tumors (-) the higher expression of this miR is in oligodendrogliomas

[0329] A combination of the expression level of any of the miRs detailed in table 9 with the expression level of hsa-miR-497 (SEQ ID NO: 208) or hsa-let-7d (SEQ ID NO: 153) also provides for classification of brain tumors as astrocytic tumors or oligodendrogliomas. This is demonstrated at node 7 of the binary-tree-classifier detailed in the invention with hsa-miR-222 (SEQ ID NO: 40) and hsa-miR-497 (SEQ ID NO: 208). In another embodiment of the invention, the expression levels of hsa-miR-222 (SEQ ID NO: 40) and hsa-let-7d (SEQ ID NO: 153) are combined to distinguish between astrocytic tumors and oligodendrogliomas.

[0330] FIG. 7 demonstrates that tumors originating in astrocytoma (marked by diamonds) are easily distinguished from tumors of oligodendroglioma origins (marked by squares) using the expression levels of hsa-miR-497 (SEQ ID NO: 208, y-axis) and hsa-miR-222 (SEQ ID NO: 40, x-axis).

TABLE-US-00010 TABLE 10 miR expression (in fluorescence units) distinguishing between the group consisting of neuroendocrine tumors and the group consisting of all non-neuroendocrine, epithelial tumors SEQ fold- ID median values change p-value NO. miR name 3.8e+004-1.5e+002 259.47 (+) 5.3e-086 56 hsa-miR-375 3.6e+003-5.2e+001 70.47 (+) 4.4e-145 65 hsa-miR-7 1.3e+003-1.8e+002 6.89 (+) 4.7e-044 175 hsa-miR-183 1.9e+003-4.4e+002 4.42 (+) 3.5e-025 152 hsa-miR-182 1.2e+003-3.0e+002 4.16 (+) 5.5e-028 155 hsa-miR-127-3p 5.6e+001-7.0e+003 124.66 (-) 1.4e-023 32 hsa-miR-205 1.5e+002-1.4e+003 9.25 (-) 1.8e-019 49 hsa-miR-31 3.4e+002-1.4e+003 4.12 (-) 9.5e-032 35 hsa-miR-21* (+) the higher expression of this miR is in the group consisting of neuroendocrine tumors (-) the higher expression of this miR is in the group consisting of all non-neuroendocrine, epithelial tumors

[0331] hsa-miR-375 (SEQ ID NO: 56), hsa-miR-7 (SEQ ID NO: 65) and hsa-miR-193a-3p (SEQ ID NO: 25) are used at node 8 of the binary-tree-classifier detailed in the invention to distinguish between the group consisting of neuroendocrine tumors and the group consisting of all non-neuroendocrine, epithelial tumors. FIG. 8 demonstrates that tumors originating in the neuroendocrine (diamonds) are easily distinguished from tumors of epithelial, origin (squares) using the expression levels of hsa-miR-193a-3p (SEQ ID NO: 25, y-axis), hsa-miR-7 (SEQ ID NO: 65, x-axis) and hsa-miR-375 (SEQ ID NO: 56, z-axis).

TABLE-US-00011 TABLE 11 miR expression (in fluorescence units) distinguishing between the group consisting of gastrointestinal (GI) epithelial tumors and the group consisting of non-GI epithelial tumors SEQ fold- ID median values change p-value NO. miR name 2.6e+003-7.1e+001 36.09 (+) 2.5e-127 27 hsa-miR-194 3.9e+003-1.2e+002 33.26 (+) 1.6e-117 177 hsa-miR-192 2.6e+003-6.7e+002 3.88 (+) 3.3e-021 4 hsa-miR-10a 5.0e+001-2.1e+004 411.76 (-) 6.5e-045 32 hsa-miR-205 (+) the higher expression of this miR is in the group consisting of GI epithelial tumors (-) the higher expression of this miR is in the group consisting of non-GI epithelial tumors

[0332] hsa-miR-194 (SEQ ID NO: 27) and hsa-miR-21* (SEQ ID NO: 35) are used at node 9 of the binary-tree-classifier detailed in the invention to distinguish between GI epithelial tumors and non-GI epithelial tumors.

[0333] FIG. 9 demonstrates that tumors originating in gastro-intestinal (GI) (marked by diamonds) are easily distinguished from tumors of non GI origins (marked by squares) using the expression levels of hsa-miR-21* (SEQ ID NO: 35, y-axis) and hsa-miR-194 (SEQ ID NO: 27, x-axis).

TABLE-US-00012 TABLE 12 miR expression (in fluorescence units) distinguishing between prostate tumors and all other non-GI epithelial tumors SEQ fold- ID median values change p-value NO. miR name 5.1e+003-5.2e+001 96.76 (+) 3.7e-016 56 hsa-miR-375 1.0e+003-5.5e+001 18.27 (+) 4.0e-025 199 hsa-miR-363 6.8e+004-7.2e+003 9.41 (+) 1.0e-025 14 hsa-miR-143 1.2e+005-1.4e+004 8.14 (+) 7.8e-022 15 hsa-miR-145 2.8e+003-3.5e+002 7.89 (+) 1.5e-012 165 hsa-miR-143* 2.1e+004-4.4e+003 4.76 (+) 2.2e-011 231 hsa-miR-99a 4.6e+002-2.1e+003 4.58 (-) 8.0e-007 36 hsa-miR-210 2.7e+002-1.1e+003 3.84 (-) 7.8e-017 154 hsa-miR-181b 1.2e+003-4.3e+003 3.76 (-) 1.2e-014 21 hsa-miR-181a 5.5e+002-2.0e+003 3.63 (-) 2.3e-002 49 hsa-miR-31 (+) the higher expression of this miR is in prostate tumors (-) the higher expression of this miR is in the group consisting of all other non-GI epithelial tumors

[0334] hsa-miR-143 (SEQ ID NO: 14) and hsa-miR-181a (SEQ ID NO: 21) are used at node 10 of the binary-tree-classifier detailed in the invention to distinguish between prostate tumors and all other non-GI epithelial tumors.

[0335] FIG. 10 demonstrates that tumors originating in prostate adenocarcinoma (marked by diamonds) are easily distinguished from tumors of non prostate origins (marked by squares) using the expression levels of hsa-miR-181a (SEQ ID NO: 21, y-axis) and hsa-miR-143 (SEQ ID NO: 14, x-axis).

TABLE-US-00013 TABLE 13 miR expression (in fluorescence units) distinguishing between seminomatous and non- seminomatous testicular tumors SEQ fold- ID median values change p-value NO. miR name 4.3e+003-7.6e+002 5.63 (+) 6.6e-004 152 hsa-miR-182 1.0e+002-2.1e+003 20.46 (-) 6.2e-005 216 hsa-miR-518e 7.8e+001-1.2e+003 15.29 (-) 4.5e-005 212 hsa-miR-516b 6.8e+001-8.2e+002 11.94 (-) 2.2e-005 224 hsa-miR-527 2.1e+002-2.2e+003 10.40 (-) 1.9e-006 13 hsa-miR-141 5.3e+002-5.0e+003 9.48 (-) 5.0e-004 194 hsa-miR-302d 1.4e+002-1.3e+003 8.97 (-) 4.1e-006 192 hsa-miR-302a 2.7e+002-2.3e+003 8.78 (-) 2.9e-003 221 hsa-miR-520c-3p 1.3e+002-1.2e+003 8.65 (-) 8.3e-004 217 hsa-miR-518f* 3.4e+003-2.9e+004 5.98 (-) 2.6e-007 205 hsa-miR-451 2.8e+002-1.7e+003 5.98 (-) 1.1e-002 219 hsa-miR-519d 2.0e+002-1.2e+003 5.90 (-) 6.8e-005 32 hsa-miR-205 2.0e+002-1.1e+003 5.59 (-) 5.8e-006 193 hsa-miR-302a* 1.9e+002-1.0e+003 5.27 (-) 6.7e-003 223 hsa-miR-524-5p 1.5e+002-8.0e+002 5.22 (-) 5.4e-003 220 hsa-miR-520a-5p 2.2e+002-1.1e+003 5.21 (-) 4.1e-003 210 hsa-miR-512-5p 3.2e+002-1.4e+003 4.57 (-) 9.2e-003 209 hsa-miR-498 7.2e+002-3.2e+003 4.51 (-) 3.1e-002 213 hsa-miR-517a 6.4e+002-2.9e+003 4.47 (-) 2.9e-002 163 hsa-miR-1323 9.5e+002-4.1e+003 4.29 (-) 1.3e-004 30 hsa-miR-200c (+) the higher expression of this miR is in seminoma tumors (-) the higher expression of this miR is in non-seminoma tumors

[0336] A combination of the expression level of any of the miRs detailed in table 13 with the expression level of hsa-miR-200b (SEQ ID NO: 29), hsa-miR-200a (SEQ ID NO: 28), hsa-miR-516a-5p (SEQ ID NO: 211), hsa-miR-767-5p (SEQ ID NO: 227), hsa-miR-518a-3p (SEQ ID NO: 215), hsa-miR-520d-5p (SEQ ID NO: 222), hsa-miR-519a (SEQ ID NO: 218) and hsa-miR-517c (SEQ ID NO: 214) also provides for classification of seminoma and non-seminoma testis-tumors.

[0337] hsa-miR-516a-5p (SEQ ID NO: 211) and hsa-miR-200b (SEQ ID NO: 29) are used at node 12 of the binary-tree-classifier detailed in the invention to distinguish between seminoma and non-seminoma testis-tumors.

[0338] FIG. 11 demonstrates that tumors originating in seminomatous testicular germ cell (marked by diamonds) are easily distinguished from tumors of non seminomatous origins (marked by squares) using the expression levels of hsa-miR-516a-5p (SEQ ID NO: 211, y-axis) and hsa-miR-200b (SEQ ID NO: 29, x-axis).

TABLE-US-00014 TABLE 14 miR expression (in fluorescence units) distinguishing between the group consisting of squamous cell carcinoma (SCC), transitional cell carcinoma (TCC), thymoma and the group consisting of non gastrointestinal (GI) adenocarcinoma tumors SEQ ID fold- miR name NO. p-value change median values hsa-miR-205 32 1.6e-059 321.76 (+) 4.6e+004-1.4e+002 hsa-miR-210 36 8.6e-015 5.96 (+) 2.9e+003-4.9e+002 hsa-miR-193b 178 2.6e-016 3.82 (+) 2.5e+003-6.6e+002 MID-16869 243 1.8e-008 3.67 (+) 2.5e+003-6.8e+002 MID-16489 242 2.2e-011 3.53 (+) 3.4e+003-9.7e+002 hsa-miR-31 49 8.2e-004 2.82 (+) 3.7e+003-1.3e+003 MID-15965 240 1.7e-010 2.78 (+) 6.4e+003-2.3e+003 hsa-miR-378 57 2.7e-017 2.71 (+) 1.4e+003-5.2e+002 hsa-miR-138 11 2.9e-023 8.05 (-) 2.8e+002-2.2e+003 hsa-miR-30a 46 1.5e-018 3.70 (-) 7.3e+002-2.7e+003 hsa-miR-146b-5p 17 4.0e-013 2.60 (-) 8.6e+002-2.2e+003 hsa-miR-30d 47 1.5e-021 2.44 (-) 1.8e+003-4.3e+003 hsa-miR-345 51 2.6e-019 2.38 (-) 4.5e+002-1.1e+003 hsa-miR-125a-5p 7 3.8e-014 2.30 (-) 4.2e+003-9.6e+003 hsa-miR-125b 8 3.0e-009 2.26 (-) 1.9e+004-4.3e+004 hsa-miR-181b 154 3.2e-008 2.24 (-) 9.4e+002-2.1e+003 hsa-miR-29b 190 3.3e-010 2.13 (-) 7.4e+002-1.6e+003 hsa-let-7i 157 4.6e-014 2.04 (-) 6.6e+003-1.3e+004 hsa-miR-30c 196 7.9e-010 2.04 (-) 2.4e+003-4.8e+003 (+) the higher expression of this miR is in SCC, TCC and thymoma (-) the higher expression of this miR is in non GI adenocarcinoma

[0339] Node 13 of the binary-tree-classifier separates tissues with high expression of miR-205 (SCC marker) such as SCC, TCC and thymomas from adenocarcinomas.

[0340] Breast adenocarcinoma and ovarian carcinoma are excluded from this separation due to a wide range of expression of miR-205.

[0341] A combination of the expression level of any of the miRs detailed in table 14 with the expression level of hsa-miR-331-3p (SEQ ID NO: 197) also provides for this classification.

[0342] hsa-miR-205 (SEQ ID NO: 32), hsa-miR-345 (SEQ ID NO: 51) and hsa-miR-125a-5p (SEQ ID NO: 7) are used at node 13 of the binary-tree-classifier detailed in the invention.

TABLE-US-00015 TABLE 15 miR expression (in fluorescence units) distinguishing between the group consisting of breast adenocarcinoma and the group consisting of SCC, TCC, thymomas and ovarian carcinoma SEQ ID fold- miR name NO. p-value change median values hsa-miR-375 56 4.1e-029 25.95 (+) 1.3e+003-5.0e+001 hsa-miR-30a 46 1.6e-014 3.25 (+) 2.7e+003-8.3e+002 hsa-miR-193a-3p 25 9.5e-022 3.09 (+) 4.4e+003-1.4e+003 hsa-miR-182 152 2.1e-009 2.94 (+) 1.2e+003-4.1e+002 hsa-miR-342-3p 50 7.3e-014 2.48 (+) 5.5e+003-2.2e+003 hsa-miR-29c* 45 6.3e-008 2.48 (+) 6.6e+002-2.7e+002 hsa-miR-29c 191 5.8e-007 2.26 (+) 5.0e+002-2.2e+002 hsa-miR-199a-3p 181 1.3e-004 2.19 (+) 7.3e+003-3.3e+003 hsa-miR-195 179 2.0e-006 2.05 (+) 2.2e+003-1.1e+003 hsa-miR-31 49 9.6e-014 13.81 (-) 2.2e+002-3.1e+003 hsa-miR-205 32 9.3e-008 6.32 (-) 6.3e+003-4.0e+004 hsa-miR-224 42 5.3e-010 5.72 (-) 8.9e+001-5.1e+002 hsa-miR-203 184 6.7e-007 4.05 (-) 1.5e+002-6.3e+002 hsa-miR-222 40 5.8e-018 2.64 (-) 5.7e+003-1.5e+004 hsa-miR-221 147 1.0e-018 2.41 (-) 3.8e+003-9.2e+003 MID-00689 236 4.9e-010 2.39 (-) 4.8e+002-1.1e+003 hsa-miR-378 57 1.2e-010 2.37 (-) 6.0e+002-1.4e+003 hsa-miR-422a 203 2.2e-008 2.24 (-) 2.3e+002-5.2e+002 hsa-miR-210 36 1.5e-007 2.22 (-) 1.2e+003-2.6e+003 (+) the higher expression of this miR is in breast adenocarcinoma (-) the higher expression of this miR is in SCC, TCC, thymomas and ovarian carcinoma

[0343] hsa-miR-193a-3p (SEQ ID NO: 25), hsa-miR-375 (SEQ ID NO: 56) and hsa-miR-342-3p (SEQ ID NO: 50) are used at node 14 of the binary-tree-classifier detailed in the invention. According to another embodiment, hsa-miR-193a-3p (SEQ ID NO: 25), hsa-miR-375 (SEQ ID NO: 56) and hsa-miR-224 (SEQ ID NO: 42) may be used at node 14 of the binary-tree-classifier detailed in the invention.

TABLE-US-00016 TABLE 16 miR expression (in fluorescence units) distinguishing between the group consisting of ovarian carcinoma and the group consisting of SCC, TCC and thymomas SEQ ID fold- miR name NO. p-value change median values hsa-miR-10a 4 1.2e-012 5.57 (+) 3.1e+003-5.5e+002 hsa-miR-130a 10 1.7e-014 3.41 (+) 5.1e+003-1.5e+003 hsa-miR-30a* 195 1.3e-014 3.39 (+) 2.5e+002-7.5e+001 hsa-miR-10b 5 5.5e-009 2.68 (+) 2.4e+003-8.8e+002 hsa-miR-625 226 2.5e-012 2.48 (+) 2.9e+002-1.2e+002 hsa-let-7e 2 7.7e-012 2.28 (+) 8.8e+003-3.9e+003 hsa-miR-30a 46 3.6e-007 2.20 (+) 1.6e+003-7.3e+002 hsa-miR-205 32 1.0e-033 37.52 (-) 1.2e+003-4.6e+004 hsa-miR-205* 185 4.2e-018 5.42 (-) 5.0e+001-2.7e+002 hsa-miR-138 11 1.1e-009 4.63 (-) 6.0e+001 2.8e+002 hsa-miR-150 168 5.2e-010 4.18 (-) 5.7e+002-2.4e+003 hsa-miR-203 184 2.5e-003 2.74 (-) 2.9e+002-8.0e+002 hsa-miR-146a 16 2.1e-006 2.62 (-) 2.9e+002-7.6e+002 MID-16489 242 1.9e-007 2.49 (-) 1.4e+003-3.4e+003 hsa-miR-140-3p 12 2.3e-015 2.42 (-) 9.5e+002-2.3e+003 MID-15684 237 5.2e-006 2.37 (-) 7.6e+002-1.8e+003 MID-16869 243 9.8e-005 2.23 (-) 1.1e+003-2.5e+003 MID-20703 250 5.9e-004 2.18 (-) 2.1e+003-4.6e+003 hsa-miR-22 39 6.5e-012 2.13 (-) 2.7e+003-5.8e+003 MID-23256 253 1.0e-003 2.13 (-) 4.3e+002-9.2e+002 hsa-miR-31 49 1.4e-002 2.12 (-) 1.7e+003-3.7e+003 MID-18422 246 1.3e-003 2.06 (-) 1.7e+003-3.6e+003 hsa-miR-149* 167 1.4e-007 2.04 (-) 2.1e+003-4.4e+003 (+) the higher expression of this miR is in ovarian carcinoma (-) the higher expression of this miR is in SCC, TCC and thymomas

[0344] hsa-miR-205 (SEQ ID NO: 32), hsa-miR-10a (SEQ ID NO: 4) and hsa-miR-22 (SEQ ID NO: 39) are used at node 15 of the binary-tree-classifier detailed in the invention.

TABLE-US-00017 TABLE 17 miR expression (in fluorescence units) distinguishing between the group consisting of thyroid carcinoma (follicular and papillary) and the group consisting of breast adenocarcinoma, lung large cell carcinoma, lung adenocarcinoma and ovarian carcinoma SEQ ID fold- miR name NO. p-value change median values hsa-miR-138 11 7.4e-033 33.86 (+) 4.1e+003-1.2e+002 hsa-miR-221 147 1.4e-009 5.03 (+) 3.4e+004-6.7e+003 hsa-miR-146b-5p 17 1.0e-006 4.74 (+) 4.9e+003-1.0e+003 hsa-let-7i 157 2.8e-027 3.71 (+) 2.2e+004-5.8e+003 hsa-miR-222 40 7.9e-009 3.63 (+) 3.9e+004-1.1e+004 hsa-miR-125b 8 1.5e-014 2.78 (+) 5.4e+004-1.9e+004 hsa-miR-31 49 5.0e-003 2.78 (+) 1.3e+003-4.9e+002 hsa-miR-126 9 1.3e-008 2.48 (+) 6.9e+003-2.8e+003 hsa-miR-29c 191 4.8e-007 2.36 (+) 8.1e+002-3.4e+002 hsa-miR-451 205 3.8e-003 2.33 (+) 1.3e+004-5.7e+003 hsa-miR-486-5p 207 1.3e-003 2.16 (+) 4.8e+002-2.2e+002 hsa-miR-30a* 195 8.0e-006 2.12 (+) 4.9e+002-2.3e+002 hsa-miR-345 51 2.7e-011 2.11 (+) 1.4e+003-6.6e+002 hsa-miR-30a 46 1.5e-006 2.10 (+) .5e+003-1.7e+003 hsa-miR-29c* 45 1.5e-006 1.97 (+) 6.4e+002-3.2e+002 hsa-miR-34a 52 3.4e-007 1.88 (+) 7.4e+003-4.0e+003 hsa-miR-1977 234 5.9e-008 1.88 (+) 6.1e+003-3.3e+003 hsa-miR-99a 231 1.5e-005 1.85 (+) 7.5e+003-4.1e+003 hsa-miR-181a 21 2.7e-004 1.85 (+) 7.6e+003-4.1e+003 hsa-miR-152 169 5.1e-008 1.82 (+) 1.0e+003-5.5e+002 hsa-miR-29a 43 3.7e-008 1.79 (+) 1.16+004-6.0e+003 hsa-miR-100 3 2.9e-005 1.77 (+) 5.4e+003-3.0e+003 hsa-miR-30c 196 7.3e-007 1.73 (+) 5.9e+003-3.4e+003 hsa-miR-181b 154 4.6e-004 1.69 (+) 2.1e+003-1.2e+003 MID-00405 390 6.4e-003 1.66 (+) 4.4e+002-2.6e+002 hsa-miR-15a 171 5.3e-006 1.65 (+) 5.5e+002-3.3e+002 MID-23794 255 6.0e-003 1.60 (+) 1.3e+003-8.1e+002 hsa-miR-331-3p 197 8.3e-007 1.57 (+) 2.3e+003-1.4e+003 hsa-miR-29b 190 1.4e-005 1.57 (+) 1.8e+003-1.1e+003 hsa-miR-27b 189 1.4e-003 1.56 (+) 3.6e+003-2.3e+003 hsa-miR-22 39 3.6e-005 1.52 (+) 6.6e+003-4.3e+003 hsa-miR-125a-5p 7 2.2e-006 1.52 (+) 9.9e+003-6.5e+003 hsa-miR-30e 48 6.2e-006 1.51 (+) 7.8e+002-5.2e+002 hsa-miR-30d 47 1.2e-002 1.51 (+) 4.3e+003-2.8e+003 hsa-miR-205 32 2.2e-005 22.05 (-) 1.0e+002-2.2e+003 hsa-miR-210 36 4.8e-020 8.83 (-) 2.1e+002-1.8e+003 hsa-miR-10a 4 8.7e-011 4.34 (-) 3.2e+002-1.4e+003 hsa-miR-193b 178 4.6e-014 3.59 (-) 4.9e+002-1.8e+003 hsa-miR-214 37 8.3e-006 2.74 (-) 1.0e+003-2.8e+003 hsa-miR-199a-3p 181 4.4e-005 2.67 (-) 2.0e+003-5.5e+003 hsa-miR-193a-3p 25 3.1e-011 2.65 (-) 1.1e+003-2.8e+003 hsa-miR-199b-5p 183 1.3e-004 2.63 (-) 2.7e+002-7.2e+002 hsa-miR-199a-5p 182 1.6e-005 2.57 (-) 3.1e+003-8.1e+003 hsa-miR-21* 35 4.7e-006 2.41 (-) 5.1e+002-1.2e+003 MID-15965 240 9.3e-003 2.39 (-) 2.0e+003-4.8e+003 hsa-miR-378 57 3.0e-006 2.35 (-) 3.4e+002-8.0e+002 MID-16489 242 1.2e-003 2.25 (-) 8.2e+002-1.9e+003 hsa-miR-425 204 1.4e-011 2.18 (-) 6.0e+002-1.3e+003 MID-00689 236 5.3e-006 2.06 (-) 3.0e+002-6.1e+002 hsa-miR-18a 176 2.0e-006 1.96 (-) 2.3e+002-4.5e+002 hsa-miR-106a 158 3.5e-006 1.85 (-) 2.3e+003-4.3e+003 hsa-miR-93 148 1.9e-010 1.79 (-) 2.4e+003-4.4e+003 hsa-miR-455-3p 206 1.9e-007 1.79 (-) 2.9e+002-5.2e+002 hsa-miR-342-3p 50 7.8e-005 1.78 (-) 1.3e+003-2.3e+003 hsa-miR-17 20 6.0e-006 1.75 (-) 1.4e+003-2.5e+003 hsa-miR-21 34 5.1e-006 1.75 (-) 2.8e+004-4.8e+004 hsa-miR-20a 186 1.1e-004 1.74 (-) 1.4e+003-2.4e+003 MID-15907 239 8.4e-004 1.72 (-) 2.4e+002-4.1e+002 MID-21271 251 9.9e-003 1.62 (-) 3.0e+002-4.8e+002 MID-17144 244 4.3e-002 1.60 (-) 2.2e+003-3.6e+003 hsa-miR-191 24 4.7e-006 1.59 (-) 3.8e+003-6.1e+003 hsa-miR-25 188 2.0e-005 1.59 (-) 1.0e+003-1.6e+003 hsa-miR-15b 172 1.9e-002 1.57 (-) 2.1e+003-3.2e+003 MID-15867 238 9.5e-003 1.56 (-) 2.6e+003-4.0e+003 (+) the higher expression of this miR is in thyroid carcinoma (-) the higher expression of this miR is in breast adenocarcinoma, lung large cell carcinoma, lung adenocarcinoma and ovarian carcinoma

[0345] FIG. 12 demonstrates binary decisions at node #16 of the decision-tree. Tumors originating in thyroid carcinoma (diamonds) are easily distinguished from tumors of adenocarcinoma of the lung, breast and ovarian origin (squares) using the expression levels of hsa-miR-93 (SEQ ID NO: 148, y-axis), hsa-miR-138 (SEQ ID NO: 11, x-axis) and hsa-miR-10a (SEQ ID NO: 4, z-axis).

TABLE-US-00018 TABLE 18 miR expression (in fluorescence units) distinguishing between follicular thyroid carcinoma and papillary thyroid carcinoma SEQ ID fold- miR name NO. p-value change median values MID-20524 249 4.5e-011 9.34 (+) 6.6e+003-7.1e+002 hsa-miR-1973 180 1.9e-008 7.80 (+) 1.7e+003-2.2e+002 hsa-miR-7 65 8.3e-005 7.58 (+) 4.5e+002-5.9e+001 hsa-miR-1978 235 4.8e-007 6.52 (+) 2.5e+003-3.8e+002 MID-16318 241 1.5e-008 6.14 (+) 2.2e+003-3.6e+002 MID-19533 248 3.0e-004 6.00 (+) 4.2e+002-7.1e+001 MID-23291 254 1.6e-008 5.76 (+) 9.6e+002-1.7e+002 MID-19340 247 3.2e-005 5.33 (+) 9.9e+002-1.9e+002 hsa-miR-1248 160 6.8e-009 5.17 (+) 6.4e+002-1.2e+002 MID-16869 243 1.1e-006 4.97 (+) 1.5e+003-3.0e+002 MID-18336 245 1.4e-010 4.48 (+) 2.7e+003-6.1e+002 MID-22664 252 7.0e-004 4.00 (+) 5.0e+002-1.2e+002 hsa-miR-146b-5p 17 6.7e-011 62.88 (-) 4.0e+002-2.5e+004 hsa-miR-31 49 2.5e-008 18.72 (-) 4.4e+002-8.2e+003 hsa-miR-146b-3p 166 5.0e-012 18.69 (-) 5.0e+001-9.3e+002 hsa-miR-551b 225 4.8e-006 10.86 (-) 7.6e+001-8.3e+002 hsa-miR-150 168 3.2e-007 10.71 (-) 3.1e+002-3.3e+003 hsa-miR-21 34 3.4e-007 4.40 (-) 1.1e+004-4.7e+004 (+) the higher expression of this miR is in follicular thyroid carcinoma (-) the higher expression of this miR is in papillary thyroid carcinoma

[0346] FIG. 13 demonstrates binary decisions at node #17 of the decision-tree. Tumors originating in follicular thyroid carcinoma (marked by diamonds) are easily distinguished from tumors of papillary thyroid carcinoma origins (marked by squares) using the expression levels of hsa-miR-21 (SEQ ID NO: 34, y-axis) and hsa-miR-146b-5p (SEQ ID NO: 17, x-axis).

TABLE-US-00019 TABLE 19 miR expression (in fluorescence units) distinguishing between the group consisting of breast adenocarcinoma and the group consisting of lung adenocarcinoma and ovarian carcinoma SEQ ID fold- miR name NO. p-value change median values hsa-miR-205 32 8.8e-005 10.55 (+) 6.3e+003-6.0e+002 hsa-miR-375 56 7.9e-006 8.43 (+) 1.3e+003-1.5e+002 hsa-miR-342-3p 50 7.7e-012 3.17 (+) 5.5e+003-1.7e+003 hsa-miR-29c* 45 2.2e-008 2.52 (+) 6.6e+002-2.6e+002 hsa-miR-193a-3p 25 2.2e-012 2.23 (+) 4.4e+003-2.0e+003 MID-23256 253 7.9e-005 2.20 (+) 9.5e+002-4.3e+002 hsa-miR-182 152 4.9e-005 2.15 (+) 1.2e+003-5.6e+002 hsa-miR-126 9 5.3e-005 1.94 (+) 3.7e+003-1.9e+003 hsa-miR-30a 46 2.9e-004 1.90 (+) 2.7e+003-1.4e+003 hsa-miR-29c 191 1.1e-004 1.78 (+) 5.0e+002-2.8e+002 hsa-miR-193b 178 1.5e-006 1.72 (+) 2.4e+003-1.4e+003 hsa-miR-31 49 7.7e-006 5.79 (-) 2.2e+002-1.3e+003 hsa-miR-222 40 3.0e-010 2.69 (-) 5.7e+003-1.5e+004 hsa-miR-130a 10 1.6e-006 2.50 (-) 1.4e+003-3.4e+003 hsa-miR-221 147 1.2e-009 2.41 (-) 3.8e+003-9.3e+003 hsa-miR-10a 4 2.2e-002 2.33 (-) 9.3e+002-2.2e+003 hsa-miR-210 36 1.5e-005 2.09 (-) 1.2e+003-2.5e+003 hsa-miR-886-3p 228 9.9e-005 2.01 (-) 1.3e+003-2.6e+003 MID-00689 236 4.6e-004 1.95 (-) 4.8e+002-9.4e+002 hsa-miR-886-5p 230 6.3e-004 1.92 (-) 5.3e+002-1.0e+003 hsa-miR-27b 189 7.1e-005 1.86 (-) 1.8e+003-3.3e+003 MID-15965 240 6.1e-003 1.79 (-) 3.6e+003-6.4e+003 hsa-miR-92a 67 3.5e-005 1.75 (-) 2.7e+003-4.7e+003 hsa-miR-378 202 3.0e-004 1.73 (-) 6.0e+002-1.0e+003 hsa-miR-146b-5p 17 2.9e-004 1.71 (-) 8.0e+002-1.4e+003 (+) the higher expression of this miR is in breast adenocarcinoma (-) the higher expression of this miR is in lung adenocarcinoma and ovarian carcinoma

[0347] FIG. 14 demonstrates binary decisions at node #18 of the decision-tree. Tumors originating in breast (diamonds) are easily distinguished from tumors of lung and ovarian origin (squares) using the expression levels of hsa-miR-92a (SEQ ID NO: 67, y-axis), hsa-miR-193a-3p (SEQ ID NO: 25, x-axis) and hsa-miR-31 (SEQ ID NO: 49, z-axis).

TABLE-US-00020 TABLE 20 miR expression (in fluorescence units) distinguishing between lung adenocarcinoma and ovarian carcinoma SEQ fold- ID median values change p-value NO. miR name 1.4e+003-5.2e+001 27.96 (+) 3.5e-008 56 hsa-miR-375 5.5e+002-6.0e+001 9.19 (+) 8.1e-009 11 hsa-miR-138 3.2e+003-5.7e+002 5.65 (+) 5.6e-004 168 hsa-miR-150 9.7e+002-2.9e+002 3.35 (+) 2.2e-004 16 hsa-miR-146a 2.3e+003-7.6e+002 2.96 (+) 3.2e-003 237 MID-15684 8.4e+003-2.9e+003 2.88 (+) 1.7e-010 21 hsa-miR-181a 7.0e+003-2.6e+003 2.69 (+) 9.2e-008 52 hsa-miR-34a 2.4e+003-9.5e+002 2.58 (+) 3.2e-007 12 hsa-miR-140-3p 2.1e+003-8.8e+002 2.39 (+) 2.1e-007 154 hsa-miR-181b 1.4e+005-6.3e+004 2.28 (+) 1.9e-003 279 hsa-miR-1826 3.2e+003-1.4e+003 2.25 (+) 6.1e-003 9 hsa-miR-126 6.1e+003-2.7e+003 2.24 (+) 2.3e-006 39 hsa-miR-22 4.2e+003-2.2e+003 1.93 (+) 8.9e-005 47 hsa-miR-30d 1.9e+003-1.0e+003 1.90 (+) 3.5e-006 23 hsa-miR-185 2.8e+003-1.5e+003 1.88 (+) 1.4e-003 50 hsa-miR-342-3p 3.6e+003-2.1e+003 1.69 (+) 1.8e-002 167 hsa-miR-149* 9.7e+002-5.9e+002 1.66 (+) 8.7e-005 383 MID-22912 6.8e+004-4.2e+004 1.64 (+) 5.4e-004 34 hsa-miR-21 1.8e+003-1.2e+003 1.55 (+) 5.1e-004 35 hsa-miR-21* 1.4e+003-8.9e+002 1.54 (+) 1.7e-004 388 hsa-miR-423-5p 4.4e+002-2.4e+003 5.38 (-) 1.0e-007 5 hsa-miR-10b 1.9e+002-7.2e+002 3.77 (-) 3.3e-006 359 hsa-miR-708 6.4e+002-2.1e+003 3.27 (-) 5.8e-003 245 MID-18336 1.8e+002-5.4e+002 2.96 (-) 6.6e-003 254 MID-23291 1.7e+003-5.1e+003 2.95 (-) 3.4e-005 10 hsa-miR-130a 2.8e+003-8.1e+003 2.86 (-) 3.7e-003 240 MID-15965 5.1e+002-1.3e+003 2.65 (-) 3.7e-004 236 MID-00689 6.1e+002-1.6e+003 2.63 (-) 1.8e-004 202 hsa-miR-378 1.3e+003-3.1e+003 2.39 (-) 1.2e-002 4 hsa-miR-41a 1.0e+003-2.3e+003 2.30 (-) 1.8e-006 25 hsa-miR-193a-3p 2.6e+002-5.6e+002 2.15 (-) 4.1e-004 203 hsa-miR-422a 3.0e+003-6.1e+003 2.04 (-) 1.8e-002 231 hsa-miR-99a 2.0e+003-3.9e+003 2.01 (-) 3.5e-005 20 hsa-miR-17 3.3e+003-6.4e+003 1.96 (-) 3.5e-005 158 hsa-miR-106a 1.8e+003-3.5e+003 1.88 (-) 3.1e-005 186 hsa-miR-20a 3.2e+003-5.9e+003 1.85 (-) 1.2e-004 258 hsa-let-7f 2.4e+003-4.4e+003 1.85 (-) 2.0e-003 244 MID-17144 2.0e+003-3.5e+003 1.72 (-) 8.7e-004 172 hsa-miR-15b 5.2e+003-8.8e+003 1.69 (-) 5.3e-004 2 hsa-let-7e 4.1e+002-6.8e+002 1.66 (-) 1.2e-002 235 hsa-miR-1978 3.3e+004-5.4e+004 1.66 (-) 3.3e-005 256 hsa-let-7a 2.8e+003-4.7e+003 1.65 (-) 3.5e-003 28 hsa-miR-200a 5.2e+002-8.5e+002 1.62 (-) 1.6e-004 277 hsa-miR-17* 3.7e+002-5.8e+002 1.57 (-) 1.9e-003 296 hsa-miR-26b 1.0e+005-1.6e+005 1.56 (-) 3.7e-002 374 MID-16748 6.0e+003-9.1e+003 1.53 (-) 2.1e-005 153 hsa-let-7d 3.8e+003-5.7e+003 1.50 (-) 4.1e-003 181 hsa-miR-199a-3p (+) the higher expression of this miR is in lung adenocarcinoma (-) the higher expression of this miR is in ovarian carcinoma

[0348] FIG. 15 demonstrates binary decisions at node #19 of the decision-tree. Tumors originating in lung adenocarcinoma (diamonds) are easily distinguished from tumors of ovarian carcinoma origin (squares) using the expression levels of hsa-miR-21 (SEQ ID NO: 34, y-axis), hsa-miR-378 (SEQ ID NO: 202, x-axis) and hsa-miR-138 (SEQ ID NO: 11, z-axis).

TABLE-US-00021 TABLE 21 miR expression (in fluorescence units) distinguishing between the group consisting of thymic carcinoma and the group consisting of TCC and SCC SEQ fold- ID median values change p-value NO. miR name 5.3e+002-5.9e+001 9.00 (+) 5.7e-026 161 hsa-miR-128 7.4e+002-9.2e+001 8.04 (+) 2.2e-007 164 hsa-miR-142-5p 6.8e+002-8.8e+001 7.82 (+) 2.6e-021 22 hsa-miR-181a* 7.1e+002-1.2e+002 6.09 (+) 1.2e-006 53 hsa-miR-34c-5p 9.1e+002-1.8e+002 5.06 (+) 2.2e-008 285 hsa-miR-20b 1.3e+004-2.8e+003 4.59 (+) 7.5e-014 3 hsa-miR-100 1.6e+003-3.6e+002 4.39 (+) 7.1e-007 152 hsa-miR-182 8.7e+002-2.0e+002 4.37 (+) 6.4e-010 191 hsa-miR-29c 3.7e+003-9.1e+002 4.09 (+) 2.6e-014 154 hsa-miR-181b 1.5e+004-3.8e+003 3.82 (+) 4.8e-009 21 hsa-miR-181a 2.1e+003-6.4e+002 3.25 (+) 6.4e-006 206 hsa-miR-455-3p 8.7e+002-2.7e+002 3.23 (+) 1.5e-010 174 hsa-miR-181d 9.4e+002-2.9e+002 3.23 (+) 1.9e-004 19 hsa-miR-149 7.3e+002-2.6e+002 2.80 (+) 2.5e-008 45 hsa-miR-29c* 8.7e+002-3.2e+002 2.69 (+) 2.2e-007 171 hsa-miR-15a 2.7e+003-1.0e+003 2.66 (+) 5.1e-005 179 hsa-miR-195 4.4e+004-1.8e+004 2.46 (+) 8.2e-008 8 hsa-miR-125b 7.3e+002-3.2e+002 2.26 (+) 2.8e-004 296 hsa-miR-26b 2.7e+003-1.2e+003 2.22 (+) 2.4e-003 284 hsa-miR-19b 5.7e+002-2.6e+002 2.17 (+) 7.8e-005 18 hsa-miR-148a 9.1e+002-4.4e+002 2.06 (+) 1.4e-006 51 hsa-miR-345 7.5e+003-3.8e+003 2.00 (+) 1.6e-002 258 hsa-let-7f 1.8e+002-4.4e+003 24.66 (-) 3.9e-008 49 hsa-miR-31 5.8e+001-1.0e+003 17.65 (-) 1.0e-007 184 hsa-miR-203 2.2e+002-1.6e+003 7.43 (-) 4.5e-018 35 hsa-miR-21* 1.1e+004-5.5e+004 4.97 (-) 1.5e-032 34 hsa-miR-21 6.2e+002-2.5e+003 4.06 (-) 2.0e-008 37 hsa-miR-214 1.6e+002-5.8e+002 3.69 (-) 6.3e-005 42 hsa-miR-224 6.9e+002-2.5e+003 3.58 (-) 2.3e-009 228 hsa-miR-886-3p 4.8e+003-1.7e+004 3.47 (-) 3.9e-009 15 hsa-miR-145 2.7e+003-8.2e+003 3.08 (-) 2.7e-008 14 hsa-miR-143 1.3e+003-3.7e+003 2.93 (-) 6.7e-005 242 MID-16489 2.5e+002-7.4e+002 2.91 (-) 4.3e-006 230 hsa-miR-886-5p 3.5e+002-1.0e+003 2.90 (-) 5.6e-004 253 MID-23256 1.1e+003-3.0e+003 2.82 (-) 7.9e-006 36 hsa-miR-210 2.2e+003-5.8e+003 2.63 (-) 1.2e-007 182 hsa-miR-199a-5p 5.0e+003-1.2e+004 2.48 (-) 3.8e-006 293 hsa-miR-23b 7.5e+003-1.8e+004 2.44 (-) 1.1e-008 292 hsa-miR-23a 2.7e+002-6.6e+002 2.43 (-) 5.4e-003 4 hsa-miR-10a 9.0e+003-2.2e+004 2.43 (-) 5.5e-015 294 hsa-miR-24 3.8e+003-8.8e+003 2.35 (-) 3.0e-004 297 hsa-miR-27a 2.3e+002-5.2e+002 2.28 (-) 1.6e-002 354 hsa-miR-612 1.8e+003-3.9e+003 2.22 (-) 1.1e-006 377 MID-17866 1.6e+003-3.3e+003 2.11 (-) 2.4e-004 189 hsa-miR-27b 6.9e+003-1.4e+004 2.08 (-) 1.5e-004 30 hsa-miR-200c 3.8e+004-7.9e+004 2.05 (-) 1.8e-008 386 MID-23178 1.5e+003-3.1e+003 2.04 (-) 2.6e-002 249 MID-20524 4.6e+002-9.3e+002 2.03 (-) 5.6e-002 5 hsa-miR-10b 2.5e+002-5.0e+002 2.03 (-) 3.2e-005 274 hsa-miR-151-3p (+) the higher expression of this miR is in thymic carcinoma (-) the higher expression of this miR is in TCC and SCC

[0349] FIG. 16 demonstrates binary decisions at node #20 of the decision-tree. Tumors originating in thymic carcinoma (marked by diamonds) are easily distinguished from tumors of urothelial carcinoma, transitional cell carcinoma (TCC) carcinoma and squamous cell carcinoma (SCC) origins (marked by squares) using the expression levels of hsa-miR-21 (SEQ ID NO: 34, y-axis) and hsa-miR-100 (SEQ ID NO: 3, x-axis).

TABLE-US-00022 TABLE 22 miR expression (in fluorescence units) distinguishing between TCC and SCC (of anus, skin, lung, head&neck, esophagus or uterine cervix) SEQ fold- ID median values change p-value NO. miR name 2.5e+002-5.0e+001 5.05 (+) 7.4e-036 69 hsa-miR-934 9.1e+003-2.2e+003 4.14 (+) 1.2e-012 28 hsa-miR-200a 2.1e+002-5.3e+001 3.87 (+) 8.4e-007 280 hsa-miR-187 6.0e+003-1.9e+003 3.19 (+) 9.8e-008 13 hsa-miR-141 5.6e+002-1.8e+002 3.15 (+) 4.7e-013 191 hsa-miR-29c 9.4e+002-3.0e+002 3.13 (+) 8.1e-008 152 hsa-miR-182 1.8e+004-6.2e+003 2.99 (+) 3.9e-010 29 hsa-miR-200b 3.2e+002-1.1e+002 2.81 (+) 8.8e-009 175 hsa-miR-183 3.1e+004-1.2e+004 2.65 (+) 8.1e-005 30 hsa-miR-200c 2.1e+003-8.1e+002 2.63 (+) 6.2e-020 204 hsa-miR-425 1.2e+003-4.8e+002 2.41 (+) 6.3e-007 4 hsa-miR-10a 8.5e+003-3.7e+003 2.30 (+) 2.4e-024 24 hsa-miR-191 1.8e+003-8.4e+002 2.14 (+) 2.7e-006 5 hsa-miR-10b 3.5e+002-1.7e+002 2.09 (+) 2.0e-006 329 hsa-miR-425* 3.4e+002-1.6e+002 2.08 (+) 8.0e-005 273 hsa-miR-148b 3.1e+002-8.1e+002 2.60 (-) 1.6e-005 170 hsa-miR-155 5.0e+002-1.2e+003 2.49 (-) 1.0e-002 184 hsa-miR-203 1.5e+002-3.5e+002 2.39 (-) 3.0e-011 26 hsa-miR-193a-5p 1.9e+003-4.5e+003 2.35 (-) 1.3e-008 231 hsa-miR-99a 1.2e+002-2.7e+002 2.28 (-) 1.6e-003 368 MID-00672 1.4e+003-3.2e+003 2.25 (-) 1.3e-005 37 hsa-miR-214 3.9e+002-8.7e+002 2.23 (-) 1.5e-004 16 hsa-miR-146a 3.5e+002-7.6e+002 2.15 (-) 4.2e-005 169 hsa-miR-152 1.5e+002-3.3e+002 2.15 (-) 4.4e-004 155 hsa-miR-127-3p 8.4e+002-1.7e+003 2.08 (-) 1.6e-005 35 hsa-miR-21* 7.7e+003-1.6e+004 2.07 (-) 5.3e-012 40 hsa-miR-222 4.4e+002-9.0e+002 2.06 (-) 1.5e-005 17 hsa-miR-146b-5p (+) the higher expression of this miR is in TCC (-) the higher expression of this miR is in SCC

[0350] hsa-miR-934 (SEQ ID NO: 69), hsa-miR-191 (SEQ ID NO: 24) and hsa-miR-29c (SEQ ID NO: 191) are used at node #21 of the binary-tree-classifier detailed in the invention to distinguish between TCC and SCC.

TABLE-US-00023 TABLE 23 miR expression (in fluorescence units) distinguishing between SCC of the uterine cervix and other SCC tumors (anus, skin, lung, head& neck or esophagus) median values auROC fold-change p-value SEQ ID NO. miR name 2.4e+002-9.2e+001 0.65 2.57 (+) 2.0e-002 164 hsa-miR-142-5p 1.6e+003-7.6e+002 0.85 2.13 (+) 1.7e-005 5 hsa-miR-10b 8.9e+003-4.4e+003 0.74 2.01 (+) 2.1e-004 231 hsa-miR-99a 1.2e+003-9.8e+002 0.71 1.24 (+) 1.2e-002 54 hsa-miR-361-5p 3.4e+004-2.7e+004 0.71 1.24 (+) 3.9e-004 1 hsa-let-7c 1.3e+003-4.3e+003 0.81 3.39 (-) 9.9e-006 242 MID-16489 3.9e+002-1.2e+003 0.74 3.10 (-) 2.1e-003 372 MID-16469 1.1e+003-3.3e+003 0.84 3.09 (-) 1.3e-008 249 MID-20524 1.7e+003-5.2e+003 0.78 3.01 (-) 2.4e-005 167 hsa-miR-149* 2.7e+002-8.0e+002 0.79 2.97 (-) 1.4e-004 254 MID-23291 2.2e+002-6.2e+002 0.76 2.77 (-) 1.7e-004 354 hsa-miR-612 5.7e+002-1.5e+003 0.76 2.65 (-) 7.8e-006 381 MID-19962 2.3e+002-6.0e+002 0.79 2.63 (-) 2.1e-005 380 MID-19898 9.8e+002-2.4e+003 0.78 2.44 (-) 2.8e-005 245 MID-18336 1.2e+002-2.8e+002 0.73 2.34 (-) 5.3e-003 358 hsa-miR-665 6.1e+002-1.4e+003 0.70 2.31 (-) 6.1e-003 364 MID-00064 2.9e+003-6.7e+003 0.81 2.30 (-) 8.8e-008 240 MID-15965 1.2e+002-2.8e+002 0.66 2.26 (-) 1.5e-002 11 hsa-miR-138 1.0e+002-2.3e+002 0.77 2.24 (-) 8.9e-005 378 MID-18307 (+) the higher expression of this miR is in SCC of the uterine cervix (-) the higher expression of this miR is in other SCC tumors

[0351] FIG. 17 demonstrates binary decisions at node #22 of the decision-tree. Tumors originating in SCC of the uterine cervix (diamonds) are easily distinguished from tumors of other SCC origin (squares) using the expression levels of hsa-miR-361-5p (SEQ ID NO: 54, y-axis), hsa-let-7c (SEQ ID NO: 1, x-axis) and hsa-miR-10b (SEQ ID NO: 5, z-axis).

TABLE-US-00024 TABLE 24 miR expression (in fluorescence units) distinguishing between anus or skin SCC and upper SCC tumors (lung, head& neck or esophagus) median values auROC fold-change p-value SEQ ID NO. miR name 3.2e+002-5.0e+001 0.78 6.38 (+) 3.0e-006 305 hsa-miR-31* 4.3e+003-8.0e+002 0.80 5.39 (+) 1.8e-006 184 hsa-miR-203 8.6e+002-2.5e+002 0.78 3.49 (+) 1.8e-006 41 hsa-miR-223 1.7e+003-5.4e+002 0.80 3.12 (+) 3.5e-006 183 hsa-miR-199b-5p 9.4e+003-3.5e+003 0.70 2.73 (+) 2.4e-003 49 hsa-miR-31 8.7e+003-3.2e+003 0.86 2.71 (+) 3.6e-007 382 MID-22331 1.9e+003-7.1e+002 0.87 2.68 (+) 1.7e-008 235 hsa-miR-1978 2.4e+002-9.2e+001 0.83 2.55 (+) 9.6e-009 291 hsa-miR-222* 6.8e+003-2.9e+003 0.74 2.31 (+) 7.4e-004 181 hsa-miR-199a-3p 1.5e+003-6.7e+002 0.88 2.28 (+) 7.1e-007 5 hsa-miR-10b 5.3e+002-2.4e+002 0.75 2.21 (+) 1.4e-004 296 hsa-miR-26b 3.4e+002-1.6e+002 0.74 2.19 (+) 7.7e-005 289 hsa-miR-22* 1.3e+003-6.0e+002 0.71 2.13 (+) 1.2e-003 206 hsa-miR-455-3p 7.9e+003-3.8e+003 0.84 2.11 (+) 4.2e-006 338 hsa-miR-494 2.9e+002-1.4e+002 0.73 2.08 (+) 1.1e-004 334 hsa-miR-483-5p 2.8e+003-1.3e+003 0.82 2.07 (+) 4.5e-006 25 hsa-miR-193a-3p 1.1e+002-3.3e+002 0.77 3.03 (-) 2.3e-005 11 hsa-miR-138 1.3e+002-3.1e+002 0.65 2.29 (-) 1.5e-002 19 hsa-miR-149 9.7e+001-2.1e+002 0.75 2.16 (-) 4.0e-005 198 hsa-miR-342-5p 1.1e+003-1.8e+003 0.83 1.63 (-) 1.1e-006 23 hsa-miR-185 (+) the higher expression of this miR is in anus or skin SCC (-) the higher expression of this miR is in upper SCC tumors

[0352] hsa-miR-10b (SEQ ID NO: 5), hsa-miR-138 (SEQ ID NO: 11) and hsa-miR-185 (SEQ ID NO: 23) are used at node 23 of the binary-tree-classifier detailed in the invention to distinguish between anus or skin SCC and upper SCC tumors.

TABLE-US-00025 TABLE 25 miR expression (in fluorescence units) distinguishing between melanoma and lymphoma (B-cell or T-cell) tumors median values auROC fold-change p-value SEQ ID NO. miR name 1.7e+003-3.0e+002 0.89 5.81 (+) 2.8e-010 4 hsa-miR-10a 1.9e+003-6.0e+002 0.80 3.13 (+) 7.9e-005 11 hsa-miR-138 1.7e+003-5.7e+002 0.94 2.98 (+) 2.3e-011 46 hsa-miR-30a 2.5e+004-8.8e+003 0.87 2.83 (+) 1.1e-009 8 hsa-miR-125b 6.2e+002-2.3e+002 0.94 2.74 (+) 9.2e-011 274 hsa-miR-151-3p 9.2e+002-3.4e+002 0.87 2.70 (+) 1.9e-007 169 hsa-miR-152 1.6e+003-6.0e+002 0.77 2.60 (+) 2.0e-004 36 hsa-miR-210 4.8e+003-1.9e+003 0.90 2.56 (+) 2.1e-011 47 hsa-miR-30d 1.2e+003-5.5e+002 0.88 2.26 (+) 2.5e-008 363 hsa-miR-99b 2.4e+003-1.1e+003 0.85 2.24 (+) 1.4e-006 231 hsa-miR-99a 6.5e+003-3.0e+003 0.80 2.17 (+) 2.2e-005 303 hsa-miR-30b 6.4e+002-3.0e+002 0.86 2.14 (+) 2.9e-008 349 hsa-miR-532-5p 2.1e+003-1.0e+003 0.86 2.08 (+) 1.6e-006 10 hsa-miR-130a 5.4e+003-2.6e+003 0.81 2.06 (+) 8.3e-006 7 hsa-miR-125a-5p 3.6e+003-1.8e+003 0.82 2.05 (+) 2.5e-006 3 hsa-miR-100 7.9e+003-3.9e+003 0.69 2.04 (+) 1.5e-002 16 hsa-miR-146a 1.6e+002-2.2e+003 0.93 13.84 (-) 1.1e-014 164 hsa-miR-142-5p 7.2e+002-7.5e+003 0.93 10.40 (-) 1.1e-013 170 hsa-miR-155 2.0e+003-1.4e+004 0.90 7.18 (-) 2.2e-010 168 hsa-miR-150 1.7e+002-7.0e+002 0.91 4.14 (-) 5.2e-011 198 hsa-miR-342-5p 2.2e+003-8.3e+003 0.97 3.83 (-) 6.2e-019 50 hsa-miR-342-3p 9.1e+002-2.6e+003 0.86 2.87 (-) 1.3e-008 245 MID-18336 2.3e+002-6.4e+002 0.77 2.74 (-) 2.4e-004 365 MID-00078 1.9e+002-5.2e+002 0.78 2.68 (-) 4.7e-004 45 hsa-miR-29c* 2.8e+003-6.6e+003 0.74 2.34 (-) 1.6e-003 382 MID-22331 3.4e+003-7.9e+003 0.85 2.30 (-) 1.2e-004 259 hsa-let-7g 3.8e+002-8.5e+002 0.75 2.25 (-) 4.3e-003 296 hsa-miR-26b 7.5e+002-1.6e+003 0.78 2.16 (-) 2.3e-004 364 MID-00064 6.3e+002-1.3e+003 0.79 2.08 (-) 7.9e-005 314 hsa-miR-361-3p 2.7e+003-5.4e+003 0.80 2.05 (-) 7.5e-007 12 hsa-miR-140-3p (+) the higher expression of this miR is in melanoma (-) the higher expression of this miR is in lymphoma

[0353] FIG. 18 demonstrates binary decisions at node #24 of the decision-tree. Tumors originating in melanoma (diamonds) are easily distinguished from tumors of lymphoma origin (squares) using the expression levels of hsa-miR-342-3p (SEQ ID NO: 50, y-axis) and hsa-miR-30d (SEQ ID NO: 47, x-axis).

TABLE-US-00026 TABLE 26 miR expression (in fluorescence units) distinguishing between B-cell lymphoma and T-cell lymphoma median values auROC fold-change p-value SEQ ID NO. miR name 8.3e+002-2.8e+002 0.74 2.96 (+) 3.7e-005 11 hsa-miR-138 6.7e+002-2.8e+002 0.72 2.37 (+) 2.2e-003 191 hsa-miR-29c 1.2e+003-5.9e+002 0.76 2.02 (+) 1.4e-003 48 hsa-miR-30e 6.7e+002-1.8e+003 0.79 2.77 (-) 1.1e-006 35 hsa-miR-21* 1.5e+003-3.9e+003 0.68 2.68 (-) 2.6e-003 228 hsa-miR-886-3p (+) the higher expression of this miR is in B-cell lymphoma (-) the higher expression of this miR is in T-cell lymphoma

[0354] hsa-miR-30e (SEQ ID NO: 48) and hsa-miR-21* (SEQ ID NO: 35) are used at node 25 of the binary-tree-classifier detailed in the invention to distinguish between B-cell lymphoma and T-cell lymphoma.

TABLE-US-00027 TABLE 27 miR expression (in fluorescence units) distinguishing between lung small cell carcinoma and other neuroendocrine tumors selected from the group consisting of lung carcinoid, medullary thyroid carcinoma, gastrointestinal tract carcinoid and pancreatic islet cell tumor median values auROC fold-change p-value SEQ ID NO. miR name 1.2e+004-1.2e+003 0.99 9.68 (+) 3.3e-021 158 hsa-miR-106a 7.3e+003-7.9e+002 1.00 9.17 (+) 3.4e-022 20 hsa-miR-17 1.4e+003-1.6e+002 0.99 8.53 (+) 8.2e-022 176 hsa-miR-18a 5.8e+003-7.0e+002 1.00 8.38 (+) 7.4e-021 186 hsa-miR-20a 1.1e+004-1.5e+003 0.98 7.71 (+) 1.7e-022 148 hsa-miR-93 4.7e+003-6.7e+002 0.89 6.99 (+) 1.0e-008 36 hsa-miR-210 2.2e+003-3.7e+002 0.95 5.87 (+) 2.8e-016 51 hsa-miR-345 8.9e+003-1.8e+003 0.95 4.96 (+) 1.6e-010 172 hsa-miR-15b 8.2e+003-1.8e+003 0.98 4.68 (+) 6.3e-020 260 hsa-miR-106b 1.1e+003-2.4e+002 0.91 4.62 (+) 7.7e-010 265 hsa-miR-130b 8.0e+003-1.8e+003 0.94 4.33 (+) 2.7e-013 67 hsa-miR-92a 4.1e+003-9.8e+002 0.98 4.15 (+) 2.6e-019 188 hsa-miR-25 1.1e+003-3.4e+002 0.98 3.40 (+) 7.9e-016 277 hsa-miR-17* 2.5e+003-8.3e+002 0.99 2.96 (+) 1.8e-011 284 hsa-miR-19b 5.1e+002-1.8e+002 0.74 2.84 (+) 6.1e-004 302 hsa-miR-301a 7.9e+002-2.9e+002 0.91 2.78 (+) 8.7e-010 68 hsa-miR-92b 9.9e+002-4.3e+002 0.69 2.28 (+) 5.1e-002 168 hsa-miR-150 2.5e+003-1.1e+003 0.70 2.24 (+) 4.5e-003 242 MID-16489 1.4e+003-6.6e+002 0.91 2.12 (+) 1.1e-009 204 hsa-miR-425 5.0e+001-1.6e+003 0.91 31.23 (-) 4.5e-009 162 hsa-miR-129-3p 1.1e+002-1.6e+003 0.91 14.13 (-) 8.6e-009 177 hsa-miR-192 7.6e+001-7.9e+002 0.91 10.42 (-) 1.5e-008 27 hsa-miR-194 5.5e+002-5.0e+003 0.92 9.14 (-) 1.7e-009 65 hsa-miR-7 7.1e+001-5.7e+002 0.78 8.02 (-) 5.8e-005 263 hsa-miR-129* 2.5e+002-1.6e+003 0.80 6.30 (-) 3.5e-005 155 hsa-miR-127-3p 1.5e+002-9.1e+002 0.96 6.05 (-) 3.5e-015 191 hsa-miR-29c 3.3e+002-2.0e+003 0.93 5.99 (-) 3.3e-013 190 hsa-miR-29b 1.7e+002-9.9e+002 0.99 5.76 (-) 1.3e-020 45 hsa-miR-29c* 1.2e+002-6.6e+002 0.75 5.60 (-) 8.0e-004 59 hsa-miR-487b 1.8e+003-8.0e+003 0.90 4.44 (-) 1.6e-012 43 hsa-miR-29a 1.3e+004-4.9e+004 0.88 3.87 (-) 3.3e-006 56 hsa-miR-375 1.6e+002-5.5e+002 0.95 3.37 (-) 9.6e-011 266 hsa-miR-132 4.0e+003-1.2e+004 0.82 2.98 (-) 9.4e-006 14 hsa-miR-143 7.8e+003-2.3e+004 0.85 2.89 (-) 6.1e-006 15 hsa-miR-145 1.2e+004-3.4e+004 0.79 2.83 (-) 4.3e-005 8 hsa-miR-125b 4.5e+003-1.2e+004 0.97 2.70 (-) 1.7e-014 7 hsa-miR-125a-5p 1.9e+003-5.0e+003 0.89 2.67 (-) 3.6e-010 39 hsa-miR-22 2.5e+003-5.7e+003 0.79 2.25 (-) 8.7e-004 189 hsa-miR-27b 1.1e+003-2.4e+003 0.64 2.18 (-) 4.1e-002 249 MID-20524 2.2e+003-4.8e+003 0.72 2.14 (-) 8.5e-003 231 hsa-miR-99a 9.6e+003-2.0e+004 0.82 2.12 (-) 1.3e-003 293 hsa-miR-23b 5.1e+003-1.0e+004 0.80 2.01 (-) 6.3e-005 2 hsa-let-7e (+) the higher expression of this miR is in lung small cell carcinoma (-) the higher expression of this miR is in other neuroendocrine tumors

[0355] hsa-miR-17 (SEQ ID NO: 20) and hsa-miR-29c* (SEQ ID NO: 45) are used at node #26 of the binary-tree-classifier detailed in the invention to distinguish between lung small cell carcinoma and other neuroendocrine tumors.

TABLE-US-00028 TABLE 28 miR expression (in fluorescence units) distinguishing between medullary thyroid carcinoma and other neuroendocrine tumors selected from the group consisting of lung carcinoid, gastrointestinal tract carcinoid and pancreatic islet cell tumor median values auROC fold-change p-value SEQ ID NO. miR name 4.4e+003-5.5e+001 0.84 79.70 (+) 1.5e-007 159 hsa-miR-124 4.0e+004-4.9e+003 0.98 8.07 (+) 1.6e-015 40 hsa-miR-222 1.9e+004-2.8e+003 0.98 6.85 (+) 4.8e-016 147 hsa-miR-221 1.1e+003-2.0e+002 0.70 5.55 (+) 1.1e-003 11 hsa-miR-138 3.2e+002-7.8e+001 0.83 4.12 (+) 7.6e-007 311 hsa-miR-335 5.8e+003-1.5e+003 0.86 3.91 (+) 1.3e-006 4 hsa-miR-10a 6.3e+004-1.7e+004 0.83 3.61 (+) 3.9e-006 8 hsa-miR-125b 1.1e+004-3.2e+003 0.79 3.43 (+) 5.5e-005 231 hsa-miR-99a 4.3e+002-2.0e+002 0.78 2.10 (+) 2.8e-004 301 hsa-miR-29b-2* 7.9e+003-3.8e+003 0.82 2.06 (+) 4.4e-005 297 hsa-miR-27a 1.4e+002-4.0e+002 0.95 2.95 (-) 7.5e-011 68 hsa-miR-92b 1.1e+003-2.8e+003 0.87 2.50 (-) 3.2e-006 67 hsa-miR-92a 1.8e+002-3.7e+002 0.76 2.07 (-) 2.0e-003 265 hsa-miR-130b 4.4e+002-9.0e+002 0.75 2.04 (-) 2.1e-003 36 hsa-miR-210 (+) the higher expression of this miR is in medullary thyroid carcinoma (-) the higher expression of this miR is in other neuroendocrine tumors

[0356] FIG. 19 demonstrates binary decisions at node #27 of the decision-tree. Tumors originating in medullary thyroid carcinoma (diamonds) are easily distinguished from tumors of other neuroendocrine origin (squares) using the expression levels of hsa-miR-92b (SEQ ID NO: 68, y-axis), hsa-miR-222 (SEQ ID NO: 40, x-axis) and hsa-miR-92a (SEQ ID NO: 67, z-axis).

TABLE-US-00029 TABLE 29 miR expression (in fluorescence units) distinguishing between lung carcinoid tumors and GI neuroendocrine tumors selected from the group consisting of gastrointestinal tract carcinoid and pancreatic islet cell tumor median values auROC fold-change p-value SEQ ID NO. miR name 4.0e+003-9.9e+001 0.90 40.08 (+) 1.9e-010 331 hsa-miR-432 6.0e+003-1.5e+002 0.86 39.24 (+) 4.6e-008 162 hsa-miR-129-3p 6.3e+003-1.9e+002 0.87 34.16 (+) 7.8e-009 59 hsa-miR-487b 1.3e+003-5.5e+001 0.88 23.36 (+) 2.9e-010 326 hsa-miR-409-5p 1.1e+003-5.0e+001 0.88 21.14 (+) 5.2e-010 306 hsa-miR-323-3p 1.0e+003-5.5e+001 0.87 18.59 (+) 1.5e-009 350 hsa-miR-539 7.9e+002-5.6e+001 0.84 14.25 (+) 1.4e-008 317 hsa-miR-369-5p 1.0e+004-7.2e+002 0.86 13.95 (+) 3.2e-007 155 hsa-miR-127-3p 1.7e+003-1.2e+002 0.86 13.60 (+) 2.1e-008 325 hsa-miR-409-3p 1.6e+003-1.2e+002 0.88 13.10 (+) 4.2e-009 318 hsa-miR-370 9.5e+002-7.3e+001 0.81 13.03 (+) 3.1e-006 339 hsa-miR-495 9.5e+002-7.4e+001 0.84 12.92 (+) 5.7e-007 264 hsa-miR-129-5p 6.4e+002-5.0e+001 0.91 12.84 (+) 1.6e-013 332 hsa-miR-433 6.5e+002-5.7e+001 0.88 11.52 (+) 5.1e-011 262 hsa-miR-127-5p 5.6e+002-5.2e+001 0.90 10.76 (+) 2.7e-012 336 hsa-miR-485-5p 2.0e+003-1.9e+002 0.86 10.44 (+) 4.2e-008 324 hsa-miR-382 7.9e+002-7.8e+001 0.83 10.20 (+) 1.3e-007 322 hsa-miR-379 6.0e+002-5.9e+001 0.89 10.15 (+) 9.6e-012 330 hsa-miR-431* 4.7e+002-5.0e+001 0.90 9.41 (+) 6.1e-012 321 hsa-miR-377* 1.3e+003-1.4e+002 0.80 9.40 (+) 1.5e-005 263 hsa-miR-129* 4.7e+002-5.0e+001 0.86 9.35 (+) 1.8e-008 309 hsa-miR-329 4.9e+002-5.3e+001 0.79 9.24 (+) 3.1e-005 53 hsa-miR-34c-5p 1.1e+003-1.2e+002 0.83 9.05 (+) 6.4e-007 320 hsa-miR-376c 1.1e+003-1.2e+002 0.86 8.81 (+) 2.3e-008 275 hsa-miR-154 6.5e+002-8.4e+001 0.83 7.73 (+) 8.1e-007 352 hsa-miR-543 9.9e+002-1.3e+002 0.82 7.49 (+) 3.2e-007 312 hsa-miR-337-5p 6.2e+002-8.8e+001 0.86 7.10 (+) 3.0e-008 355 hsa-miR-654-3p 3.5e+002-5.0e+001 0.91 7.05 (+) 3.2e-013 367 MID-00465 6.0e+002-1.0e+002 0.84 5.76 (+) 5.9e-007 269 hsa-miR-134 3.2e+003-8.5e+002 0.91 3.84 (+) 1.2e-011 64 hsa-miR-652 3.2e+002-1.1e+002 0.83 2.84 (+) 2.3e-005 308 hsa-miR-328 2.6e+003-9.4e+002 0.74 2.78 (+) 1.1e-003 175 hsa-miR-183 2.8e+003-1.0e+003 0.87 2.73 (+) 3.0e-006 190 hsa-miR-29b 3.9e+003-1.6e+003 0.88 2.49 (+) 6.9e-010 54 hsa-miR-361-5p 4.1e+002-1.7e+002 0.67 2.44 (+) 2.1e-002 302 hsa-miR-301a 4.0e+003-1.7e+003 0.79 2.41 (+) 5.9e-004 152 hsa-miR-182 4.0e+002-1.7e+002 0.88 2.39 (+) 4.7e-007 301 hsa-miR-29b-2* 8.7e+002-3.7e+002 0.77 2.36 (+) 6.8e-005 266 hsa-miR-132 7.7e+003-3.3e+003 0.82 2.34 (+) 5.4e-006 47 hsa-miR-30d 3.7e+002-1.6e+002 0.70 2.32 (+) 5.8e-003 313 hsa-miR-338-3p 3.3e+002-1.5e+002 0.66 2.16 (+) 1.3e-002 359 hsa-miR-708 5.5e+003-2.5e+003 0.68 2.16 (+) 4.2e-002 65 hsa-miR-7 2.1e+003-9.9e+002 0.78 2.13 (+) 7.0e-005 307 hsa-miR-324-5p 1.2e+003-5.9e+002 0.81 2.02 (+) 1.6e-004 191 hsa-miR-29c 3.5e+002-1.9e+003 0.88 5.36 (-) 1.0e-007 242 MID-16489 6.5e+002-1.9e+003 0.76 2.96 (-) 4.9e-004 4 hsa-miR-10a 1.3e+003-3.6e+003 0.84 2.79 (-) 1.9e-006 147 hsa-miR-221 2.2e+003-5.9e+003 0.81 2.75 (-) 8.5e-006 40 hsa-miR-222 2.6e+002-6.8e+002 0.76 2.56 (-) 7.4e-004 372 MID-16469 3.5e+002-8.9e+002 0.71 2.56 (-) 4.7e-003 168 hsa-miR-150 1.5e+002-3.7e+002 0.83 2.55 (-) 4.8e-005 16 hsa-miR-146a 1.9e+003-4.7e+003 0.82 2.40 (-) 1.7e-005 182 hsa-miR-199a-5p 1.3e+003-3.0e+003 0.79 2.35 (-) 1.2e-004 167 hsa-miR-149* 2.1e+002-4.8e+002 0.84 2.26 (-) 3.1e-005 356 hsa-miR-658 1.2e+003-2.8e+003 0.74 2.25 (-) 1.4e-003 148 hsa-miR-93 1.4e+003-3.1e+003 0.70 2.23 (-) 1.9e-002 382 MID-22331 8.0e+002-1.8e+003 0.83 2.21 (-) 2.9e-005 37 hsa-miR-214 4.4e+002-8.9e+002 0.79 2.01 (-) 2.1e-004 364 MID-00064 2.1e+002-4.2e+002 0.78 2.01 (-) 1.1e-003 35 hsa-miR-21* (+) the higher expression of this miR is in lung carcinoid tumors (-) the higher expression of this miR is in GI neuroendocrine tumors

[0357] hsa-miR-652 (SEQ ID NO: 64), hsa-miR-34c-5p (SEQ ID NO: 53) and hsa-miR-214 (SEQ ID NO: 37) are used at node 28 of the binary-tree-classifier detailed in the invention to distinguish between lung carcinoid tumors and GI neuroendocrine tumors.

TABLE-US-00030 TABLE 30 miR expression (in fluorescence units) distinguishing between pancreatic islet cell tumors and GI neuroendocrine carcinoid tumors selected from the group consisting of small intestine and duodenum; appendicitis, stomach and pancreas fold- miR name SEQ ID NO. p-value change auROC median values hsa-miR-129* 263 2.8e-004 20.91 (+) 0.80 2.3e+003 1.1e+002 hsa-miR-217 288 6.6e-003 9.61 (+) 0.72 4.8e+002 5.0e+001 hsa-miR-148a 18 6.8e-006 8.54 (+) 0.90 1.6e+003 1.9e+002 hsa-miR-216a 286 2.7e-002 8.34 (+) 0.68 4.3e+002 5.2e+001 hsa-miR-129-3p 162 4.4e-003 7.22 (+) 0.74 1.8e+003 2.5e+002 hsa-miR-551b 225 2.3e-003 6.65 (+) 0.74 6.6e+002 9.9e+001 hsa-miR-216b 287 5.4e-003 6.04 (+) 0.75 3.0e+002 5.0e+001 hsa-miR-455-3p 206 7.3e-007 3.75 (+) 0.92 7.1e+002 1.9e+002 hsa-miR-451 205 2.5e-003 3.65 (+) 0.79 1.3e+004 3.4e+003 hsa-miR-26b 296 2.8e-004 3.43 (+) 0.83 8.9e+002 2.6e+002 hsa-let-7f 258 3.6e-004 3.29 (+) 0.91 8.7e+003 2.6e+003 hsa-miR-338-3p 313 3.2e-003 3.25 (+) 0.78 5.2e+002 1.6e+002 MID-17866 377 5.0e-005 2.71 (+) 0.85 6.6e+003 2.4e+003 MID-16582 373 1.2e-005 2.45 (+) 0.88 1.9e+004 7.6e+003 hsa-let-7a 256 1.0e-003 2.42 (+) 0.80 6.4e+004 2.7e+004 hsa-let-7d 153 1.8e-004 2.36 (+) 0.89 1.1e+004 4.7e+003 hsa-let-7g 259 2.2e-003 2.28 (+) 0.86 6.4e+003 2.8e+003 hsa-miR-130b 265 1.0e-002 2.11 (+) 0.69 4.8e+002 2.3e+002 hsa-miR-30b 303 7.5e-003 2.09 (+) 0.75 6.4e+003 3.1e+003 hsa-miR-133b 268 5.4e-004 9.40 (-) 0.81 1.0e+002 9.7e+002 hsa-miR-133a 267 5.4e-004 9.22 (-) 0.80 1.1e+002 1.0e+003 hsa-miR-143* 165 2.1e-006 8.37 (-) 0.93 2.3e+002 1.9e+003 hsa-miR-145 15 3.7e-008 8.18 (-) 0.94 1.1e+004 9.2e+004 hsa-miR-145* 272 7.0e-006 8.05 (-) 0.91 6.6e+001 5.3e+002 hsa-miR-143 14 3.7e-009 7.30 (-) 0.96 5.2e+003 3.8e+004 hsa-miR-378 202 6.2e-006 6.35 (-) 0.88 3.1e+002 2.0e+003 MID-00689 236 8.1e-006 4.99 (-) 0.88 2.9e+002 1.4e+003 hsa-miR-422a 203 7.9e-006 4.74 (-) 0.88 1.4e+002 6.4e+002 hsa-miR-10a 4 2.4e-004 3.91 (-) 0.82 9.2e+002 3.6e+003 hsa-miR-150 168 4.4e-003 3.78 (-) 0.78 3.0e+002 1.1e+003 hsa-miR-330-3p 310 3.4e-004 3.23 (-) 0.81 1.1e+002 3.6e+002 hsa-miR-28-3p 298 4.6e-007 3.16 (-) 0.95 2.1e+002 6.7e+002 hsa-miR-194 27 1.4e-002 3.09 (-) 0.74 7.2e+002 2.2e+003 hsa-miR-200b 29 7.6e-005 2.72 (-) 0.91 7.5e+003 2.1e+004 hsa-miR-21 34 2.8e-006 2.57 (-) 0.87 8.5e+003 2.2e+004 hsa-miR-886-3p 228 8.0e-003 2.56 (-) 0.74 6.7e+002 1.7e+003 hsa-miR-100 3 3.6e-003 2.50 (-) 0.77 1.5e+003 3.8e+003 hsa-miR-532-5p 349 4.3e-007 2.14 (-) 0.94 2.5e+002 5.3e+002 hsa-miR-21* 35 8.5e-004 2.06 (-) 0.82 2.4e+002 5.0e+002 hsa-miR-193a-5p 26 5.7e-003 2.01 (-) 0.77 1.6e+002 3.3e+002 (+) the higher expression of this miR is in pancreatic islet cell tumors (-) the higher expression of this miR is in GI neuroendocrine carcinoid tumors

[0358] hsa-miR-21 (SEQ ID NO: 34), and hsa-miR-148a (SEQ ID NO: 18) are used at node 29 of the binary-tree-classifier detailed in the invention to distinguish between pancreatic islet cell tumors and GI neuroendocrine carcinoid tumors.

TABLE-US-00031 TABLE 31 miR expression (in fluorescence units) distinguishing between gastric or esophageal adenocarcinoma and other adenocarcinoma tumors of the gastrointestinal system selected from the group consisting of cholangiocarcinoma or adenocarcinoma of extrahepatic biliary tract, pancreatic adenocarcinoma and colorectal adenocarcinoma SEQ ID fold- miR name NO. p-value change auROC median values hsa-miR-133a 267 4.6e-008 9.14 (+) 0.74 6.2e+002 6.7e+001 hsa-miR-133b 268 3.9e-008 8.73 (+) 0.74 5.5e+002 6.3e+001 hsa-miR-143* 165 3.9e-007 4.26 (+) 0.75 2.5e+003 5.9e+002 hsa-miR-145 15 4.5e-004 2.82 (+) 0.71 7.9e+004 2.8e+004 hsa-miR-143 14 1.3e-003 2.55 (+) 0.68 3.2e+004 1.3e+004 hsa-miR-658 356 8.2e-004 2.53 (+) 0.71 1.3e+003 5.1e+002 hsa-miR-149* 167 2.2e-004 2.33 (+) 0.72 7.2e+003 3.1e+003 MID-17576 376 7.2e-004 2.22 (+) 0.69 3.1e+003 1.4e+003 MID-16469 372 3.0e-004 2.20 (+) 0.71 1.4e+003 6.5e+002 hsa-miR-145* 272 3.0e-004 2.14 (+) 0.69 3.2e+002 1.5e+002 MID-15986 370 3.8e-004 2.11 (+) 0.74 2.9e+003 1.4e+003 hsa-miR-224 42 5.4e-008 6.57 (-) 0.83 5.5e+001 3.6e+002 hsa-miR-223 41 1.1e-004 2.61 (-) 0.73 1.5e+002 4.0e+002 hsa-miR-1201 146 1.2e-002 1.28 (-) 0.67 9.0e+002 1.2e+003 (+) the higher expression of this miR is in gastric or esophageal adenocarcinoma (-) the higher expression of this miR is in other adenocarcinoma tumors of the gastrointestinal system

[0359] FIG. 20 demonstrates binary decisions at node #30 of the decision-tree. Tumors originating in gastric or esophageal adenocarcinoma (diamonds) are easily distinguished from tumors of other GI adenocarcinoma origin (squares) using the expression levels of hsa-miR-1201 (SEQ ID NO: 146, y-axis), hsa-miR-224 (SEQ ID NO: 42, x-axis) and hsa-miR-210 (SEQ ID NO: 36, z-axis).

TABLE-US-00032 TABLE 32 miR expression (in fluorescence units) distinguishing between colorectal adenocarcinoma and cholangiocarcinoma or adenocarcinoma of biliary tract or pancreas SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-224 42 4.0e-003 2.55 (+) 0.69 5.4e+002 2.1e+002 hsa-miR-203 184 1.2e-003 2.28 (+) 0.70 4.2e+002 1.8e+002 hsa-miR-92a 67 5.1e-007 1.91 (+) 0.77 6.2e+003 3.2e+003 hsa-miR-106a 158 4.6e-007 1.81 (+) 0.81 5.6e+003 3.1e+003 hsa-miR-17 20 1.3e-007 1.81 (+) 0.81 3.2e+003 1.8e+003 hsa-miR-20a 186 7.9e-005 1.80 (+) 0.76 3.2e+003 1.8e+003 hsa-miR-19b 284 1.4e-005 1.75 (+) 0.76 1.9e+003 1.1e+003 MID-17356 389 3.0e-003 1.67 (+) 0.70 2.6e+003 1.6e+003 hsa-miR-422a 203 2.1e-005 1.63 (+) 0.75 5.1e+002 3.1e+002 MID-15965 240 5.6e-003 1.60 (+) 0.67 7.2e+003 4.5e+003 MID-00689 236 1.7e-005 1.59 (+) 0.76 1.1e+003 6.9e+002 hsa-miR-1201 146 2.5e-003 1.53 (+) 0.68 1.6e+003 1.1e+003 hsa-miR-425 204 5.2e-004 1.49 (+) 0.69 1.4e+003 9.1e+002 hsa-miR-29a 43 1.2e-005 1.44 (+) 0.77 9.3e+003 6.5e+003 hsa-miR-18a 176 7.3e-006 1.44 (+) 0.75 6.4e+002 4.5e+002 hsa-miR-378 202 1.4e-004 1.41 (+) 0.72 1.3e+003 9.1e+002 hsa-miR-31 49 2.0e-003 3.39 (-) 0.69 5.3e+002 1.8e+003 hsa-miR-30a 46 2.2e-008 2.39 (-) 0.82 8.2e+002 2.0e+003 hsa-miR-214* 38 1.3e-002 1.47 (-) 0.66 2.5e+002 3.7e+002 hsa-miR-99b 363 2.2e-003 1.41 (-) 0.73 9.0e+002 1.3e+003 (+) the higher expression of this miR is in colorectal adenocarcinoma (-) the higher expression of this miR is in other cholangiocarcinoma or adenocarcinoma tumors of biliary tract or pancreas

[0360] FIG. 21 demonstrates binary decisions at node #31 of the decision-tree. Tumors originating in colorectal adenocarcinoma (diamonds) are easily distinguished from tumors of cholangiocarcinoma or adenocarcinoma of biliary tract or pancreas origin (squares) using the expression levels of hsa-miR-30a (SEQ ID NO: 46, y-axis), hsa-miR-17 (SEQ ID NO: 20, x-axis) and hsa-miR-29a (SEQ ID NO: 43, z-axis).

TABLE-US-00033 TABLE 33 miR expression (in fluorescence units) distinguishing between cholangiocarcinoma or adenocarcinoma of extrahepatic biliary tract and pancreatic adenocarcinoma SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-31 49 1.5e-003 3.06 (+) 0.81 3.4e+003 1.1e+003 hsa-miR-138 11 1.1e-002 2.36 (+) 0.71 3.3e+002 1.4e+002 hsa-miR-141 13 1.7e-002 1.77 (+) 0.70 3.0e+003 1.7e+003 MID-16582 373 1.5e-002 1.65 (+) 0.70 1.8e+004 1.1e+004 hsa-miR-181b 154 9.6e-002 1.63 (+) 0.69 1.4e+003 8.4e+002 hsa-miR-10b 5 5.1e-001 1.62 (+) 0.69 7.0e+002 4.3e+002 hsa-miR-200c 30 7.4e-002 1.61 (+) 0.68 1.5e+004 9.3e+003 hsa-miR-29c* 45 1.3e-002 1.58 (+) 0.72 4.2e+002 2.7e+002 hsa-miR-193b 178 1.1e-001 1.47 (+) 0.66 1.5e+003 1.0e+003 hsa-miR-221 147 1.2e-002 1.36 (+) 0.75 9.0e+003 6.6e+003 hsa-miR-151-3p 274 4.0e-002 1.36 (+) 0.70 6.4e+002 4.7e+002 hsa-miR-146a 16 2.4e-002 1.34 (+) 0.66 7.3e+002 5.4e+002 hsa-miR-222 40 3.7e-002 1.32 (+) 0.71 1.5e+004 1.1e+004 hsa-miR-181a 21 8.4e-002 1.30 (+) 0.71 4.9e+003 3.8e+003 hsa-miR-29a 43 6.3e-002 1.14 (+) 0.66 6.8e+003 6.0e+003 MID-23256 253 2.1e-002 1.81 (-) 0.74 3.3e+002 5.9e+002 MID-18336 245 8.0e-002 1.70 (-) 0.66 1.1e+003 1.9e+003 hsa-let-7a 256 7.4e-003 1.68 (-) 0.73 2.7e+004 4.5e+004 hsa-miR-140-3p 12 9.2e-002 1.51 (-) 0.65 1.8e+003 2.7e+003 MID-16748 374 5.4e-003 1.47 (-) 0.75 9.3e+004 1.4e+005 MID-18395 379 2.9e-002 1.45 (-) 0.66 6.1e+004 8.9e+004 hsa-miR-1973 180 6.6e-002 1.41 (-) 0.69 3.3e+002 4.7e+002 hsa-let-7d 153 2.6e-002 1.40 (-) 0.68 4.3e+003 6.0e+003 hsa-miR-345 51 7.1e-002 1.39 (-) 0.75 3.2e+002 4.4e+002 hsa-miR-34a 52 3.9e-002 1.38 (-) 0.70 4.4e+003 6.1e+003 hsa-let-7c 1 1.4e-002 1.37 (-) 0.73 2.4e+004 3.3e+004 hsa-miR-26a 295 2.8e-002 1.36 (-) 0.68 1.5e+004 2.0e+004 hsa-let-7b 257 3.3e-003 1.35 (-) 0.77 2.9e+004 3.9e+004 MID-23168 385 6.5e-002 1.26 (-) 0.66 4.8e+003 6.1e+003 hsa-miR-23b 293 9.0e-002 1.21 (-) 0.67 1.0e+004 1.3e+004 hsa-miR-24 294 2.6e-002 1.18 (-) 0.68 2.1e+004 2.4e+004 (+) the higher expression of this miR is in pancreatic adenocarcinoma (-) the higher expression of this miR is in cholangiocarcinoma or adenocarcinoma of extrahepatic biliary tract

[0361] hsa-miR-345 (SEQ ID NO: 51), hsa-miR-31 (SEQ ID NO: 49) and hsa-miR-146a (SEQ ID NO: 16) are used at node #32 of the binary-tree-classifier detailed in the invention to distinguish between cholangio cancer or adenocarcinoma of extrahepatic biliary tract and pancreatic adenocarcinoma.

TABLE-US-00034 TABLE 34 miR expression (in fluorescence units) distinguishing between kidney tumors selected from the group consisting of chromophobe renal cell carcinoma, clear cell renal cell carcinoma and papillary renal cell carcinoma and other tumors selected from the group consisting of sarcoma, adrenal (pheochromocytoma, adrenocortical carcinoma) and mesothelioma (pleural mesothelioma) SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-200b 29 7.6e-042 96.12 (+) 0.94 4.8e+003 5.0e+001 hsa-miR-200a 28 3.3e-044 45.03 (+) 0.94 2.3e+003 5.0e+001 hsa-miR-200c 30 1.1e-015 15.36 (+) 0.82 7.7e+002 5.0e+001 hsa-miR-30a 46 8.6e-041 9.73 (+) 0.96 1.1e+004 1.2e+003 hsa-miR-31 49 1.1e-008 9.21 (+) 0.74 1.1e+003 1.2e+002 hsa-miR-30a* 195 8.6e-039 8.87 (+) 0.94 1.7e+003 1.9e+002 hsa-miR-182 152 1.1e-009 6.58 (+) 0.74 5.0e+002 7.5e+001 hsa-miR-183 175 9.8e-011 5.07 (+) 0.76 2.5e+002 5.0e+001 hsa-miR-30d 47 5.0e-033 3.81 (+) 0.92 8.3e+003 2.2e+003 hsa-miR-10a 4 2.5e-016 3.52 (+) 0.83 5.1e+003 1.5e+003 MID-23751 387 6.4e-011 3.15 (+) 0.75 2.3e+002 7.3e+001 hsa-miR-30c 196 1.1e-025 2.95 (+) 0.89 9.5e+003 3.2e+003 hsa-miR-192 177 2.1e-012 2.80 (+) 0.76 4.0e+002 1.4e+002 MID-17375 375 1.7e-015 2.52 (+) 0.79 2.5e+002 9.8e+001 hsa-miR-194 27 2.1e-012 2.43 (+) 0.75 2.2e+002 9.0e+001 hsa-miR-30e* 304 6.5e-013 2.40 (+) 0.76 2.8e+002 1.2e+002 hsa-miR-222 40 1.6e-012 2.37 (+) 0.75 1.6e+004 6.7e+003 hsa-miR-29c 191 9.5e-006 2.33 (+) 0.69 5.9e+002 2.5e+002 hsa-miR-221 147 1.4e-011 2.20 (+) 0.74 9.8e+003 4.5e+003 hsa-miR-21* 35 7.6e-003 2.19 (+) 0.61 9.8e+002 4.5e+002 hsa-miR-146a 16 3.7e-007 2.15 (+) 0.71 6.1e+002 2.8e+002 hsa-miR-21 34 1.2e-003 2.07 (+) 0.64 4.9e+004 2.4e+004 hsa-miR-10b 5 5.4e-004 2.06 (+) 0.66 4.7e+003 2.3e+003 hsa-miR-127-3p 155 2.1e-018 9.53 (-) 0.85 1.2e+002 1.2e+003 hsa-miR-199a-3p 181 4.9e-023 7.57 (-) 0.89 1.6e+003 1.2e+004 hsa-miR-337-5p 312 1.2e-019 7.45 (-) 0.86 5.0e+001 3.7e+002 hsa-miR-199b-5p 183 1.7e-015 7.21 (-) 0.82 1.4e+002 1.0e+003 hsa-miR-199a-5p 182 1.4e-017 6.48 (-) 0.86 2.6e+003 1.7e+004 hsa-miR-376c 320 3.5e-019 5.73 (-) 0.86 5.0e+001 2.9e+002 hsa-miR-487b 59 2.6e-016 5.23 (-) 0.86 6.5e+001 3.4e+002 hsa-miR-214* 38 9.7e-016 5.18 (-) 0.82 9.4e+001 4.9e+002 hsa-miR-382 324 2.2e-017 4.83 (-) 0.86 5.0e+001 2.4e+002 hsa-miR-381 323 6.6e-017 4.27 (-) 0.83 5.0e+001 2.1e+002 hsa-miR-214 37 2.7e-013 4.22 (-) 0.81 1.2e+003 5.0e+003 hsa-miR-379 322 8.5e-018 4.21 (-) 0.86 5.0e+001 2.1e+002 hsa-miR-409-3p 325 1.2e-015 4.14 (-) 0.83 5.0e+001 2.1e+002 hsa-miR-149 19 2.8e-016 3.76 (-) 0.86 6.7e+001 2.5e+002 hsa-miR-224 42 2.8e-007 3.51 (-) 0.71 7.5e+001 2.6e+002 hsa-miR-483-5p 334 1.3e-011 3.25 (-) 0.79 9.9e+001 3.2e+002 hsa-miR-130b 265 9.5e-012 2.08 (-) 0.79 1.5e+002 3.0e+002 hsa-miR-181a* 22 4.8e-009 2.00 (-) 0.76 1.0e+002 2.0e+002 (+) the higher expression of this miR is in kidney tumors (-) the higher expression of this miR is in sarcoma, adrenal and mesothelioma tumors

[0362] FIG. 22 demonstrates binary decisions at node #33 of the decision-tree. Tumors originating in kidney (diamonds) are easily distinguished from tumors of adrenal, mesothelioma and sarcoma origin (squares) using the expression levels of hsa-miR-200b (SEQ ID NO: 29, y-axis), hsa-miR-30a (SEQ ID NO: 46, x-axis) and hsa-miR-149 (SEQ ID NO: 19, z-axis).

TABLE-US-00035 TABLE 35 miR expression (in fluorescence units) distinguishing between pheochromocytoma (neuroendocrine tumor of the adrenal) and all sarcoma, adrenal carcinoma and mesothelioma tumors miR name SEQ ID NO. p-value fold-change auROC median values hsa-miR-7 65 6.7e-067 295.36 (+) 0.96 1.5e+004 5.0e+001 hsa-miR-375 56 5.0e-036 196.58 (+) 0.91 9.8e+003 5.0e+001 hsa-miR-138 11 3.2e-009 29.73 (+) 0.85 4.0e+003 1.3e+002 hsa-miR-129-3p 162 1.5e-021 20.53 (+) 0.94 1.0e+003 5.0e+001 hsa-miR-487b 59 3.0e-008 15.11 (+) 0.84 4.0e+003 2.7e+002 hsa-miR-432 331 7.4e-008 14.54 (+) 0.81 2.2e+003 1.5e+002 hsa-miR-539 350 9.4e-011 12.45 (+) 0.84 8.7e+002 7.0e+001 hsa-miR-127-3p 155 8.3e-005 12.36 (+) 0.80 1.2e+004 9.6e+002 hsa-miR-485-3p 335 1.2e-008 11.61 (+) 0.80 6.8e+002 5.8e+001 hsa-miR-124 159 2.7e-008 11.48 (+) 0.87 5.7e+002 5.0e+001 hsa-miR-485-5p 336 2.3e-014 10.67 (+) 0.86 5.6e+002 5.3e+001 hsa-miR-433 332 1.2e-012 10.38 (+) 0.83 5.2e+002 5.0e+001 hsa-miR-129* 263 1.1e-029 10.28 (+) 0.94 5.1e+002 5.0e+001 hsa-miR-323-3p 306 9.6e-008 9.55 (+) 0.82 4.8e+002 5.0e+001 hsa-miR-495 339 1.0e-005 9.22 (+) 0.79 1.2e+003 1.2e+002 hsa-miR-487a 337 4.4e-006 9.01 (+) 0.80 5.2e+002 5.8e+001 hsa-miR-154 275 4.8e-006 8.75 (+) 0.80 1.4e+003 1.6e+002 hsa-miR-29b-2* 301 6.0e-013 8.56 (+) 0.90 6.1e+002 7.1e+001 hsa-miR-154* 276 3.7e-005 8.54 (+) 0.78 4.5e+002 5.3e+001 hsa-miR-431* 330 4.7e-009 7.77 (+) 0.83 4.1e+002 5.3e+001 hsa-miR-369-5p 317 1.3e-006 7.56 (+) 0.81 7.4e+002 9.8e+001 hsa-miR-329 309 1.4e-007 7.56 (+) 0.80 4.8e+002 6.4e+001 hsa-miR-29c* 45 1.4e-009 7.28 (+) 0.90 1.8e+003 2.4e+002 hsa-miR-370 318 1.5e-005 7.24 (+) 0.79 1.1e+003 1.5e+002 hsa-miR-382 324 1.5e-005 6.74 (+) 0.80 1.5e+003 2.2e+002 hsa-miR-543 352 3.1e-004 6.53 (+) 0.76 8.6e+002 1.3e+002 hsa-miR-29c 191 2.6e-008 6.44 (+) 0.89 1.5e+003 2.3e+002 hsa-miR-127-5p 262 1.1e-005 6.40 (+) 0.79 6.2e+002 9.6e+001 hsa-miR-134 269 2.3e-004 6.39 (+) 0.77 9.6e+002 1.5e+002 hsa-miR-338-3p 313 2.1e-012 6.03 (+) 0.90 3.3e+002 5.4e+001 hsa-miR-149 19 4.8e-008 5.80 (+) 0.84 1.3e+003 2.2e+002 MID-00465 367 7.5e-012 5.32 (+) 0.82 2.7e+002 5.0e+001 hsa-miR-409-5p 326 5.3e-005 5.27 (+) 0.78 5.6e+002 1.1e+002 hsa-miR-409-3p 325 3.1e-004 5.26 (+) 0.76 9.6e+002 1.8e+002 hsa-miR-379 322 8.2e-004 5.25 (+) 0.76 9.2e+002 1.8e+002 hsa-miR-410 327 4.6e-008 5.05 (+) 0.79 2.5e+002 5.0e+001 hsa-miR-29b 190 3.4e-011 4.95 (+) 0.97 4.0e+003 8.1e+002 hsa-miR-1180 261 7.6e-019 4.85 (+) 0.93 3.7e+002 7.6e+001 hsa-miR-377* 321 1.0e-005 4.43 (+) 0.79 2.8e+002 6.4e+001 hsa-miR-873 360 2.9e-009 4.09 (+) 0.81 2.0e+002 5.0e+001 hsa-miR-598 353 1.2e-012 4.08 (+) 0.88 2.1e+002 5.3e+001 hsa-miR-337-5p 312 3.0e-003 4.01 (+) 0.73 1.3e+003 3.3e+002 MID-16270 371 8.8e-005 3.96 (+) 0.77 2.6e+002 6.6e+001 hsa-miR-10b 5 6.2e-005 3.51 (+) 0.85 7.2e+003 2.1e+003 hsa-miR-411 328 2.6e-002 3.40 (+) 0.68 2.3e+002 6.7e+001 hsa-miR-451 205 4.3e-004 3.17 (+) 0.77 2.2e+004 6.9e+003 hsa-miR-199b-5p 183 2.9e-008 12.33 (-) 0.90 1.1e+002 1.3e+003 hsa-miR-214* 38 2.4e-006 4.75 (-) 0.86 1.1e+002 5.5e+002 hsa-miR-199a-3p 181 2.9e-005 4.48 (-) 0.84 3.2e+003 1.5e+004 hsa-miR-214 37 6.8e-005 4.47 (-) 0.85 1.3e+003 5.9e+003 hsa-miR-222 40 5.9e-004 4.23 (-) 0.80 1.8e+003 7.8e+003 hsa-miR-199a-5p 182 9.0e-006 4.13 (-) 0.87 4.5e+003 1.8e+004 hsa-miR-221 147 6.4e-004 4.11 (-) 0.80 1.3e+003 5.2e+003 hsa-miR-146b-5p 17 4.7e-007 3.72 (-) 0.85 1.9e+002 7.0e+002 hsa-miR-224 42 2.3e-003 3.48 (-) 0.72 8.9e+001 3.1e+002 MID-22331 382 3.8e-005 3.42 (-) 0.81 8.0e+002 2.7e+003 hsa-miR-21 34 1.1e-004 3.35 (-) 0.79 7.9e+003 2.6e+004 hsa-miR-148a 18 1.3e-003 3.08 (-) 0.74 1.4e+002 4.2e+002 hsa-miR-100 3 2.6e-005 3.03 (-) 0.83 2.1e+003 6.3e+003 (+) the higher expression of this miR is in pheochromocytoma (-) the higher expression of this miR is in sarcoma, adrenal carcinoma and mesothelioma tumors

[0363] FIG. 23 demonstrates binary decisions at node #34 of the decision-tree. Tumors originating in pheochromocytoma (diamonds) are easily distinguished from tumors of adrenal, mesothelioma and sarcoma origin (squares) using the expression levels of hsa-miR-375 (SEQ ID NO: 56, y-axis) and hsa-miR-7 (SEQ ID NO: 65, x-axis).

TABLE-US-00036 TABLE 36 miR expression (in fluorescence units) distinguishing between adrenal carcinoma and mesothelioma or sarcoma tumors SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-509-3p 61 1.3e-040 51.10 (+) 0.98 2.6e+003 5.0e+001 hsa-miR-483-3p 333 4.9e-007 24.55 (+) 0.76 1.3e+003 5.4e+001 hsa-miR-202 31 8.1e-066 24.01 (+) 0.99 1.2e+003 5.0e+001 hsa-miR-513a-5p 347 2.6e-024 21.83 (+) 0.95 1.4e+003 6.4e+001 hsa-miR-509-3-5p 346 9.3e-030 12.08 (+) 0.96 6.0e+002 5.0e+001 hsa-miR-503 344 2.2e-016 11.82 (+) 0.92 2.2e+003 1.9e+002 hsa-miR-506 345 3.8e-033 10.25 (+) 0.98 5.1e+002 5.0e+001 MID-23751 387 1.2e-026 9.70 (+) 0.96 5.9e+002 6.1e+001 hsa-miR-483-5p 334 6.0e-005 8.66 (+) 0.71 2.7e+003 3.1e+002 hsa-miR-542-5p 351 1.1e-015 7.79 (+) 0.91 1.1e+003 1.4e+002 hsa-miR-382 324 8.5e-005 5.77 (+) 0.72 1.2e+003 2.0e+002 hsa-miR-409-5p 326 3.1e-007 5.44 (+) 0.75 5.5e+002 1.0e+002 hsa-miR-134 269 2.7e-004 5.31 (+) 0.73 7.2e+002 1.4e+002 hsa-miR-127-3p 155 8.9e-003 4.98 (+) 0.69 3.9e+003 7.9e+002 hsa-miR-376c 320 6.4e-003 4.93 (+) 0.68 1.0e+003 2.1e+002 hsa-miR-379 322 2.6e-003 4.84 (+) 0.69 7.8e+002 1.6e+002 hsa-miR-487b 59 4.9e-005 4.53 (+) 0.72 1.0e+003 2.2e+002 hsa-miR-370 318 1.3e-003 4.49 (+) 0.69 6.6e+002 1.5e+002 hsa-miR-409-3p 325 2.9e-004 4.45 (+) 0.71 7.8e+002 1.7e+002 MID-18336 245 3.9e-011 4.19 (+) 0.92 4.7e+003 1.1e+003 MID-23291 254 1.8e-007 3.79 (+) 0.84 1.1e+003 2.9e+002 hsa-miR-432 331 1.5e-004 3.71 (+) 0.70 5.0e+002 1.4e+002 hsa-miR-154 275 1.6e-003 3.60 (+) 0.69 5.3e+002 1.5e+002 hsa-miR-1973 180 7.2e-011 3.48 (+) 0.90 1.1e+003 3.1e+002 hsa-miR-654-3p 355 6.9e-002 3.22 (+) 0.63 5.4e+002 1.7e+002 MID-15986 370 8.6e-009 3.14 (+) 0.86 3.5e+003 1.1e+003 hsa-miR-381 323 4.5e-002 3.07 (+) 0.64 5.4e+002 1.8e+002 hsa-miR-337-5p 312 5.8e-002 3.07 (+) 0.63 8.9e+002 2.9e+002 hsa-miR-193b 178 1.0e-008 3.03 (+) 0.88 5.6e+003 1.8e+003 MID-20524 249 1.6e-006 3.02 (+) 0.81 4.2e+003 1.4e+003 hsa-miR-199b-5p 183 1.3e-015 18.32 (-) 0.96 9.5e+001 1.7e+003 hsa-miR-199a-3p 181 9.7e-014 10.80 (-) 0.95 1.7e+003 1.9e+004 hsa-miR-214* 38 2.6e-016 10.75 (-) 0.97 6.1e+001 6.5e+002 hsa-miR-199a-5p 182 1.9e-015 9.43 (-) 0.97 2.5e+003 2.4e+004 hsa-miR-214 37 1.2e-011 7.89 (-) 0.96 9.0e+002 7.1e+003 hsa-miR-100 3 1.8e-012 4.87 (-) 0.90 1.5e+003 7.5e+003 hsa-miR-193a-3p 25 3.6e-006 3.37 (-) 0.83 7.6e+002 2.5e+003 hsa-miR-152 169 2.5e-006 3.05 (-) 0.80 4.3e+002 1.3e+003 (+) the higher expression of this miR is in adrenal carcinoma (-) the higher expression of this miR is in sarcoma and mesothelioma tumors

[0364] hsa-miR-202 (SEQ ID NO: 31), hsa-miR-509-3p (SEQ ID NO: 61) and hsa-miR-214* (SEQ ID NO: 38) are used at node 35 of the binary-tree-classifier detailed in the invention to distinguish between adrenal carcinoma and sarcoma or mesothelioma tumors.

TABLE-US-00037 TABLE 37 miR expression (in fluorescence units) distinguishing between GIST and mesothelioma or sarcoma tumors fold- SEQ median values auROC change p-value ID NO. miR name 2.4e+002 5.6e+003 0.97 23.39 (+) 4.2e-033 165 hsa-miR-143* 4.9e+003 1.0e+005 0.97 21.41 (+) 1.7e-025 14 hsa-miR-143 8.1e+003 1.5e+005 0.99 18.42 (+) 4.2e-026 15 hsa-miR-145 5.0e+001 7.9e+002 0.87 15.77 (+) 1.5e-010 333 hsa-miR-483-3p 6.2e+001 8.4e+002 0.98 13.54 (+) 1.5e-037 272 hsa-miR-145* 1.6e+002 1.6e+003 0.99 9.58 (+) 2.7e-024 270 hsa-miR-139-5p 1.8e+002 1.8e+003 0.96 9.49 (+) 2.7e-019 45 hsa-miR-29c* 6.1e+001 5.8e+002 0.95 9.48 (+) 7.9e-028 301 hsa-miR-29b-2* 1.9e+002 1.5e+003 0.94 7.89 (+) 1.9e-015 191 hsa-miR-29c 1.2e+003 6.3e+003 0.96 5.12 (+) 8.8e-014 46 hsa-miR-30a 1.9e+002 7.3e+002 0.93 3.84 (+) 6.6e-013 195 hsa-miR-30a* 2.4e+002 8.7e+002 0.92 3.66 (+) 1.8e-008 266 hsa-miR-132 6.1e+002 2.2e+003 0.91 3.52 (+) 4.2e-008 190 hsa-miR-29b 1.9e+002 6.5e+002 0.82 3.50 (+) 1.5e-006 19 hsa-miR-149 2.6e+002 9.0e+002 0.82 3.47 (+) 4.8e-005 334 hsa-miR-483-5p 9.3e+002 1.9e+002 0.70 5.00 (-) 2.1e-003 155 hsa-miR-127-3p 3.4e+003 7.1e+002 0.88 4.74 (-) 2.3e-007 25 hsa-miR-193a-3p 7.0e+003 1.9e+003 0.79 3.64 (-) 5.5e-004 147 hsa-miR-221 9.8e+003 2.8e+003 0.78 3.54 (-) 1.1e-003 40 hsa-miR-222 3.2e+004 9.8e+003 0.75 3.26 (-) 1.1e-003 34 hsa-miR-21 (+) the higher expression of this miR is in GIST (-) the higher expression of this miR is in sarcoma and mesothelioma tumors

[0365] hsa-miR-29C* (SEQ ID NO: 45) and hsa-miR-143 (SEQ ID NO: 14) are used at node 36 of the binary-tree-classifier detailed in the invention to distinguish between GIST and sarcoma or mesothelioma tumors.

TABLE-US-00038 TABLE 38 miR expression (in fluorescence units) distinguishing between chromophobe renal cell carcinoma tumors and clear cell or papillary renal cell carcinoma tumors fold- SEQ median values auROC change p-value ID NO. miR name 8.8e+001 2.1e+003 0.99 23.68 (+) 4.7e-017 13 hsa-miR-141 3.0e+002 5.7e+003 0.99 18.81 (+) 8.4e-012 30 hsa-miR-200c 6.6e+001 9.8e+002 0.99 14.85 (+) 7.5e-019 361 hsa-miR-874 5.0e+001 7.4e+002 0.97 14.80 (+) 1.0e-014 280 hsa-miR-187 5.0e+001 7.2e+002 0.98 14.47 (+) 4.7e-018 362 hsa-miR-891a 5.3e+003 7.4e+004 0.98 13.97 (+) 5.3e-017 147 hsa-miR-221 7.6e+003 9.0e+004 0.97 11.89 (+) 1.4e-015 40 hsa-miR-222 5.3e+001 5.1e+002 0.98 9.66 (+) 1.2e-017 291 hsa-miR-222* 1.4e+002 1.1e+003 0.94 8.01 (+) 2.7e-010 387 MID-23751 7.4e+001 5.4e+002 0.97 7.32 (+) 4.4e-015 290 hsa-miR-221* 1.1e+002 5.6e+002 0.93 4.97 (+) 3.2e-010 299 hsa-miR-296-5p 3.2e+002 1.5e+003 0.90 4.90 (+) 3.1e-007 152 hsa-miR-182 8.4e+002 3.3e+003 0.73 3.89 (+) 6.3e-003 178 hsa-miR-193b 5.4e+003 1.7e+004 0.92 3.26 (+) 2.2e-007 303 hsa-miR-30b 1.1e+003 3.5e+003 0.74 3.20 (+) 8.1e-003 242 MID-16489 6.2e+003 3.3e+002 0.85 18.53 (-) 4.3e-006 49 hsa-miR-31 6.1e+002 5.0e+001 0.90 12.13 (-) 2.1e-007 11 hsa-miR-138 1.8e+003 2.2e+002 0.98 8.38 (-) 6.7e-014 35 hsa-miR-21* 7.9e+004 1.0e+004 0.95 7.54 (-) 3.1e-013 34 hsa-miR-21 3.7e+003 5.4e+002 0.92 6.79 (-) 3.1e-009 36 hsa-miR-210 9.8e+002 1.7e+002 0.97 5.71 (-) 3.1e-013 206 hsa-miR-455-3p 1.0e+003 2.5e+002 0.91 4.07 (-) 4.7e-008 16 hsa-miR-146a 6.0e+002 1.7e+002 0.89 3.64 (-) 7.1e-007 170 hsa-miR-155 7.5e+002 2.1e+002 0.78 3.48 (-) 1.6e-003 177 hsa-miR-192 8.6e+002 2.5e+002 0.86 3.39 (-) 6.6e-006 17 hsa-miR-146b-5p (+) the higher expression of this miR is in chromophobe renal cell carcinoma tumors (-) the higher expression of this miR is in clear cell or papillary renal cell carcinoma tumors

[0366] hsa-miR-210 (SEQ ID NO: 36) and hsa-miR-221 (SEQ ID NO: 147) are used at node #37 of the binary-tree-classifier detailed in the invention to distinguish between chromophobe renal cell carcinoma tumors and clear cell or papillary renal cell carcinoma tumors.

TABLE-US-00039 TABLE 39 miR expression (in fluorescence units) distinguishing between clear cell and papillary renal cell carcinoma tumors SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-503 344 2.3e-005 4.81 (+) 0.89 5.7e+002 1.2e+002 MID-22331 382 5.8e-003 3.65 (+) 0.81 5.9e+003 1.6e+003 hsa-miR-126 9 1.1e-005 3.54 (+) 0.94 6.4e+003 1.8e+003 hsa-miR-494 338 3.0e-003 3.45 (+) 0.82 5.7e+003 1.7e+003 hsa-miR-200b 29 3.1e-004 8.35 (-) 0.87 1.3e+003 1.1e+004 hsa-miR-31 49 3.0e-002 6.61 (-) 0.81 1.3e+003 8.7e+003 hsa-miR-200a 28 5.0e-005 5.30 (-) 0.92 9.5e+002 5.1e+003 hsa-miR-30a* 195 1.1e-009 4.10 (-) 1.00 5.1e+002 2.1e+003 hsa-miR-30a 46 4.5e-010 3.70 (-) 1.00 5.0e+003 1.9e+004 hsa-miR-10a 4 6.9e-004 3.39 (-) 0.86 1.6e+003 5.3e+003 hsa-miR-138 11 2.0e-002 3.23 (-) 0.76 2.3e+002 7.6e+002 MID-23291 254 7.4e-003 3.17 (-) 0.79 2.0e+002 6.4e+002 (+) the higher expression of this miR is in renal clear cell carcinoma tumors (-) the higher expression of this miR is in papillary renal cell carcinoma tumors

[0367] hsa-miR-31 (SEQ ID NO: 49) and hsa-miR-126 (SEQ ID NO: 9) are used at node 38 of the binary-tree-classifier detailed in the invention to distinguish between renal clear cell and papillary cell carcinoma tumors.

TABLE-US-00040 TABLE 40 miR expression (in fluorescence units) distinguishing between pleural mesothelioma and sarcoma tumors SEQ ID fold- miR name NO. p-value change auROC median values hsa-miR-31 49 1.7e-006 13.97 (+) 0.78 1.7e+003 1.2e+002 hsa-miR-21* 35 2.0e-011 5.01 (+) 0.89 2.1e+003 4.3e+002 hsa-miR-146b-5p 17 2.1e-008 2.75 (+) 0.84 1.6e+003 5.9e+002 hsa-miR-21 34 4.8e-010 2.71 (+) 0.89 6.9e+004 2.5e+004 hsa-miR-193a-3p 25 2.3e-005 2.57 (+) 0.77 6.1e+003 2.4e+003 hsa-miR-210 36 5.9e-005 2.49 (+) 0.75 3.1e+003 1.2e+003 hsa-miR-150 168 9.7e-004 2.33 (+) 0.70 1.1e+003 4.6e+002 hsa-miR-155 170 1.4e-004 2.33 (+) 0.75 6.8e+002 2.9e+002 hsa-miR-193a-5p 26 1.4e-005 2.25 (+) 0.76 7.3e+002 3.2e+002 hsa-miR-10a 4 1.7e-004 2.13 (+) 0.76 2.4e+003 1.1e+003 hsa-miR-29b 190 1.1e-003 2.03 (+) 0.70 1.0e+003 5.1e+002 hsa-miR-30a 46 1.5e-005 1.99 (+) 0.77 1.8e+003 9.0e+002 hsa-miR-130a 10 8.9e-003 1.90 (+) 0.71 4.9e+003 2.6e+003 MID-15965 240 1.6e-003 1.88 (+) 0.69 5.2e+003 2.8e+003 hsa-miR-29a 43 1.5e-002 1.71 (+) 0.67 7.3e+003 4.3e+003 hsa-miR-22 39 1.1e-004 1.65 (+) 0.72 8.2e+003 5.0e+003 MID-23168 385 2.9e-002 1.57 (+) 0.64 5.9e+003 3.8e+003 hsa-miR-574-5p 63 1.4e-002 1.55 (+) 0.66 1.5e+003 9.9e+002 hsa-miR-378 202 2.8e-002 1.53 (+) 0.66 1.2e+003 7.5e+002 hsa-miR-199a-3p 181 3.3e-007 3.62 (-) 0.84 7.4e+003 2.7e+004 hsa-miR-214 37 3.2e-006 3.45 (-) 0.81 3.1e+003 1.1e+004 hsa-miR-10b 5 6.7e-008 3.36 (-) 0.85 8.3e+002 2.8e+003 hsa-miR-199b-5p 183 2.0e-004 3.19 (-) 0.74 9.1e+002 2.9e+003 hsa-miR-199a-5p 182 6.0e-005 2.84 (-) 0.80 1.1e+004 3.0e+004 hsa-miR-214* 38 5.7e-005 2.22 (-) 0.78 3.8e+002 8.4e+002 hsa-miR-455-3p 206 2.8e-004 1.69 (-) 0.75 7.0e+002 1.2e+003 hsa-miR-26b 296 2.9e-003 1.61 (-) 0.75 4.4e+002 7.0e+002 hsa-let-7c 1 4.3e-003 1.58 (-) 0.71 3.2e+004 5.0e+004 (+) the higher expression of this miR is in pleural mesothelioma tumors (-) the higher expression of this miR is in sarcoma tumors

[0368] hsa-miR-21* (SEQ ID NO: 35) hsa-miR-130a (SEQ ID NO: 10) and hsa-miR-10b (SEQ ID NO: 5) are used at node 39 of the binary-tree-classifier detailed in the invention to distinguish between pleural mesothelioma tumors and sarcoma tumors.

TABLE-US-00041 TABLE 41 miR expression (in fluorescence units) distinguishing between synovial sarcoma and other sarcoma tumors SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-182 152 2.9e-009 25.03 (+) 0.89 1.3e+003 5.1e+001 hsa-miR-200b 29 9.2e-009 21.59 (+) 0.92 1.1e+003 5.0e+001 hsa-miR-124 159 5.9e-007 19.35 (+) 0.88 9.7e+002 5.0e+001 hsa-miR-200a 28 1.5e-008 12.81 (+) 0.92 6.4e+002 5.0e+001 hsa-miR-495 339 8.2e-005 7.00 (+) 0.88 8.9e+002 1.3e+002 hsa-miR-154 275 1.6e-005 6.94 (+) 0.89 1.1e+003 1.6e+002 hsa-miR-543 352 2.8e-004 6.53 (+) 0.87 7.3e+002 1.1e+002 hsa-miR-149 19 6.1e-006 6.51 (+) 0.91 8.8e+002 1.4e+002 hsa-miR-376c 320 9.6e-005 6.22 (+) 0.86 2.0e+003 3.3e+002 hsa-miR-127-3p 155 1.5e-003 6.05 (+) 0.84 6.2e+003 1.0e+003 hsa-miR-127-5p 262 1.7e-004 5.40 (+) 0.84 5.2e+002 9.7e+001 hsa-miR-214* 38 2.5e-004 5.33 (+) 0.86 4.2e+003 7.8e+002 hsa-miR-214 37 7.3e-003 5.29 (+) 0.84 5.0e+004 9.5e+003 hsa-miR-199a-5p 182 4.7e-003 4.90 (+) 0.87 1.4e+005 2.9e+004 hsa-miR-432 331 1.9e-003 4.56 (+) 0.81 6.4e+002 1.4e+002 hsa-miR-369-5p 317 2.7e-004 4.43 (+) 0.84 5.0e+002 1.1e+002 hsa-miR-381 323 9.6e-003 4.19 (+) 0.78 8.9e+002 2.1e+002 hsa-miR-654-3p 355 2.7e-003 3.96 (+) 0.81 7.7e+002 1.9e+002 hsa-miR-100 3 9.6e-006 3.79 (+) 0.91 2.1e+004 5.6e+003 hsa-miR-196a 282 2.8e-004 3.76 (+) 0.86 6.5e+002 1.7e+002 hsa-miR-337-5p 312 6.0e-003 3.55 (+) 0.80 1.5e+003 4.1e+002 hsa-miR-199a-3p 181 7.4e-003 3.52 (+) 0.86 9.1e+004 2.6e+004 hsa-miR-134 269 6.5e-003 3.41 (+) 0.80 5.5e+002 1.6e+002 hsa-miR-370 318 8.3e-003 3.32 (+) 0.79 6.4e+002 1.9e+002 hsa-miR-487b 59 7.6e-003 3.08 (+) 0.78 1.0e+003 3.2e+002 hsa-miR-132 266 7.1e-007 3.02 (+) 0.87 6.4e+002 2.1e+002 hsa-miR-379 322 1.1e-002 2.92 (+) 0.78 6.3e+002 2.2e+002 hsa-miR-125b 8 1.6e-004 2.83 (+) 0.92 1.2e+005 4.4e+004 hsa-miR-382 324 8.4e-003 2.58 (+) 0.78 6.1e+002 2.4e+002 hsa-miR-130a 10 1.8e-003 2.45 (+) 0.80 6.0e+003 2.4e+003 hsa-miR-222 40 2.1e-010 10.95 (-) 0.96 8.4e+002 9.2e+003 hsa-miR-221 147 7.9e-010 10.19 (-) 0.96 6.7e+002 6.8e+003 hsa-miR-152 169 2.1e-005 4.92 (-) 0.88 3.6e+002 1.8e+003 hsa-miR-451 205 2.8e-002 4.72 (-) 0.72 1.7e+003 7.9e+003 hsa-miR-21 34 4.4e-005 4.59 (-) 0.84 5.9e+003 2.7e+004 hsa-miR-150 168 9.2e-004 4.32 (-) 0.83 1.4e+002 5.9e+002 hsa-miR-143 14 2.2e-007 4.15 (-) 0.92 1.3e+003 5.5e+003 hsa-miR-145 15 2.7e-007 3.30 (-) 0.93 2.9e+003 9.5e+003 hsa-miR-140-3p 12 1.6e-003 3.30 (-) 0.91 1.0e+003 3.5e+003 hsa-miR-30a 46 6.1e-003 2.92 (-) 0.78 3.5e+002 1.0e+003 MID-23794 255 4.1e-004 2.86 (-) 0.82 3.6e+002 1.0e+003 hsa-miR-22 39 1.9e-005 2.82 (-) 0.88 2.0e+003 5.8e+003 hsa-miR-185 23 3.9e-003 2.74 (-) 0.77 4.0e+002 1.1e+003 hsa-miR-29b 190 6.8e-003 2.45 (-) 0.80 2.4e+002 5.8e+002 MID-00689 236 4.9e-003 2.27 (-) 0.79 3.1e+002 7.1e+002 MID-23178 386 2.2e-004 2.10 (-) 0.85 3.2e+004 6.7e+004 MID-18395 379 2.9e-003 2.08 (-) 0.80 3.7e+004 7.7e+004 hsa-miR-378 202 7.4e-003 2.07 (-) 0.78 3.8e+002 7.8e+002 (+) the higher expression of this miR is in synovial sarcoma tumors (-) the higher expression of this miR is in other sarcoma tumors

[0369] hsa-miR-100 (SEQ ID NO: 3) hsa-miR-145 (SEQ ID NO: 15) and hsa-miR-222 (SEQ ID NO: 40) are used at node 40 of the binary-tree-classifier detailed in the invention to distinguish between synovial sarcoma tumors and other sarcoma tumors.

TABLE-US-00042 TABLE 42 miR expression (in fluorescence units) distinguishing between chondrosarcoma and other non synovial sarcoma tumors SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-140-3p 12 2.1e-022 75.69 (+) 1.00 2.2e+005 2.9e+003 hsa-miR-140-5p 271 8.5e-015 35.23 (+) 0.91 5.1e+003 1.5e+002 hsa-miR-455-3p 206 6.1e-015 14.49 (+) 0.98 1.6e+004 1.1e+003 hsa-miR-483-3p 333 3.1e-003 11.03 (+) 0.71 5.5e+002 5.0e+001 hsa-miR-138 11 1.2e-006 11.01 (+) 0.88 1.1e+003 9.5e+001 hsa-miR-455-5p 58 6.3e-012 8.87 (+) 0.87 8.2e+002 9.2e+001 hsa-miR-210 36 1.5e-006 4.37 (+) 0.91 4.7e+003 1.1e+003 hsa-miR-148a 18 3.1e-004 3.98 (+) 0.83 1.4e+003 3.6e+002 hsa-miR-193b 178 2.3e-002 2.36 (+) 0.72 3.6e+003 1.5e+003 hsa-miR-23b 293 1.5e-004 2.13 (+) 0.84 2.8e+004 1.3e+004 hsa-miR-27b 189 5.8e-004 2.05 (+) 0.80 5.5e+003 2.7e+003 MID-22331 382 1.1e-004 5.01 (-) 0.70 6.7e+002 3.4e+003 MID-19962 381 1.2e-004 3.91 (-) 0.81 1.7e+002 6.6e+002 MID-15965 240 1.9e-004 3.76 (-) 0.83 8.5e+002 3.2e+003 MID-20524 249 8.0e-004 3.47 (-) 0.79 4.2e+002 1.5e+003 hsa-miR-10b 5 1.3e-005 3.27 (-) 0.85 9.0e+002 2.9e+003 MID-17866 377 6.9e-005 2.92 (-) 0.78 1.0e+003 2.9e+003 hsa-miR-1978 235 1.3e-003 2.62 (-) 0.75 2.7e+002 7.1e+002 hsa-miR-146b-5p 17 4.4e-005 2.48 (-) 0.81 2.8e+002 7.0e+002 hsa-miR-17 20 2.7e-002 2.36 (-) 0.71 5.7e+002 1.3e+003 MID-23168 385 8.2e-003 2.36 (-) 0.73 1.9e+003 4.5e+003 MID-23017 384 4.8e-003 2.16 (-) 0.74 5.0e+003 1.1e+004 hsa-miR-30a 46 3.2e-004 2.04 (-) 0.79 5.4e+002 1.1e+003 hsa-miR-1979 283 3.0e-004 2.02 (-) 0.83 8.1e+003 1.6e+004 (+) the higher expression of this miR is in chondrosarcoma tumors (-) the higher expression of this miR is in other non-synovial sarcoma tumors

[0370] hsa-miR-140-3p (SEQ ID NO: 12) and hsa-miR-455-5p (SEQ ID NO: 58) are used at node 41 of the binary-tree-classifier detailed in the invention to distinguish between chondrosarcoma tumors and other non-synovial sarcoma tumors.

TABLE-US-00043 TABLE 43 miR expression (in fluorescence units) distinguishing between liposarcoma and other non chondrosarcoma and non synovial sarcoma tumors SEQ fold- miR name ID NO. p-value change auROC median values hsa-miR-26a 295 1.6e-011 6.18 (+) 0.93 1.2e+005 1.9e+004 hsa-miR-451 205 8.1e-003 4.20 (+) 0.73 1.8e+004 4.2e+003 hsa-miR-193a-3p 25 6.5e-006 3.94 (+) 0.84 5.9e+003 1.5e+003 hsa-miR-193a-5p 26 7.5e-007 3.70 (+) 0.88 8.8e+002 2.4e+002 hsa-miR-99a 231 2.2e-005 3.24 (+) 0.88 2.0e+004 6.1e+003 hsa-miR-199b-5p 183 1.9e-003 2.60 (+) 0.75 5.9e+003 2.3e+003 hsa-miR-224 42 1.7e-004 2.54 (+) 0.79 7.9e+002 3.1e+002 MID-23291 254 9.9e-003 2.54 (+) 0.71 7.4e+002 2.9e+002 hsa-miR-150 168 1.5e-002 2.38 (+) 0.71 1.0e+003 4.2e+002 hsa-miR-652 64 1.1e-004 2.36 (+) 0.77 7.7e+002 3.2e+002 hsa-miR-143 14 5.4e-006 2.27 (+) 0.84 1.1e+004 4.8e+003 hsa-miR-193b 178 2.7e-004 2.20 (+) 0.76 3.0e+003 1.4e+003 hsa-miR-145 15 1.1e-004 2.13 (+) 0.78 1.7e+004 7.9e+003 hsa-miR-22 39 9.8e-004 2.12 (+) 0.79 9.7e+003 4.6e+003 hsa-miR-210 36 1.8e-004 4.49 (-) 0.79 3.1e+002 1.4e+003 hsa-miR-181b 154 1.2e-002 2.60 (-) 0.71 9.0e+002 2.4e+003 hsa-miR-130b 265 4.0e-003 2.29 (-) 0.75 2.6e+002 5.9e+002 hsa-miR-181d 174 3.2e-003 2.16 (-) 0.75 3.0e+002 6.5e+002 MID-23017 384 2.0e-004 2.14 (-) 0.79 5.6e+003 1.2e+004 hsa-miR-92a 67 6.6e-004 2.04 (-) 0.80 1.6e+003 3.3e+003 (+) the higher expression of this miR is in liposarcoma tumors (-) the higher expression of this miR is in other non-chondrosarcoma and non-synovial sarcoma tumors

[0371] hsa-miR-210 (SEQ ID NO: 36) and hsa-miR-193a-5p (SEQ ID NO: 26) are used at node 42 of the binary-tree-classifier detailed in the invention to distinguish between liposarcoma tumors and other non-chondrosarcoma and non-synovial sarcoma tumors.

TABLE-US-00044 TABLE 44 miR expression (in fluorescence units) distinguishing between Ewing sarcoma or osteosarcoma; and rhabdomyosarcoma, malignant fibrous histiocytoma (MFH) or fibrosarcoma miR name SEQ ID NO. p-value fold-change auROC median values hsa-miR-181a* 22 1.1e-006 6.62 (+) 0.87 1.2e+003 1.9e+002 hsa-miR-181b 154 8.7e-009 5.68 (+) 0.91 6.4e+003 1.1e+003 hsa-miR-181a 21 2.9e-010 5.67 (+) 0.93 2.1e+004 3.7e+003 hsa-miR-181d 174 3.5e-006 4.19 (+) 0.85 1.8e+003 4.2e+002 hsa-miR-451 205 1.2e-002 3.27 (+) 0.72 9.4e+003 2.9e+003 hsa-miR-106a 158 2.9e-003 2.63 (+) 0.78 4.7e+003 1.8e+003 hsa-miR-20a 186 2.9e-003 2.52 (+) 0.78 2.8e+003 1.1e+003 hsa-miR-93 148 9.2e-005 2.45 (+) 0.81 4.9e+003 2.0e+003 hsa-miR-17 20 5.1e-003 2.32 (+) 0.77 2.6e+003 1.1e+003 hsa-miR-487b 59 1.1e-002 4.54 (-) 0.71 1.3e+002 6.0e+002 hsa-miR-125b 8 2.9e-005 2.86 (-) 0.84 1.7e+004 4.9e+004 hsa-miR-199b-5p 183 9.4e-003 2.70 (-) 0.72 1.3e+003 3.4e+003 hsa-miR-99a 231 1.1e-003 2.34 (-) 0.76 3.5e+003 8.1e+003 (+) the higher expression of this miR is in Ewing sarcoma or osteosarcoma tumors (-) the higher expression of this miR is in rhabdomyosarcoma, malignant fibrous histiocytoma (MFH) or fibrosarcoma tumors

[0372] hsa-miR-181a (SEQ ID NO: 21) is used at node 43 of the binary-tree-classifier detailed in the invention to distinguish between Ewing sarcoma or osteosarcoma tumors and rhabdomyosarcoma, malignant fibrous histiocytoma (MFH) or fibrosarcoma tumors.

TABLE-US-00045 TABLE 45 miR expression (in fluorescence units) distinguishing between Ewing sarcoma and osteosarcoma miR name SEQ ID NO. p-value fold-change auROC median values hsa-miR-127-3p 155 3.7e-006 6.60 (+) 1.00 1.1e+003 1.6e+002 hsa-miR-195 179 8.9e-004 5.85 (+) 0.97 8.5e+003 1.4e+003 hsa-miR-29a 43 1.4e-002 4.90 (+) 0.86 1.4e+004 2.8e+003 hsa-miR-497 208 1.1e-004 4.58 (+) 1.00 6.5e+003 1.4e+003 hsa-miR-181a-2* 278 1.0e-003 4.42 (+) 0.88 7.6e+002 1.7e+002 hsa-miR-146b-5p 17 6.0e-003 4.05 (+) 0.86 1.6e+003 4.0e+002 MID-23168 385 1.4e-002 2.64 (+) 0.81 8.9e+003 3.4e+003 hsa-miR-181d 174 1.5e-002 2.60 (+) 0.77 2.1e+003 8.0e+002 hsa-miR-10b 5 1.3e-002 2.55 (+) 0.82 4.1e+003 1.6e+003 hsa-miR-34a 52 7.1e-003 2.19 (+) 0.84 4.9e+003 2.2e+003 hsa-let-7b 257 2.7e-004 2.16 (+) 0.97 5.4e+004 2.5e+004 MID-00144 366 2.1e-003 2.12 (+) 0.88 5.2e+002 2.5e+002 hsa-miR-30e 48 6.2e-003 2.06 (+) 0.84 9.4e+002 4.5e+002 hsa-miR-31 49 7.9e-005 25.44 (-) 0.96 5.0e+001 1.3e+003 hsa-miR-140-3p 12 1.4e-003 5.72 (-) 0.89 2.0e+003 1.2e+004 hsa-miR-193a-3p 25 5.2e-005 4.92 (-) 0.94 7.6e+002 3.8e+003 hsa-miR-152 169 3.3e-003 4.09 (-) 0.89 4.4e+002 1.8e+003 hsa-miR-21 34 3.2e-003 3.00 (-) 0.89 1.2e+004 3.7e+004 hsa-miR-21* 35 1.7e-003 2.96 (-) 0.83 2.7e+002 8.1e+002 hsa-miR-185 23 4.2e-003 2.55 (-) 0.88 6.7e+002 1.7e+003 MID-23017 384 1.7e-002 2.53 (-) 0.82 8.2e+003 2.1e+004 hsa-miR-27b 189 3.8e-003 2.52 (-) 0.84 1.7e+003 4.3e+003 MID-17866 377 3.0e-002 2.18 (-) 0.80 2.3e+003 5.1e+003 hsa-miR-130b 265 3.0e-002 2.17 (-) 0.78 4.4e+002 9.6e+002 hsa-miR-24 294 3.3e-003 2.07 (-) 0.82 1.8e+004 3.7e+004 hsa-miR-23b 293 9.0e-003 2.03 (-) 0.86 8.8e+003 1.8e+004 hsa-miR-23a 292 1.6e-002 2.02 (-) 0.80 1.5e+004 3.0e+004 (+) the higher expression of this miR is in Ewing sarcoma tumors (-) the higher expression of this miR is in osteosarcoma tumors

[0373] FIG. 24 demonstrates binary decisions at node #44 of the decision-tree. Tumors originating in Ewing sarcoma (diamonds) are easily distinguished from tumors of osteosarcoma origin (squares) using the expression levels of hsa-miR-31 (SEQ ID NO: 49, y-axis) and hsa-miR-193a-3p (SEQ ID NO: 25, x-axis).

TABLE-US-00046 TABLE 46 miR expression (in fluorescence units) distinguishing between rhabdomyosarcoma and malignant fibrous histiocytoma (MFH) or fibrosarcoma fold- SEQ median values auROC change p-value ID NO. miR name 5.0e+001 4.1e+003 0.96 81.34 (+) 1.9e-007 33 hsa-miR-206 5.7e+001 4.3e+003 0.89 74.89 (+) 1.8e-004 268 hsa-miR-133b 5.9e+001 3.9e+003 0.88 66.65 (+) 3.2e-004 267 hsa-miR-133a 5.0e+001 1.3e+003 0.89 25.89 (+) 3.9e-006 333 hsa-miR-483-3p 5.3e+001 5.2e+002 0.85 9.90 (+) 1.3e-004 276 hsa-miR-154* 5.8e+001 5.6e+002 0.85 9.63 (+) 1.2e-004 319 hsa-miR-376a 5.7e+001 5.1e+002 0.86 9.00 (+) 4.8e-005 306 hsa-miR-323-3p 2.5e+002 1.8e+003 0.84 7.01 (+) 2.8e-003 320 hsa-miR-376c 2.6e+002 1.7e+003 0.82 6.52 (+) 3.9e-003 334 hsa-miR-483-5p 3.1e+002 1.9e+003 0.87 6.22 (+) 5.1e-004 323 hsa-miR-381 1.0e+002 6.3e+002 0.85 6.19 (+) 5.4e-004 300 hsa-miR-299-3p 1.3e+002 7.9e+002 0.82 6.18 (+) 1.4e-003 281 hsa-miR-188-5p 4.1e+002 2.3e+003 0.86 5.73 (+) 1.4e-003 59 hsa-miR-487b 1.5e+002 8.4e+002 0.85 5.68 (+) 8.1e-004 339 hsa-miR-495 3.7e+002 1.7e+003 0.79 4.57 (+) 3.1e-002 316 hsa-miR-362-5p 2.0e+002 9.2e+002 0.80 4.49 (+) 2.4e-003 176 hsa-miR-18a 2.9e+002 1.3e+003 0.82 4.39 (+) 1.4e-003 348 hsa-miR-532-3p 1.8e+002 7.8e+002 0.85 4.27 (+) 4.0e-004 352 hsa-miR-543 4.0e+002 1.7e+003 0.81 4.18 (+) 2.3e-002 349 hsa-miR-532-5p 1.9e+003 7.8e+003 0.87 4.14 (+) 4.9e-004 67 hsa-miR-92a 5.7e+002 2.4e+003 0.86 4.13 (+) 9.2e-004 357 hsa-miR-660 1.3e+002 5.6e+002 0.78 4.13 (+) 4.2e-003 315 hsa-miR-362-3p 2.3e+002 8.6e+002 0.81 3.73 (+) 2.8e-003 343 hsa-miR-502-3p 2.0e+002 7.2e+002 0.84 3.64 (+) 1.5e-003 342 hsa-miR-501-3p 2.3e+002 8.5e+002 0.82 3.62 (+) 6.7e-003 355 hsa-miR-654-3p 1.9e+002 6.7e+002 0.79 3.56 (+) 1.3e-002 340 hsa-miR-500 2.4e+002 8.4e+002 0.80 3.56 (+) 7.9e-003 344 hsa-miR-503 2.2e+003 7.6e+003 0.78 3.53 (+) 7.2e-003 10 hsa-miR-130a 2.6e+002 8.8e+002 0.80 3.35 (+) 3.7e-003 341 hsa-miR-500* 2.6e+002 7.9e+002 0.79 3.06 (+) 7.3e-003 331 hsa-miR-432 9.3e+002 2.7e+003 0.77 2.90 (+) 1.4e-002 20 hsa-miR-17 4.3e+002 1.2e+003 0.86 2.90 (+) 1.0e-003 277 hsa-miR-17* 2.4e+002 6.7e+002 0.83 2.77 (+) 7.0e-003 318 hsa-miR-370 1.6e+003 4.5e+003 0.78 2.75 (+) 1.4e-002 158 hsa-miR-106a 4.3e+002 1.1e+003 0.83 2.67 (+) 3.0e-003 265 hsa-miR-130b 1.0e+003 2.7e+003 0.86 2.63 (+) 7.1e-004 284 hsa-miR-19b 8.6e+002 2.1e+003 0.82 2.43 (+) 8.6e-003 36 hsa-miR-210 6.1e+003 6.8e+002 0.90 8.92 (-) 1.8e-004 183 hsa-miR-199b-5p 1.9e+004 4.5e+003 0.83 4.15 (-) 8.0e-004 40 hsa-miR-222 1.1e+003 3.1e+002 0.90 3.55 (-) 5.6e-005 63 hsa-miR-574-5p 1.1e+004 3.2e+003 0.82 3.52 (-) 2.2e-003 147 hsa-miR-221 5.9e+003 1.8e+003 0.80 3.25 (-) 2.0e-003 43 hsa-miR-29a 5.2e+002 1.6e+002 0.82 3.19 (-) 5.4e-003 289 hsa-miR-22* 5.1e+003 1.7e+003 0.82 3.04 (-) 7.0e-003 52 hsa-miR-34a 8.1e+002 2.9e+002 0.76 2.81 (-) 1.4e-002 190 hsa-miR-29b 1.2e+003 4.5e+002 0.86 2.67 (-) 4.7e-003 4 hsa-miR-10a 3.7e+003 1.5e+003 0.86 2.43 (-) 1.3e-003 5 hsa-miR-10b 7.0e+003 2.9e+003 0.85 2.39 (-) 1.5e-003 39 hsa-miR-22 1.6e+003 6.9e+002 0.78 2.25 (-) 1.5e-002 169 hsa-miR-152 2.9e+003 1.3e+003 0.76 2.19 (-) 2.8e-002 208 hsa-miR-497 (+) the higher expression of this miR is in rhabdomyosarcoma tumors (-) the higher expression of this miR is in MFH or fibrosarcoma tumors

[0374] FIG. 25 demonstrates binary decisions at node #45 of the decision-tree. Tumors originating in Rhabdomyosarcoma (diamonds) are easily distinguished from tumors of malignant fibrous histiocytoma (MFH) or fibrosarcoma origin (squares) using the expression levels of hsa-miR-206 (SEQ ID NO: 33, y-axis), hsa-miR-22 (SEQ ID NO: 39, x-axis) and hsa-miR-487b (SEQ ID NO: 59, z-axis).

TABLE-US-00047 TABLE 47 .beta. values of the decision tree classifier The classification at node 11 is based on the gender of subject rather than on beta values; accordingly, no data is provided for this node. P.sub.TH = 0.5 for all node miR 1 miR 2 miR 3 .beta.0 SEQ SEQ SEQ Node intercept miR hsa- ID NO .beta.1 miR hsa- ID NO .beta.2 miR hsa- ID NO .beta.3 1 -23.3111 miR-372 55 2.3127 2 -26.9408 miR-122 6 2.3127 3 -3.8519 miR-200b 29 1.8567 miR-126 9 -1.379 4 -8.2646 miR-200c 30 1.9582 miR-30a 46 -1.2306 5 17.4706 miR-146a 16 1.1979 let-7e 2 -1.7697 miR-30a 46 -0.88435 6 -32.5621 miR-9* 66 1.5475 miR-92b 68 1.7188 7 -9.5521 miR-222 40 -1.1606 miR-497 208 2.0005 8 -23.053 miR-193a-3p 25 -1.0267 miR-7 65 1.2404 miR-375 56 1.6602 9 -29.3207 miR-194 27 2.0115 miR-21* 35 1.1414 10 1.244 miR-181a 21 -1.5458 miR-143 14 0.9879 12 21.3416 miR-200b 29 -1.942 miR-516a-5p 211 -1.256 13 10.3775 miR-125a-5p 7 -1.1455 miR-205 32 1.1064 miR-345 51 -1.0128 14 -40.666 miR-193a-3p 25 1.9505 miR-342-3p 50 0.93196 miR-375 56 0.82076 15 26.2937 miR-22 39 -1.8153 miR-10a 4 0.61098 miR-205 32 -0.91632 16 9.4008 miR-93 148 -1.3023 miR-138 11 1.5494 miR-10a 4 -1.119 17 42.5529 miR-21 34 -1.801 miR-146b-5p 17 -1.4509 18 0.52521 miR-193a-3p 25 1.7974 miR-31 49 -0.63021 miR-92a 67 -1.3119 19 -20.7179 miR-138 11 0.9662 miR-378 202 -1.3077 miR-21 34 1.6447 20 15.0039 miR-100 3 1.0814 miR-21 34 -2.0444 21 -31.6015 miR-191 24 1.5137 miR-29c 191 0.22547 miR-934 69 1.734 22 -44.3141 miR-10b 5 1.41 let-7c 1 0.86212 miR-361-5p 54 1.6178 23 7.6168 miR-138 11 -0.32773 miR-10b 5 1.3275 miR-185 23 -1.8652 24 2.4904 miR-342-3p 50 -1.7146 miR-30d 47 1.5521 26 -10.0563 miR-17 20 1.9063 miR-29c* 45 -1.3096 27 -2.3904 miR-222 40 1.5531 miR-92b 68 -1.5907 miR-92a 67 -0.63749 28 -22.027 miR-652 64 1.9688 miR-214 37 -0.65807 miR-34c-5p 53 1.0197 29 -11.4697 miR-21 34 1.8457 miR-148a 18 -1.3936 30 21.7628 miR-224 42 -1.3059 miR-210 36 -0.79749 1201 146 -0.50909 31 -17.747 miR-17 20 0.95763 miR-29a 43 1.6268 miR-30a 46 -1.3361 32 -2.3716 miR-31 49 1.0661 miR-146a 16 0.62041 miR-345 51 -1.8214 33 -4.226 miR-200b 29 0.48415 miR-149 19 -2.0172 miR-30a 46 1.0224 34 -29.6828 miR-7 65 2.1394 miR-375 56 0.87847 35 -23.6445 miR-202 31 2.1832 miR-509-3p 61 0.76095 miR-214* 38 -0.057027 36 -41.4047 miR-29c* 45 1.2571 miR-143 14 1.9413 37 -25.1227 miR-221 147 2.2247 miR-210 36 -0.63202 38 -24.5409 miR-31 49 -0.19797 miR-126 9 2.3043 39 -20.7495 miR-130a 10 1.014 miR-10b 5 -1.0484 miR-21* 35 1.7948 40 -6.0971 miR-100 3 1.9198 miR-222 40 -1.0289 miR-145 15 -0.77759 41 -38.5059 miR-140-3p 12 1.6462 miR-455-5p 58 1.6244 42 -10.7873 miR-210 36 -0.84091 miR-193a-5p 26 1.9298 43 -30.4778 miR-181a 21 2.3127 44 31.0975 miR-193a-3p 25 -2.0358 miR-31 49 -1.0974 45 -17.5516 miR-22 39 -0.91078 miR-487b 59 1.0201 miR-206 487 1.8651

TABLE-US-00048 TABLE 48 Using fine-needle aspiration (FNA), pleural effusion or bronchial brushing for the identification of cancer tissue of origin Class Biopsy identified Site Histological Type Sampling Method lung-small Lymph Neuroendocrine; Small percutaneous FNA Node UpperSCC Lung Non-small; squamous percutaneous FNA UpperSCC Lung Non-small; adenocarcinoma percutaneous FNA lung-small Lung Neuroendocrine; Small percutaneous FNA lung-adeno Lung Non-small; adenocarcinoma percutaneous FNA UpperSCC Lung Non-small; squamous percutaneous FNA lung-small Lymph Neuroendocrine; Small transbronchial FNA Node lung-small Lung Neuroendocrine; Small transbronchial FNA lung-adeno Lung Non-small; adenocarcinoma Pleural effusion pleura lung-adeno Lung Non-small; adenocarcinoma Pleural effusion pleura Lung, small Lung Neuroendocrine; Small bronchial brushing Lung, small Lung Neuroendocrine; Small bronchial brushing Lung, small Lung Neuroendocrine; Small bronchial brushing Lung, small Lung Neuroendocrine; Small bronchial brushing Lung, small Lung Neuroendocrine; Small bronchial brushing

[0375] The foregoing description of the specific embodiments so fully reveals the general nature of the invention that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without undue experimentation and without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims.

[0376] It should be understood that the detailed description and specific examples, while indicating preferred embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description.

REFERENCES

[0377] 1. Bentwich, I. et al. Identification of hundreds of conserved and nonconserved human microRNAs. Nat Genet (2005). [0378] 2. Farh, K. K. et al. The Widespread Impact of Mammalian MicroRNAs on mRNA Repression and Evolution. Science (2005). [0379] 3. Griffiths-Jones, S., Grocock, R. J., van Dongen, S., Bateman, A. & Enright, A. J. miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res 34, D140-4 (2006). [0380] 4. He, L. et al. A microRNA polycistron as a potential human oncogene. Nature 435, 828-33 (2005). [0381] 5. Baskerville, S. & Bartel, D. P. Microarray profiling of microRNAs reveals frequent coexpression with neighboring miRNAs and host genes. Rna 11, 241-7 (2005). [0382] 6. Landgraf, P. et al. A Mammalian microRNA Expression Atlas Based on Small RNA Library Sequencing. Cell 129, 1401-14 (2007). [0383] 7. Volinia, S. et al. A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl Acad Sci USA (2006). [0384] 8. Lu, J. et al. MicroRNA expression profiles classify human cancers. Nature 435, 834-8 (2005). [0385] 9. Varadhachary, G. R., Abbruzzese, J. L. & Lenzi, R. Diagnostic strategies for unknown primary cancer. Cancer 100, 1776-85 (2004). [0386] 10. Pimiento, J. M., Teso, D., Malkan, A., Dudrick, S. J. & Palesty, J. A. Cancer of unknown primary origin: a decade of experience in a community-based hospital. Am J Surg 194, 833-7; discussion 837-8 (2007). [0387] 11. Shaw, P. H., Adams, R., Jordan, C. & Crosby, T. D. A clinical review of the investigation and management of carcinoma of unknown primary in a single cancer network. Clin Oncol (R Coll Radiol) 19, 87-95 (2007). [0388] 12. Hainsworth, J. D. & Greco, F. A. Treatment of patients with cancer of an unknown primary site. N Engl J Med 329, 257-63 (1993). [0389] 13. Blaszyk, H., Hartmann, A. & Bjornsson, J. Cancer of unknown primary: clinicopathologic correlations. Apmis 111, 1089-94 (2003). [0390] 14. Bloom, G. et al. Multi-platform, multi-site, microarray-based human tumor classification. Am J Pathol 164, 9-16 (2004). [0391] 15. Ma, X. J. et al. Molecular classification of human cancers using a 92-gene real-time quantitative polymerase chain reaction assay. Arch Pathol Lab Med 130, 465-73 (2006). [0392] 16. Talantov, D. et al. A quantitative reverse transcriptase-polymerase chain reaction assay to identify metastatic carcinoma tissue of origin. J Mol Diagn 8, 320-9 (2006). [0393] 17. Tothill, R. W. et al. An expression-based site of origin diagnostic method designed for clinical application to cancer of unknown origin. Cancer Res 65, 4031-40 (2005). [0394] 18. Shedden, K. A. et al. Accurate molecular classification of human cancers based on gene expression using a simple classifier with a pathological tree-based framework. Am J Pathol 163, 1985-95 (2003). [0395] 19. Raver-Shapira, N. et al. Transcriptional Activation of miR-34a Contributes to p53-Mediated Apoptosis. Mol Cell (2007). [0396] 20. Xiao, C. et al. MiR-150 Controls B Cell Differentiation by Targeting the Transcription Factor c-Myb. Cell 131, 146-59 (2007).

Sequence CWU 1

1

390122RNAHomo sapiens 1ugagguagua gguuguaugg uu 22222RNAHomo sapiens 2ugagguagga gguuguauag uu 22322RNAHomo sapiens 3aacccguaga uccgaacuug ug 22423RNAHomo sapiens 4uacccuguag auccgaauuu gug 23523RNAHomo sapiens 5uacccuguag aaccgaauuu gug 23622RNAHomo sapiens 6uggaguguga caaugguguu ug 22724RNAHomo sapiens 7ucccugagac ccuuuaaccu guga 24822RNAHomo sapiens 8ucccugagac ccuaacuugu ga 22922RNAHomo sapiens 9ucguaccgug aguaauaaug cg 221022RNAHomo sapiens 10cagugcaaug uuaaaagggc au 221123RNAHomo sapiens 11agcugguguu gugaaucagg ccg 231221RNAHomo sapiens 12uaccacaggg uagaaccacg g 211322RNAHomo sapiens 13uaacacuguc ugguaaagau gg 221421RNAHomo sapiens 14ugagaugaag cacuguagcu c 211523RNAHomo sapiens 15guccaguuuu cccaggaauc ccu 231622RNAHomo sapiens 16ugagaacuga auuccauggg uu 221722RNAHomo sapiens 17ugagaacuga auuccauagg cu 221822RNAHomo sapiens 18ucagugcacu acagaacuuu gu 221923RNAHomo sapiens 19ucuggcuccg ugucuucacu ccc 232023RNAHomo sapiens 20caaagugcuu acagugcagg uag 232123RNAHomo sapiens 21aacauucaac gcugucggug agu 232222RNAHomo sapiens 22accaucgacc guugauugua cc 222322RNAHomo sapiens 23uggagagaaa ggcaguuccu ga 222423RNAHomo sapiens 24caacggaauc ccaaaagcag cug 232522RNAHomo sapiens 25aacuggccua caaaguccca gu 222622RNAHomo sapiens 26ugggucuuug cgggcgagau ga 222722RNAHomo sapiens 27uguaacagca acuccaugug ga 222822RNAHomo sapiens 28uaacacuguc ugguaacgau gu 222922RNAHomo sapiens 29uaauacugcc ugguaaugau ga 223023RNAHomo sapiens 30uaauacugcc ggguaaugau gga 233120RNAHomo sapiens 31agagguauag ggcaugggaa 203222RNAHomo sapiens 32uccuucauuc caccggaguc ug 223322RNAHomo sapiens 33uggaauguaa ggaagugugu gg 223422RNAHomo sapiens 34uagcuuauca gacugauguu ga 223521RNAHomo sapiens 35caacaccagu cgaugggcug u 213622RNAHomo sapiens 36cugugcgugu gacagcggcu ga 223722RNAHomo sapiens 37acagcaggca cagacaggca gu 223822RNAHomo sapiens 38ugccugucua cacuugcugu gc 223922RNAHomo sapiens 39aagcugccag uugaagaacu gu 224021RNAHomo sapiens 40agcuacaucu ggcuacuggg u 214122RNAHomo sapiens 41ugucaguuug ucaaauaccc ca 224221RNAHomo sapiens 42caagucacua gugguuccgu u 214322RNAHomo sapiens 43uagcaccauc ugaaaucggu ua 224422RNAHomo sapiens 44uagcaccauu ugaaaucggu ua 224522RNAHomo sapiens 45ugaccgauuu cuccuggugu uc 224622RNAHomo sapiens 46uguaaacauc cucgacugga ag 224722RNAHomo sapiens 47uguaaacauc cccgacugga ag 224822RNAHomo sapiens 48uguaaacauc cuugacugga ag 224921RNAHomo sapiens 49aggcaagaug cuggcauagc u 215023RNAHomo sapiens 50ucucacacag aaaucgcacc cgu 235122RNAHomo sapiens 51gcugacuccu aguccagggc uc 225222RNAHomo sapiens 52uggcaguguc uuagcugguu gu 225323RNAHomo sapiens 53aggcagugua guuagcugau ugc 235422RNAHomo sapiens 54uuaucagaau cuccaggggu ac 225523RNAHomo sapiens 55aaagugcugc gacauuugag cgu 235622RNAHomo sapiens 56uuuguucguu cggcucgcgu ga 225721RNAHomo sapiens 57acuggacuug gagucagaag g 215822RNAHomo sapiens 58uaugugccuu uggacuacau cg 225922RNAHomo sapiens 59aaucguacag ggucauccac uu 226021RNAHomo sapiens 60cagcagcaca cugugguuug u 216122RNAHomo sapiens 61ugauugguac gucugugggu ag 226223RNAHomo sapiens 62uucucgagga aagaagcacu uuc 236323RNAHomo sapiens 63ugagugugug ugugugagug ugu 236421RNAHomo sapiens 64aauggcgcca cuaggguugu g 216523RNAHomo sapiens 65uggaagacua gugauuuugu ugu 236622RNAHomo sapiens 66auaaagcuag auaaccgaaa gu 226722RNAHomo sapiens 67uauugcacuu gucccggccu gu 226822RNAHomo sapiens 68uauugcacuc gucccggccu cc 226922RNAHomo sapiens 69ugucuacuac uggagacacu gg 227084RNAHomo sapiens 70gcauccgggu ugagguagua gguuguaugg uuuagaguua cacccuggga guuaacugua 60caaccuucua gcuuuccuug gagc 847179RNAHomo sapiens 71cccgggcuga gguaggaggu uguauaguug aggaggacac ccaaggagau cacuauacgg 60ccuccuagcu uuccccagg 797280RNAHomo sapiens 72ccuguugcca caaacccgua gauccgaacu ugugguauua guccgcacaa gcuuguaucu 60auagguaugu gucuguuagg 8073110RNAHomo sapiens 73gaucugucug ucuucuguau auacccugua gauccgaauu uguguaagga auuuuguggu 60cacaaauucg uaucuagggg aauauguagu ugacauaaac acuccgcucu 11074110RNAHomo sapiens 74ccagagguug uaacguuguc uauauauacc cuguagaacc gaauuugugu gguauccgua 60uagucacaga uucgauucua ggggaauaua uggucgaugc aaaaacuuca 1107585RNAHomo sapiens 75ccuuagcaga gcuguggagu gugacaaugg uguuuguguc uaaacuauca aacgccauua 60ucacacuaaa uagcuacugc uaggc 857686RNAHomo sapiens 76ugccagucuc uaggucccug agacccuuua accugugagg acauccaggg ucacagguga 60gguucuuggg agccuggcgu cuggcc 867788RNAHomo sapiens 77ugcgcuccuc ucagucccug agacccuaac uugugauguu uaccguuuaa auccacgggu 60uaggcucuug ggagcugcga gucgugcu 887889RNAHomo sapiens 78accagacuuu uccuaguccc ugagacccua acuugugagg uauuuuagua acaucacaag 60ucaggcucuu gggaccuagg cggagggga 897985RNAHomo sapiens 79cgcuggcgac gggacauuau uacuuuuggu acgcgcugug acacuucaaa cucguaccgu 60gaguaauaau gcgccgucca cggca 858089RNAHomo sapiens 80ugcugcuggc cagagcucuu uucacauugu gcuacugucu gcaccuguca cuagcagugc 60aauguuaaaa gggcauuggc cguguagug 898199RNAHomo sapiens 81cccuggcaug gugugguggg gcagcuggug uugugaauca ggccguugcc aaucagagaa 60cggcuacuuc acaacaccag ggccacacca cacuacagg 998284RNAHomo sapiens 82cguugcugca gcugguguug ugaaucaggc cgacgagcag cgcauccucu uacccggcua 60uuucacgaca ccaggguugc auca 8483100RNAHomo sapiens 83ugugucucuc ucuguguccu gccagugguu uuacccuaug guagguuacg ucaugcuguu 60cuaccacagg guagaaccac ggacaggaua ccggggcacc 1008495RNAHomo sapiens 84cggccggccc uggguccauc uuccaguaca guguuggaug gucuaauugu gaagcuccua 60acacugucug guaaagaugg cucccgggug gguuc 9585106RNAHomo sapiens 85gcgcagcgcc cugucuccca gccugaggug cagugcugca ucucugguca guugggaguc 60ugagaugaag cacuguagcu caggaagaga gaaguuguuc ugcagc 1068688RNAHomo sapiens 86caccuugucc ucacggucca guuuucccag gaaucccuua gaugcuaaga uggggauucc 60uggaaauacu guucuugagg ucaugguu 888799RNAHomo sapiens 87ccgaugugua uccucagcuu ugagaacuga auuccauggg uugugucagu gucagaccuc 60ugaaauucag uucuucagcu gggauaucuc ugucaucgu 998873RNAHomo sapiens 88ccuggcacug agaacugaau uccauaggcu gugagcucua gcaaugcccu guggacucag 60uucuggugcc cgg 738968RNAHomo sapiens 89gaggcaaagu ucugagacac uccgacucug aguaugauag aagucagugc acuacagaac 60uuugucuc 689089RNAHomo sapiens 90gccggcgccc gagcucuggc uccgugucuu cacucccgug cuuguccgag gagggaggga 60gggacggggg cugugcuggg gcagcugga 899184RNAHomo sapiens 91gucagaauaa ugucaaagug cuuacagugc agguagugau augugcaucu acugcaguga 60aggcacuugu agcauuaugg ugac 8492110RNAHomo sapiens 92ugaguuuuga gguugcuuca gugaacauuc aacgcugucg gugaguuugg aauuaaaauc 60aaaaccaucg accguugauu guacccuaug gcuaaccauc aucuacucca 11093110RNAHomo sapiens 93agaagggcua ucaggccagc cuucagagga cuccaaggaa cauucaacgc ugucggugag 60uuugggauuu gaaaaaacca cugaccguug acuguaccuu gggguccuua 1109482RNAHomo sapiens 94agggggcgag ggauuggaga gaaaggcagu uccugauggu ccccucccca ggggcuggcu 60uuccucuggu ccuucccucc ca 829592RNAHomo sapiens 95cggcuggaca gcgggcaacg gaaucccaaa agcagcuguu gucuccagag cauuccagcu 60gcgcuuggau uucguccccu gcucuccugc cu 929688RNAHomo sapiens 96cgaggauggg agcugagggc ugggucuuug cgggcgagau gagggugucg gaucaacugg 60ccuacaaagu cccaguucuc ggcccccg 889785RNAHomo sapiens 97augguguuau caaguguaac agcaacucca uguggacugu guaccaauuu ccaguggaga 60ugcuguuacu uuugaugguu accaa 859885RNAHomo sapiens 98ugguucccgc ccccuguaac agcaacucca uguggaagug cccacugguu ccaguggggc 60ugcuguuauc uggggcgagg gccag 859990RNAHomo sapiens 99ccgggccccu gugagcaucu uaccggacag ugcuggauuu cccagcuuga cucuaacacu 60gucugguaac gauguucaaa ggugacccgc 9010095RNAHomo sapiens 100ccagcucggg cagccguggc caucuuacug ggcagcauug gauggaguca ggucucuaau 60acugccuggu aaugaugacg gcggagcccu gcacg 9510168RNAHomo sapiens 101cccucgucuu acccagcagu guuugggugc gguugggagu cucuaauacu gccggguaau 60gauggagg 68102110RNAHomo sapiens 102cgccucagag ccgcccgccg uuccuuuuuc cuaugcauau acuucuuuga ggaucuggcc 60uaaagaggua uagggcaugg gaaaacgggg cggucggguc cuccccagcg 110103110RNAHomo sapiens 103aaagauccuc agacaaucca ugugcuucuc uuguccuuca uuccaccgga gucugucuca 60uacccaacca gauuucagug gagugaaguu caggaggcau ggagcugaca 11010486RNAHomo sapiens 104ugcuucccga ggccacaugc uucuuuauau ccccauaugg auuacuuugc uauggaaugu 60aaggaagugu gugguuucgg caagug 8610572RNAHomo sapiens 105ugucggguag cuuaucagac ugauguugac uguugaaucu cauggcaaca ccagucgaug 60ggcugucuga ca 72106110RNAHomo sapiens 106acccggcagu gccuccaggc gcagggcagc cccugcccac cgcacacugc gcugccccag 60acccacugug cgugugacag cggcugaucu gugccugggc agcgcgaccc 110107110RNAHomo sapiens 107ggccuggcug gacagaguug ucaugugucu gccugucuac acuugcugug cagaacaucc 60gcucaccugu acagcaggca cagacaggca gucacaugac aacccagccu 11010885RNAHomo sapiens 108ggcugagccg caguaguucu ucaguggcaa gcuuuauguc cugacccagc uaaagcugcc 60aguugaagaa cuguugcccu cugcc 85109110RNAHomo sapiens 109gcugcuggaa gguguaggua cccucaaugg cucaguagcc aguguagauc cugucuuucg 60uaaucagcag cuacaucugg cuacuggguc ucugauggca ucuucuagcu 110110110RNAHomo sapiens 110ccuggccucc ugcagugcca cgcuccgugu auuugacaag cugaguugga cacuccaugu 60gguagagugu caguuuguca aauaccccaa gugcggcaca ugcuuaccag 11011181RNAHomo sapiens 111gggcuuucaa gucacuagug guuccguuua guagaugauu gugcauuguu ucaaaauggu 60gcccuaguga cuacaaagcc c 8111264RNAHomo sapiens 112augacugauu ucuuuuggug uucagaguca auauaauuuu cuagcaccau cugaaaucgg 60uuau 6411388RNAHomo sapiens 113aucucuuaca caggcugacc gauuucuccu gguguucaga gucuguuuuu gucuagcacc 60auuugaaauc gguuaugaug uaggggga 8811471RNAHomo sapiens 114gcgacuguaa acauccucga cuggaagcug ugaagccaca gaugggcuuu cagucggaug 60uuugcagcug c 7111570RNAHomo sapiens 115guuguuguaa acauccccga cuggaagcug uaagacacag cuaagcuuuc agucagaugu 60uugcugcuac 7011692RNAHomo sapiens 116gggcagucuu ugcuacugua aacauccuug acuggaagcu guaagguguu cagaggagcu 60uucagucgga uguuuacagc ggcaggcugc ca 9211771RNAHomo sapiens 117ggagaggagg caagaugcug gcauagcugu ugaacuggga accugcuaug ccaacauauu 60gccaucuuuc c 7111899RNAHomo sapiens 118gaaacugggc ucaaggugag gggugcuauc ugugauugag ggacaugguu aauggaauug 60ucucacacag aaaucgcacc cgucaccuug gccuacuua 9911998RNAHomo sapiens 119acccaaaccc uaggucugcu gacuccuagu ccagggcucg ugauggcugg ugggcccuga 60acgagggguc uggaggccug gguuugaaua ucgacagc 98120110RNAHomo sapiens 120ggccagcugu gaguguuucu uuggcagugu cuuagcuggu uguugugagc aauaguaagg 60aagcaaucag caaguauacu gcccuagaag ugcugcacgu uguggggccc 11012177RNAHomo sapiens 121agucuaguua cuaggcagug uaguuagcug auugcuaaua guaccaauca cuaaccacac 60ggccagguaa aaagauu 7712272RNAHomo sapiens 122ggagcuuauc agaaucucca gggguacuuu auaauuucaa aaaguccccc aggugugauu 60cugauuugcu uc 7212367RNAHomo sapiens 123gugggccuca aauguggagc acuauucuga uguccaagug gaaagugcug cgacauuuga 60gcgucac 6712464RNAHomo sapiens 124ccccgcgacg agccccucgc acaaaccgga ccugagcguu uuguucguuc ggcucgcgug 60aggc 6412566RNAHomo sapiens 125agggcuccug acuccagguc cuguguguua ccuagaaaua gcacuggacu uggagucaga 60aggccu 6612696RNAHomo sapiens 126ucccuggcgu gaggguaugu gccuuuggac uacaucgugg aagccagcac caugcagucc 60augggcauau acacuugccu caaggccuau gucauc 9612784RNAHomo sapiens 127uugguacuug gagagugguu aucccugucc uguucguuuu gcucaugucg aaucguacag 60ggucauccac uuuuucagua ucaa 84128112RNAHomo sapiens 128ccaccccggu ccugcucccg ccccagcagc acacuguggu uuguacggca cuguggccac 60guccaaacca cacuguggug uuagagcgag ggugggggag gcaccgccga gg 11212994RNAHomo sapiens 129caugcugugu gugguacccu acugcagaca guggcaauca uguauaauua aaaaugauug 60guacgucugu ggguagagua cugcaugaca caug 9413091RNAHomo sapiens 130caugcugugu gugguacccu acugcagaca guggcaauca uguauaauua aaaaugauug 60guacgucugu ggguagagua cugcaugaca c 9113175RNAHomo sapiens 131gugguacccu acugcagacg uggcaaucau guauaauuaa aaaugauugg uacgucugug 60gguagaguac ugcau 7513290RNAHomo sapiens 132ucucaggcug ugaccuucuc gaggaaagaa gcacuuucug uugucugaaa gaaaagaaag 60ugcuuccuuu cagaggguua cgguuugaga 9013390RNAHomo sapiens 133ucucagguug ugaccuucuc gaggaaagaa gcacuuucug uugucugaaa gaaaagaaag 60ugcuuccuuu cagaggguua cgguuugaga 9013496RNAHomo sapiens 134gggaccugcg ugggugcggg cgugugagug ugugugugug aguguguguc gcuccggguc 60cacgcucaug cacacaccca cacgcccaca cucagg 9613598RNAHomo sapiens 135acgaauggcu augcacugca caacccuagg agagggugcc auucacauag acuauaauug 60aauggcgcca cuaggguugu gcagugcaca accuacac 98136110RNAHomo sapiens 136uuggauguug gccuaguucu guguggaaga cuagugauuu uguuguuuuu agauaacuaa 60aucgacaaca aaucacaguc ugccauaugg cacaggccau gccucuacag 110137110RNAHomo sapiens 137cuggauacag aguggaccgg cuggccccau cuggaagacu agugauuuug uuguugucuu 60acugcgcuca acaacaaauc ccagucuacc uaauggugcc agccaucgca 110138110RNAHomo sapiens 138agauuagagu

ggcugugguc uagugcugug uggaagacua gugauuuugu uguucugaug 60uacuacgaca acaagucaca gccggccuca uagcgcagac ucccuucgac 11013989RNAHomo sapiens 139cgggguuggu uguuaucuuu gguuaucuag cuguaugagu gguguggagu cuucauaaag 60cuagauaacc gaaaguaaaa auaacccca 8914087RNAHomo sapiens 140ggaagcgagu uguuaucuuu gguuaucuag cuguaugagu guauuggucu ucauaaagcu 60agauaaccga aaguaaaaac uccuuca 8714190RNAHomo sapiens 141ggaggcccgu uucucucuuu gguuaucuag cuguaugagu gccacagagc cgucauaaag 60cuagauaacc gaaaguagaa augauucuca 9014278RNAHomo sapiens 142cuuucuacac agguugggau cgguugcaau gcuguguuuc uguaugguau ugcacuuguc 60ccggccuguu gaguuugg 7814375RNAHomo sapiens 143ucaucccugg guggggauuu guugcauuac uuguguucua uauaaaguau ugcacuuguc 60ccggccugug gaaga 7514496RNAHomo sapiens 144cgggccccgg gcgggcggga gggacgggac gcggugcagu guuguuuuuu cccccgccaa 60uauugcacuc gucccggccu ccggcccccc cggccc 9614583RNAHomo sapiens 145agaaauaagg cuucugucua cuacuggaga cacugguagu auaaaaccca gagucuccag 60uaauggacgg gagccuuauu ucu 8314624RNAHomo sapiens 146agccugauua aacacaugcu cuga 2414723RNAHomo sapiens 147agcuacauug ucugcugggu uuc 2314823RNAHomo sapiens 148caaagugcug uucgugcagg uag 2314985RNAHomo sapiens 149uuuacaguuu gccaugauga aaugcauguu aaguccgugu uucagcugau cagccugauu 60aaacacaugc ucugagcaga cuaaa 85150110RNAHomo sapiens 150ugaacaucca ggucuggggc augaaccugg cauacaaugu agauuucugu guucguuagg 60caacagcuac auugucugcu ggguuucagg cuaccuggaa acauguucuc 11015180RNAHomo sapiens 151cugggggcuc caaagugcug uucgugcagg uagugugauu acccaaccua cugcugagcu 60agcacuuccc gagcccccgg 8015224RNAHomo sapiens 152uuuggcaaug guagaacuca cacu 2415322RNAHomo sapiens 153agagguagua gguugcauag uu 2215423RNAHomo sapiens 154aacauucauu gcugucggug ggu 2315522RNAHomo sapiens 155ucggauccgu cugagcuugg cu 2215622RNAHomo sapiens 156ugagguagga gguuguauag uu 2215722RNAHomo sapiens 157ugagguagua guuugugcug uu 2215823RNAHomo sapiens 158aaaagugcuu acagugcagg uag 2315920RNAHomo sapiens 159uaaggcacgc ggugaaugcc 2016027RNAHomo sapiens 160accuucuugu auaagcacug ugcuaaa 2716121RNAHomo sapiens 161ucacagugaa ccggucucuu u 2116222RNAHomo sapiens 162aagcccuuac cccaaaaagc au 2216322RNAHomo sapiens 163ucaaaacuga ggggcauuuu cu 2216421RNAHomo sapiens 164cauaaaguag aaagcacuac u 2116522RNAHomo sapiens 165ggugcagugc ugcaucucug gu 2216622RNAHomo sapiens 166ugcccugugg acucaguucu gg 2216721RNAHomo sapiens 167agggagggac gggggcugug c 2116822RNAHomo sapiens 168ucucccaacc cuuguaccag ug 2216921RNAHomo sapiens 169ucagugcaug acagaacuug g 2117023RNAHomo sapiens 170uuaaugcuaa ucgugauagg ggu 2317122RNAHomo sapiens 171uagcagcaca uaaugguuug ug 2217222RNAHomo sapiens 172uagcagcaca ucaugguuua ca 2217322RNAHomo sapiens 173aacauucaac cugucgguga gu 2217423RNAHomo sapiens 174aacauucauu guugucggug ggu 2317522RNAHomo sapiens 175uauggcacug guagaauuca cu 2217623RNAHomo sapiens 176uaaggugcau cuagugcaga uag 2317721RNAHomo sapiens 177cugaccuaug aauugacagc c 2117822RNAHomo sapiens 178aacuggcccu caaagucccg cu 2217921RNAHomo sapiens 179uagcagcaca gaaauauugg c 2118019RNAHomo sapiens 180accgugcaaa gguagcaua 1918122RNAHomo sapiens 181acaguagucu gcacauuggu ua 2218223RNAHomo sapiens 182cccaguguuc agacuaccug uuc 2318323RNAHomo sapiens 183cccaguguuu agacuaucug uuc 2318422RNAHomo sapiens 184gugaaauguu uaggaccacu ag 2218521RNAHomo sapiens 185gauuucagug gagugaaguu c 2118623RNAHomo sapiens 186uaaagugcuu auagugcagg uag 2318722RNAHomo sapiens 187agaauugugg cuggacaucu gu 2218822RNAHomo sapiens 188cauugcacuu gucucggucu ga 2218921RNAHomo sapiens 189uucacagugg cuaaguucug c 2119023RNAHomo sapiens 190uagcaccauu ugaaaucagu guu 2319122RNAHomo sapiens 191uagcaccauu ugaaaucggu ua 2219223RNAHomo sapiens 192uaagugcuuc cauguuuugg uga 2319323RNAHomo sapiens 193acuuaaacgu ggauguacuu gcu 2319423RNAHomo sapiens 194uaagugcuuc cauguuugag ugu 2319522RNAHomo sapiens 195cuuucagucg gauguuugca gc 2219623RNAHomo sapiens 196uguaaacauc cuacacucuc agc 2319721RNAHomo sapiens 197gccccugggc cuauccuaga a 2119821RNAHomo sapiens 198aggggugcua ucugugauug a 2119922RNAHomo sapiens 199aauugcacgg uauccaucug ua 2220023RNAHomo sapiens 200aagugccgcc aucuuuugag ugu 2320120RNAHomo sapiens 201acucaaacug ugggggcacu 2020221RNAHomo sapiens 202acuggacuug gagucagaag g 2120322RNAHomo sapiens 203acuggacuua gggucagaag gc 2220423RNAHomo sapiens 204aaugacacga ucacucccgu uga 2320522RNAHomo sapiens 205aaaccguuac cauuacugag uu 2220621RNAHomo sapiens 206gcaguccaug ggcauauaca c 2120722RNAHomo sapiens 207uccuguacug agcugccccg ag 2220821RNAHomo sapiens 208cagcagcaca cugugguuug u 2120923RNAHomo sapiens 209uuucaagcca gggggcguuu uuc 2321023RNAHomo sapiens 210cacucagccu ugagggcacu uuc 2321123RNAHomo sapiens 211uucucgagga aagaagcacu uuc 2321222RNAHomo sapiens 212aucuggaggu aagaagcacu uu 2221322RNAHomo sapiens 213aucgugcauc ccuuuagagu gu 2221422RNAHomo sapiens 214aucgugcauc cuuuuagagu gu 2221522RNAHomo sapiens 215gaaagcgcuu cccuuugcug ga 2221621RNAHomo sapiens 216aaagcgcuuc ccuucagagu g 2121722RNAHomo sapiens 217cucuagaggg aagcacuuuc uc 2221822RNAHomo sapiens 218aaagugcauc cuuuuagagu gu 2221922RNAHomo sapiens 219caaagugccu cccuuuagag ug 2222021RNAHomo sapiens 220cuccagaggg aaguacuuuc u 2122122RNAHomo sapiens 221aaagugcuuc cuuuuagagg gu 2222220RNAHomo sapiens 222cuacaaaggg aagcccuuuc 2022322RNAHomo sapiens 223cuacaaaggg aagcacuuuc uc 2222420RNAHomo sapiens 224cugcaaaggg aagcccuuuc 2022521RNAHomo sapiens 225gcgacccaua cuugguuuca g 2122621RNAHomo sapiens 226agggggaaag uucuauaguc c 2122723RNAHomo sapiens 227ugcaccaugg uugucugagc aug 2322821RNAHomo sapiens 228cgcgggugcu uacugacccu u 2122923RNAHomo sapiens 229ucuuugguua ucuagcugua uga 2323023RNAHomo sapiens 230cgggucggag uuagcucaag cgg 2323122RNAHomo sapiens 231aacccguaga uccgaucuug ug 2223222RNAHomo sapiens 232caagcucgcu ucuauggguc ug 2223323RNAHomo sapiens 233gaagugcuuc gauuuugggg ugu 2323422RNAHomo sapiens 234gauuagggug cuuagcuguu aa 2223521RNAHomo sapiens 235gguuuggucc uagccuuucu a 2123622RNAHomo sapiens 236uggacuugga gucaggaggc cu 2223722RNAHomo sapiens 237aaucugcagg gggagccugg gu 2223821RNAHomo sapiens 238acaugaaaag gggagagggc a 2123922RNAHomo sapiens 239acccccccca gccauacaua ga 2224025RNAHomo sapiens 240acuaccccag gaugccagca uaguu 2524122RNAHomo sapiens 241agcugguuug auggggagcc au 2224222RNAHomo sapiens 242agggugacag ggaacaguag au 2224322RNAHomo sapiens 243augugggugg uggucaccgu uu 2224422RNAHomo sapiens 244cacugauuau cgaggcgauu cu 2224520RNAHomo sapiens 245gaacccuacu ccugguacca 2024622RNAHomo sapiens 246gaauuuccug aggggagggg gc 2224722RNAHomo sapiens 247ggcaggacgg cguaggucuu ga 2224822RNAHomo sapiens 248gggcugggca gguuucagga au 2224921RNAHomo sapiens 249uaggucaagg uguagcccau a 2125022RNAHomo sapiens 250uauguacaag guggaggggg cg 2225122RNAHomo sapiens 251ucccccaccc uuagcuuaga ua 2225222RNAHomo sapiens 252uggagcaggc uggggcuuug ag 2225323RNAHomo sapiens 253ugugcuccgg aguuaccucg uuu 2325418RNAHomo sapiens 254uguggguucg aguuccau 1825518RNAHomo sapiens 255uucccggcca augcauua 1825622RNAHomo sapiens 256ugagguagua gguuguauag uu 2225722RNAHomo sapiens 257ugagguagua gguugugugg uu 2225822RNAHomo sapiens 258ugagguagua gauuguauag uu 2225922RNAHomo sapiens 259ugagguagua guuuguacag uu 2226021RNAHomo sapiens 260uaaagugcug acagugcaga u 2126122RNAHomo sapiens 261uuuccggcuc gcgugggugu gu 2226222RNAHomo sapiens 262cugaagcuca gagggcucug au 2226322RNAHomo sapiens 263aagcccuuac cccaaaaagu au 2226421RNAHomo sapiens 264cuuuuugcgg ucugggcuug c 2126522RNAHomo sapiens 265cagugcaaug augaaagggc au 2226622RNAHomo sapiens 266uaacagucua cagccauggu cg 2226722RNAHomo sapiens 267uuuggucccc uucaaccagc ug 2226822RNAHomo sapiens 268uuuggucccc uucaaccagc ua 2226922RNAHomo sapiens 269ugugacuggu ugaccagagg gg 2227022RNAHomo sapiens 270ucuacagugc acgugucucc ag 2227122RNAHomo sapiens 271cagugguuuu acccuauggu ag 2227222RNAHomo sapiens 272ggauuccugg aaauacuguu cu 2227322RNAHomo sapiens 273ucagugcauc acagaacuuu gu 2227421RNAHomo sapiens 274cuagacugaa gcuccuugag g 2127522RNAHomo sapiens 275uagguuaucc guguugccuu cg 2227622RNAHomo sapiens 276aaucauacac gguugaccua uu 2227722RNAHomo sapiens 277acugcaguga aggcacuugu ag 2227822RNAHomo sapiens 278accacugacc guugacugua cc 2227927RNAHomo sapiens 279auugaucauc gacacuucga acgcaau 2728022RNAHomo sapiens 280ucgugucuug uguugcagcc gg 2228121RNAHomo sapiens 281caucccuugc augguggagg g 2128222RNAHomo sapiens 282uagguaguuu cauguuguug gg 2228322RNAHomo sapiens 283cucccacugc uucacuugac ua 2228423RNAHomo sapiens 284ugugcaaauc caugcaaaac uga 2328523RNAHomo sapiens 285caaagugcuc auagugcagg uag 2328622RNAHomo sapiens 286uaaucucagc uggcaacugu ga 2228722RNAHomo sapiens 287aaaucucugc aggcaaaugu ga 2228823RNAHomo sapiens 288uacugcauca ggaacugauu gga 2328922RNAHomo sapiens 289aguucuucag uggcaagcuu ua 2229022RNAHomo sapiens 290accuggcaua caauguagau uu 2229122RNAHomo sapiens 291cucaguagcc aguguagauc cu 2229221RNAHomo sapiens 292aucacauugc cagggauuuc c 2129321RNAHomo sapiens 293aucacauugc cagggauuac c 2129422RNAHomo sapiens 294uggcucaguu cagcaggaac ag 2229522RNAHomo sapiens 295uucaaguaau ccaggauagg cu 2229621RNAHomo sapiens 296uucaaguaau ucaggauagg u 2129721RNAHomo sapiens 297uucacagugg cuaaguuccg c 2129822RNAHomo sapiens 298cacuagauug ugagcuccug ga 2229921RNAHomo sapiens 299agggcccccc cucaauccug u 2130022RNAHomo sapiens 300uaugugggau gguaaaccgc uu 2230122RNAHomo sapiens 301cugguuucac augguggcuu ag 2230223RNAHomo sapiens 302cagugcaaua guauugucaa agc 2330322RNAHomo sapiens 303uguaaacauc cuacacucag cu 2230422RNAHomo sapiens 304cuuucagucg gauguuuaca gc 2230522RNAHomo sapiens 305ugcuaugcca acauauugcc au 2230621RNAHomo sapiens 306cacauuacac ggucgaccuc u 2130723RNAHomo sapiens 307cgcauccccu agggcauugg ugu 2330822RNAHomo sapiens 308cuggcccucu cugcccuucc gu 2230922RNAHomo sapiens 309aacacaccug guuaaccucu uu 2231023RNAHomo sapiens 310gcaaagcaca cggccugcag aga 2331123RNAHomo sapiens 311ucaagagcaa uaacgaaaaa ugu 2331221RNAHomo sapiens 312gaacggcuuc auacaggagu u 2131322RNAHomo sapiens 313uccagcauca gugauuuugu ug 2231423RNAHomo sapiens 314ucccccaggu gugauucuga uuu 2331522RNAHomo sapiens 315aacacaccua uucaaggauu ca 2231624RNAHomo sapiens 316aauccuugga accuaggugu gagu 2431722RNAHomo sapiens 317agaucgaccg uguuauauuc gc 2231822RNAHomo sapiens 318gccugcuggg

guggaaccug gu 2231921RNAHomo sapiens 319aucauagagg aaaauccacg u 2132021RNAHomo sapiens 320aacauagagg aaauuccacg u 2132122RNAHomo sapiens 321agagguugcc cuuggugaau uc 2232221RNAHomo sapiens 322ugguagacua uggaacguag g 2132322RNAHomo sapiens 323uauacaaggg caagcucucu gu 2232422RNAHomo sapiens 324gaaguuguuc gugguggauu cg 2232522RNAHomo sapiens 325gaauguugcu cggugaaccc cu 2232623RNAHomo sapiens 326agguuacccg agcaacuuug cau 2332721RNAHomo sapiens 327aauauaacac agauggccug u 2132821RNAHomo sapiens 328uaguagaccg uauagcguac g 2132922RNAHomo sapiens 329aucgggaaug ucguguccgc cc 2233022RNAHomo sapiens 330caggucgucu ugcagggcuu cu 2233123RNAHomo sapiens 331ucuuggagua ggucauuggg ugg 2333222RNAHomo sapiens 332aucaugaugg gcuccucggu gu 2233321RNAHomo sapiens 333ucacuccucu ccucccgucu u 2133422RNAHomo sapiens 334aagacgggag gaaagaaggg ag 2233522RNAHomo sapiens 335gucauacacg gcucuccucu cu 2233622RNAHomo sapiens 336agaggcuggc cgugaugaau uc 2233722RNAHomo sapiens 337aaucauacag ggacauccag uu 2233822RNAHomo sapiens 338ugaaacauac acgggaaacc uc 2233922RNAHomo sapiens 339aaacaaacau ggugcacuuc uu 2234023RNAHomo sapiens 340uaauccuugc uaccugggug aga 2334122RNAHomo sapiens 341augcaccugg gcaaggauuc ug 2234222RNAHomo sapiens 342aaugcacccg ggcaaggauu cu 2234322RNAHomo sapiens 343aaugcaccug ggcaaggauu ca 2234423RNAHomo sapiens 344uagcagcggg aacaguucug cag 2334521RNAHomo sapiens 345uaaggcaccc uucugaguag a 2134622RNAHomo sapiens 346uacugcagac guggcaauca ug 2234718RNAHomo sapiens 347uucacaggga ggugucau 1834822RNAHomo sapiens 348ccucccacac ccaaggcuug ca 2234922RNAHomo sapiens 349caugccuuga guguaggacc gu 2235022RNAHomo sapiens 350ggagaaauua uccuuggugu gu 2235123RNAHomo sapiens 351ucggggauca ucaugucacg aga 2335222RNAHomo sapiens 352aaacauucgc ggugcacuuc uu 2235322RNAHomo sapiens 353uacgucaucg uugucaucgu ca 2235425RNAHomo sapiens 354gcugggcagg gcuucugagc uccuu 2535522RNAHomo sapiens 355uaugucugcu gaccaucacc uu 2235625RNAHomo sapiens 356ggcggaggga aguagguccg uuggu 2535722RNAHomo sapiens 357uacccauugc auaucggagu ug 2235820RNAHomo sapiens 358accaggaggc ugaggccccu 2035923RNAHomo sapiens 359aaggagcuua caaucuagcu ggg 2336021RNAHomo sapiens 360gcaggaacuu gugagucucc u 2136122RNAHomo sapiens 361cugcccuggc ccgagggacc ga 2236222RNAHomo sapiens 362ugcaacgaac cugagccacu ga 2236322RNAHomo sapiens 363cacccguaga accgaccuug cg 2236422RNAHomo sapiens 364aacuggggcg ggaaggggga ag 2236522RNAHomo sapiens 365aagugauugg aggugggugg gg 2236622RNAHomo sapiens 366agaagcugaa gggagagaga ca 2236722RNAHomo sapiens 367gugguuaucc cugcuguguu cg 2236822RNAHomo sapiens 368ugcagcuggu ggagucuggg gg 2236922RNAHomo sapiens 369aaucugcagg gggagccugg gu 2237024RNAHomo sapiens 370acucccaugu cccuugggaa gguc 2437123RNAHomo sapiens 371agcgagguug cccuuuguau auu 2337219RNAHomo sapiens 372agggcugggg acagagaug 1937319RNAHomo sapiens 373agugaagcau uggacugua 1937418RNAHomo sapiens 374aucccacucc ugacacca 1837521RNAHomo sapiens 375cauccuagcc cuaagucugg c 2137628RNAHomo sapiens 376cccaggcugg aguguagugg cgugaucu 2837722RNAHomo sapiens 377cgccugugaa uagucacugc ac 2237820RNAHomo sapiens 378gaaagcugag cgugaacgug 2037920RNAHomo sapiens 379gaaucccacu ucugacacca 2038023RNAHomo sapiens 380guuccuguug gccgagugga gac 2338119RNAHomo sapiens 381uaaaaggaac ucggcaaau 1938217RNAHomo sapiens 382ugcagaucuu gguggua 1738322RNAHomo sapiens 383uggggccucc cacagcuguu uc 2238417RNAHomo sapiens 384ugguggucua gugguua 1738525RNAHomo sapiens 385uguccaaagu aaacgcccug acgca 2538619RNAHomo sapiens 386ugucccuucg uggucgcca 1938720RNAHomo sapiens 387uucaugggga agcagauuug 2038823RNAHomo sapiens 388ugaggggcag agagcgagac uuu 2338924RNAHomo sapiens 389cauagcccgg ucgcugguac auga 2439023RNAHomo sapiens 390gccgagacua gagucacauc cug 23

Claims

1.-95. (canceled)

96. A method of identifying a tissue of origin of a cancer sample, said method comprising: (a) obtaining a biological sample from a subject in need thereof, wherein the sample is of a cancer selected from the group consisting of cancer of unknown primary (CUP), primary cancer, and metastatic cancer; (b) measuring the level of nucleic acids comprising SEQ ID NOS: 1, 2 or 156, 3-7, 9-12, 14-21, 23-27, 29-40, 42, 43, 44 or 191, 45-51, 53-56, 57 or 202, 58, 59, 60 or 208, 61, 62 or 211, 64-69, 146-148, and optionally at least one control nucleic acid in the biological sample and applying a classifier algorithm to said level of nucleic acids measured; and (c) identifying the tissue of origin of the sample based on the classification provided by the classifier algorithm.

97. The method of claim 96, wherein the classifier algorithm is selected from the group consisting of: decision tree classifier, K-nearest neighbor classifier (KNN), logistic regression classifier, nearest neighbor classifier, neural network classifier, Gaussian mixture model (GMM), Support Vector Machine (SVM) classifier, nearest centroid classifier, linear regression classifier and random forest classifier.

98. The method of claim 96, wherein the cancer is selected from the group consisting of adrenocortical carcinoma; anus or skin squamous cell carcinoma; biliary tract adenocarcinoma; Ewing sarcoma; gastrointestinal stromal tumor (GIST); gastrointestinal tract carcinoid; renal cell carcinoma: chromophobe, clear cell and papillary; pancreatic islet cell tumor; pheochromocytoma; urothelial cell carcinoma (TCC); lung, head & neck, or esophagus squamous cell carcinoma (SCC); brain: astrocytic tumor, oligodendroglioma; breast adenocarcinoma; uterine cervix squamous cell carcinoma; chondrosarcoma; germ cell cancer; sarcoma; colorectal adenocarcinoma; liposarcoma; hepatocellular carcinoma (HCC); lung large cell or adenocarcinoma; lung carcinoid; pleural mesothelioma; lung small cell carcinoma; B-cell lymphoma; T-cell lymphoma; melanoma; malignant fibrous histiocytoma (MFH) or fibrosarcoma; osteosarcoma; ovarian primitive germ cell tumor; ovarian carcinoma; pancreatic adenocarcinoma; prostate adenocarcinoma; rhabdomyosarcoma; gastric or esophageal adenocarcinoma; synovial sarcoma; non-seminomatous testicular germ cell tumor; seminomatous testicular germ cell tumor; thymoma; thymic carcinoma; follicular thyroid carcinoma; medullary thyroid carcinoma; and papillary thyroid carcinoma.

99. The method of claim 98, wherein a level of SEQ ID NOS: 55 above the reference threshold indicates a cancer of germ cell origin selected from the group consisting of an ovarian primitive cell and a testis cell, and further wherein a level of SEQ ID NOS: 29 and 62 above the reference threshold indicates a testis cell cancer origin selected from the group consisting of seminomatous testicular germ cell and non-seminomatous testicular germ cell.

100. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 9 and 29 above the reference threshold indicates a cancer origin selected from the group consisting of biliary tract adenocarcinoma and hepatocellular carcinoma.

101. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 156, 66 and 68 above the reference threshold indicates a cancer of brain origin, and further wherein a level of SEQ ID NOS: 40 and 60 above the reference threshold indicates a brain cancer origin selected from the group consisting of oligodendroglioma and astrocytoma.

102. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14 and 21 above the reference threshold indicates a cancer of prostate adenocarcinoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 27, 35, 14, 21, 32, 51, 7, 25, 50, 11, 148, 4, 49 and 67 above the reference threshold indicates a cancer of breast adenocarcinoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 27, 35, 14, 21, 32, 51, 7, 25, 4, 39, 50, 11, 148, 49, 67, 57 and 34 above the reference threshold indicates a cancer of an ovarian carcinoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 11, 148, 4, 49, 67, 57 and 34 above the reference threshold indicates a cancer of lung large cell or lung adenocarcinoma origin; and wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20 and 45 above the reference threshold indicates a cancer of lung small cell carcinoma origin.

103. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 11, 148 and 4 above the reference threshold indicates a cancer of thyroid carcinoma origin, and further wherein a level of SEQ ID NOS: 17 and 34 above the threshold indicates that the thyroid carcinoma origin is follicular or papillary.

104. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 50, 4, 39, 3 and 34 above the reference threshold indicates a cancer of a thymic carcinoma origin; or wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 14, 21, 32, 51, 7, 50, 4, 39, 3, 34, 69, 24 and 44 above the reference threshold indicates a cancer of urothelial cell carcinoma or squamous cell carcinoma origin, and further wherein a level of SEQ ID NOS: 1, 5 and 54 above the reference threshold indicates that the squamous-cell-carcinoma origin is uterine cervix squamous-cell--carcinoma or non-uterine cervix squamous cell carcinoma; or further wherein a level of SEQ ID NOS: 11 and 23 above the reference threshold indicates that the non-uterine cervix squamous cell carcinoma origin is selected from the group consisting of: a) anus or skin squamous cell carcinoma, and b) lung, head & neck, and esophagus squamous cell carcinoma.

105. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 47 and 50 above the reference threshold indicates a cancer of melanoma or lymphoma origin, and further wherein a level of SEQ ID NOS: 35 and 48 above the reference threshold indicates that the lymphoma cancer origin is selected from the group consisting of B-cell lymphoma and T-cell lymphoma.

106. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20, 45, 40, 67 and 68 above the reference threshold indicates a cancer of medullary thyroid carcinoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20, 45, 40, 67, 68, 64, 53 and 37 above the reference threshold indicates a cancer of lung carcinoid origin; and wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 20, 45, 40, 67, 68, 64, 53, 37, 34 and 18 above the reference threshold indicates a cancer of gastrointestinal tract carcinoid or pancreatic islet cell tumor origin.

107. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 42, 36 and 146 above the reference threshold indicates a cancer of gastric or esophageal adenocarcinoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 42, 36, 146, 20 and 43 above the reference threshold indicates a cancer of colorectal adenocarcinoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 56, 65, 25, 27, 35, 42, 36, 146, 20, 43, 51, 49 and 16 above the reference threshold indicates a cancer of pancreatic adenocarcinoma or biliary tract adenocarcinoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19 and 29 above the reference threshold indicates a cancer of renal cell carcinoma origin, and further wherein a level of SEQ ID NOS: 36 and 147 above the reference threshold indicates a chromophobe renal cell carcinoma origin, or further wherein a level of SEQ ID NOS: 49 and 9 above the reference threshold indicates that the renal cell carcinoma origin is clear cell or papillary.

108. The method of claim 98, wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65 and 56 above the reference threshold indicates a cancer of pheochromocytoma origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65, 56, 31, 38 and 61 above the reference threshold indicates a cancer of adrenocortical origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65, 56, 31, 38, 61, 14 and 45 above the reference threshold indicates a cancer of gastrointestinal stromal tumor origin; wherein a level of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 55, 6, 30, 46, 16, 2, 66, 68, 19, 29, 65, 56, 31, 38, 61, 14, 45, 35, 10 and 5 above the reference threshold indicates a cancer of pleural mesothelioma or sarcoma origin, and further wherein a level of SEQ ID NOS: 3, 40 and 15 above the reference threshold indicates that the sarcoma is synovial sarcoma, or further wherein a level of SEQ ID NOS: 3, 40, 15, 12 and 58 above the reference threshold indicates that the sarcoma is chondrosarcoma, or further wherein a level of SEQ ID NOS: 3, 40, 15, 12, 58, 36 and 26 above the reference threshold indicates that the sarcoma is liposarcoma, or further wherein a level of SEQ ID NOS: 3, 40, 15, 12, 58, 36, 26, 21, 25 and 49 above the reference threshold indicates that the sarcoma is Ewing sarcoma or osteosarcoma; or further wherein a level of SEQ ID NOS: 3, 40, 15, 12, 58, 36, 26, 21, 59, 39 and 33 above the reference threshold indicates that the sarcoma is selected from the group consisting of: a) rhabdomyosarcoma, and b) malignant fibrous histiocytoma and fibrosarcoma.

109. The method of claim 96, wherein the biological sample is selected from the group consisting of a bodily fluid, a cell line, a tissue sample, a biopsy sample, a needle biopsy sample, a fine needle biopsy (FNA) sample, a surgically removed sample, and a sample obtained by tissue-sampling procedures such as endoscopy, bronchoscopy, or laparoscopic methods.

110. The method of claim 109, wherein the tissue is a fresh, frozen, fixed, wax-embedded or formalin-fixed paraffin-embedded (FFPE) tissue.

111. The method of claim 96, wherein the level of the nucleic acid sequence is determined by a method selected from the group consisting of nucleic acid hybridization and nucleic acid amplification.

112. The method of claim 113, wherein nucleic acid hybridization is performed using a solid-phase nucleic acid biochip array or in situ hybridization and wherein nucleic acid amplification is real-time PCR comprising forward and reverse primers and a probe comprising a sequence selected from the group consisting of a sequence that is complementary to a sequence selected from SEQ ID NOS: 1, 2 or 156, 3-7, 9-12, 14-21, 23-27, 29-40, 42, 43, 44 or 191, 45-51, 53-56, 57 or 202, 58, 59, 60 or 208, 61, 62 or 211, 64-69, 146-148, and optionally at least one control nucleic acid and a fragment thereof.

113. A kit for performing the method of claim 96 comprising probes, wherein the probes comprise (i) DNA equivalents of nucleic acids comprising SEQ ID NOS: 1-7, 9-12, 14-21, 23-27, 29-40, 42-51, 53-57, 59-62, 64-69, 146-148, and 156, (ii) the complements thereof, or (iii) sequences at least 90% identical to (i) or (ii).