U.S. patent application number 15/853258 was filed with the patent office on 2018-05-10 for gene expression signature for classification of tissue of origin of tumor samples.
This patent application is currently assigned to Rosetta Genomics Ltd.. The applicant listed for this patent is Rosetta Genomics Ltd.. Invention is credited to Ranit Aharonov, Nitzan Rosenfeld, Shai Rosenwald.
Application Number | 20180127835 15/853258 |
Document ID | / |
Family ID | 45329188 |
Filed Date | 2018-05-10 |
United States Patent
Application |
20180127835 |
Kind Code |
A1 |
Aharonov; Ranit ; et
al. |
May 10, 2018 |
GENE EXPRESSION SIGNATURE 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.
Inventors: |
Aharonov; Ranit; (Tel Aviv,
IL) ; Rosenfeld; Nitzan; (Rehovot, IL) ;
Rosenwald; Shai; (Nes Ziona, IL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Rosetta Genomics Ltd. |
Rehovot |
|
IL |
|
|
Assignee: |
Rosetta Genomics Ltd.
Rehovot
IL
|
Family ID: |
45329188 |
Appl. No.: |
15/853258 |
Filed: |
December 22, 2017 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
14320113 |
Jun 30, 2014 |
|
|
|
15853258 |
|
|
|
|
13167489 |
Jun 23, 2011 |
8802599 |
|
|
14320113 |
|
|
|
|
PCT/IL09/01212 |
Dec 23, 2009 |
|
|
|
13167489 |
|
|
|
|
12532940 |
Sep 24, 2009 |
|
|
|
PCT/IL08/00396 |
Mar 20, 2008 |
|
|
|
13167489 |
|
|
|
|
61140642 |
Dec 24, 2008 |
|
|
|
61024565 |
Jan 30, 2008 |
|
|
|
60929244 |
Jun 19, 2007 |
|
|
|
60907266 |
Mar 27, 2007 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
C12Q 2600/158 20130101;
C12Q 1/6886 20130101; C12Q 2600/178 20130101 |
International
Class: |
C12Q 1/6886 20060101
C12Q001/6886 |
Claims
1. A method of producing thyroid medullary cancer cDNA sequences,
the method comprising: isolating RNA from a sample from a human
subject, wherein the sample comprises a tumor cell; contacting the
RNA with a polyadenylation agent under conditions that are
sufficient to form a polyadenylated RNA; reverse transcribing the
polyadenylated RNA in the presence of a universal poly(T) adapter
to produce cDNA sequences comprising a poly(T) tail; and amplifying
the cDNA sequences using a forward primer that is specific for SEQ
ID NO:42; thereby producing thyroid medullary cancer cDNA
sequences.
2. The method of claim 1, wherein the amplifying further comprises
amplifying the cDNA sequences with a reverse primer that is
complementary to the poly(T) tail.
3. The method of claim 1, wherein the sample is a biopsy.
4. The method of claim 1, wherein the sample is a fine-needle
aspiration.
5. The method of claim 1, wherein the amplifying comprises
quantitative PCR.
6. A reaction mixture for generating cDNA sequences derived from a
thyroid medullary cancer, the reaction mixture comprising: a
nucleic acid sample obtained from a biological sample from a human
subject, wherein the biological sample comprises a tumor cell; a
primer for generating cDNA sequences, wherein the primer is
specific for SEQ ID NO:42; and a detectable probe.
7. The reaction mixture of claim 6, wherein the nucleic acid sample
comprises cDNA sequences comprising a poly(T) tail, wherein the
cDNA sequences are produced from an RNA sample that has been
polyadenylated and reverse transcribed in the presence of a
universal poly(T) adapter.
8. The reaction mixture of claim 7, wherein the reaction mixture
further comprises a reverse primer that is complementary to the
poly(T) tail.
9. The reaction mixture of claim 6, wherein the biological sample
is a biopsy or a fine-needle aspiration.
10. A method of detecting gene expression in a sample from a human
subject, the method comprising: detecting the presence of SEQ ID
NO:42 in a nucleic acid sample from the subject, wherein the
detecting step comprises: contacting the nucleic acid sample with a
primer specific for SEQ ID NO:42; amplifying at least one nucleic
acid sequence in the sample; and detecting the presence of an
amplified nucleic acid sequence comprising SEQ ID NO:42; thereby
detecting the gene expression in the sample.
11. The method of claim 10, wherein the nucleic acid sample is an
RNA sample isolated from a biological sample from the subject,
wherein the biological sample comprises a tumor cell.
12. The method of claim 11, wherein the biological sample is a
biopsy.
13. The method of claim 11, wherein the biological sample is a
fine-needle aspiration.
14. The method of claim 11, wherein prior to the contacting step,
the method comprises contacting the RNA sample with a
polyadenylation agent under conditions that are sufficient to form
a polyadenylated RNA and reverse transcribing the polyadenylated
RNA in the presence of a universal poly(T) adapter to produce a
nucleic acid sample comprising cDNA sequences.
15. The method of claim 10, wherein the amplifying step comprises
quantitative PCR.
16. A method of identifying a subject as having a cancer of thyroid
medullary origin, the method comprising: obtaining a nucleic acid
sample from a human subject, wherein the nucleic acid sample is
from a biological sample comprising a tumor cell; contacting the
nucleic acid sample with a primer specific for SEQ ID NO:42;
amplifying the nucleic acid sequences in the biological sample; and
detecting the presence of an amplified nucleic acid sequence
comprising SEQ ID NO:42; thereby identifying the subject as having
a cancer of thyroid medullary origin.
17. The method of claim 16, wherein the nucleic acid sample is an
RNA sample, and wherein prior to the contacting step, the method
comprises contacting the RNA sample with a polyadenylation agent
under conditions that are sufficient to form a polyadenylated RNA
and reverse transcribing the polyadenylated RNA in the presence of
a universal poly(T) adapter to produce a nucleic acid sample
comprising cDNA sequences.
18. The method of claim 16, wherein the biological sample is a
biopsy.
19. The method of claim 16, wherein the biological sample is a
fine-needle aspiration.
20. The method of claim 16, wherein the amplifying step comprises
quantitative PCR.
21. The method of claim 16, wherein the detecting step comprises
measuring the relative abundance of an amplified nucleic acid
sequence comprising SEQ ID NO:42, relative to a reference value,
and identifying the subject as having a cancer of thyroid medullary
origin based on the relative abundance of the amplified nucleic
acid sequence comprising SEQ ID NO:42.
Description
FIELD OF THE INVENTION
[0001] The present invention relates to methods 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 further 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 903 paraffin-embedded samples
from 26 different tumor classes, corresponding to 18 distinct
tissues and organs, including primary and metastatic tumors.
microRNA microarray, of the samples as well as qRT-PCR data, were
used to construct a classifier, based on 48 tissue-specific
microRNAs, each linked to specific differential-diagnosis
roles.
[0010] The overall sensitivity of the independent blinded test in
identifying the tumor tissue of origin is 84%, with 97%
specificity. High confidence predictions reach 90% sensitivity with
99% specificity.
[0011] 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.
According to a first aspect, the present invention provides a
method of identifying a tissue of origin of a biological sample,
the method comprising: obtaining a biological sample from a
subject; determining an expression profile of individual nucleic
acids for a predetermined set of microRNAs; and classifying the
tissue of origin for said sample by a classifier. According to one
embodiment, said classifier is a decision tree model.
[0012] According to another aspect, the present invention provides
a method of classifying a tissue of origin of a biological sample,
the method comprising: obtaining a biological sample from a
subject; determining an expression profile in said sample of
nucleic acid sequences selected from the group consisting of SEQ ID
NOS: 1-49, or a sequence having at least about 80% identity
thereto; and comparing said expression profile to a reference
expression profile by using a classifier algorithm; whereby the
expression of any of said nucleic acid sequences or combinations
thereof allows the identification of the tissue of origin of said
sample.
[0013] According to one embodiment, said classifier algorithm is a
decision tree classifier, logistic regression classifier, linear
regression classifier, nearest neighbor classifier (including K
nearest neighbors), neural network classifier, Gaussian mixture
model (GMM) classifier and Support Vector Machine (SVM) classifier,
nearest centroid classifier, random forest classifier or any
boosting or bootstrap aggregating (bagging) of those
classifiers.
[0014] According to certain embodiments, said tissue is selected
from the group consisting of liver, lung, bladder, prostate,
breast, colon, ovary, testis, stomach, thyroid, pancreas, brain,
head and neck, kidney, melanocytes, thymus, biliary tract and
esophagus.
[0015] According to some embodiments said biological sample is a
cancerous sample.
[0016] According to another aspect, the present invention provides
a method of classifying 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: 1-49 or a sequence having at
least about 80% identity thereto; and comparing said obtained
measurement to reference values representing abundance of said
nucleic acid sequences by using a classifier algorithm; whereby the
relative abundance of said nucleic acid sequences allows the
classification of said cancer.
[0017] According to some embodiments, said reference values are
predetermined thresholds.
[0018] According to one embodiment, said sample is obtained from a
subject with a metastatic cancer. According to another embodiment,
said sample is obtained from a subject with cancer of unknown
primary (CUP). According to a further embodiment, said sample is
obtained from a subject with a primary cancer. According to still
another embodiment, said sample is a tumor of unidentified origin,
a metastatic tumor or a primary tumor.
[0019] According to certain embodiments, said cancer is selected
from the group consisting of liver cancer, biliary tract cancer,
lung cancer, bladder cancer, prostate cancer, breast cancer, colon
cancer, ovarian cancer, testicular cancer, stomach cancer, thyroid
cancer, pancreas cancer, brain cancer, head and neck cancer, kidney
cancer, melanoma, thymus cancer and esophagus cancer.
[0020] According to some embodiments, said lung cancer is selected
from the group consisting of lung carcinoid, lung small cell
carcinoma, lung adenocarcinoma, and lung squamous cell
carcinoma.
[0021] According to some embodiments, said brain cancer is selected
from the group consisting of brain astrocytoma and brain
oligodendroglioma.
[0022] According to some embodiments, said thyroid cancer is
selected from the group consisting of thyroid follicular, thyroid
papillary and thyroid medullary cancer.
[0023] According to some embodiments, said ovarian cancer is
selected from the group consisting of ovarian endometrioid and
ovarian serous cancer.
[0024] According to some embodiments, said testicular cancer is
selected from the group consisting of testicular non-seminoma and
testicular seminoma.
[0025] According to some embodiments, said esophagus cancer is
selected from the group consisting of esophagus adenocarcinoma and
esophagus squamous cell carcinoma.
[0026] According to some embodiments, said head and neck cancer is
selected from the group consisting of larynx carcinoma, pharynx
carcinoma and nose carcinoma.
[0027] According to some embodiments, said biliary tract cancer is
selected from the group consisting of cholangiocarcinoma and
gallbladder adenocarcinoma.
[0028] According to other embodiments, said 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
surgically removed sample, and a sample obtained by tissue-sampling
procedures. According to some embodiments the biological sample is
a fine needle aspiration (FNA) sample. According to some
embodiments, said tissue is a fresh, frozen, fixed, wax-embedded or
formalin-fixed paraffin-embedded (FFPE) tissue.
[0029] The classification method of the present invention comprises
the use of at least one classifier algorithm, said classifier
algorithm is selected from the group consisting of decision tree
classifier, logistic regression classifier, linear regression
classifier, nearest neighbor classifier (including K nearest
neighbors), neural network classifier, Gaussian mixture model (GMM)
classifier and Support Vector Machine (SVM) classifier, nearest
centroid classifier, random forest classifier or any boosting or
bootstrap aggregating (bagging) of those classifiers.
[0030] The classifier may use a decision tree structure (including
binary tree) or a voting (including weighted voting) scheme to
compare the classification of one or more classifier algorithms in
order to reach a unified or majority decision.
[0031] The invention further provides a method for classifying a
cancer of liver origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 6, 9, 25, 26, or a sequence having
at least about 80% identity thereto in a sample obtained from a
subject; wherein the abundance of said nucleic acid sequence is
indicative of a cancer of liver origin.
[0032] The invention further provides a method for classifying a
cancer of testicular origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 6, 26, 41, or a sequence having at
least about 80% identity thereto in a sample obtained from a
subject; wherein the abundance of said nucleic acid sequence is
indicative of a cancer of testicular origin.
[0033] The invention further provides a method for classifying a
cancer of testicular seminoma origin, the method comprising
measuring the relative abundance of a nucleic acid sequence
selected from the group consisting of SEQ ID NOS: 6, 26, 31, 41,
45, 48 or a sequence having at least about 80% identity thereto in
a sample obtained from a subject; wherein the abundance of said
nucleic acid sequence is indicative of a cancer of testicular
seminoma origin.
[0034] The invention further provides a method for classifying a
cancer of melanoma origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 6, 15, 17, 26, 41, 46, or a
sequence having at least about 80% identity thereto in a sample
obtained from a subject; wherein the abundance of said nucleic acid
sequence is indicative of a cancer of melanoma origin.
[0035] The invention further provides a method for classifying a
cancer of kidney origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 6, 7, 15, 17, 26, 41, 46, 47, or a
sequence having at least about 80% identity thereto in a sample
obtained from a subject; wherein the abundance of said nucleic acid
sequence is indicative of a cancer of kidney origin.
[0036] The invention further provides a method for classifying 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: 6, 7, 15, 17, 26, 41, 46, 47, or a
sequence having at least about 80% identity thereto in a sample
obtained from a subject; wherein the abundance of said nucleic acid
sequence is indicative of a cancer of brain origin.
[0037] The invention further provides a method for classifying a
cancer of brain astrocytoma origin, the method comprising measuring
the relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 6, 7, 10, 15, 17, 26, 41, 46, 47,
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 astrocytoma origin.
[0038] The invention further provides a method for classifying a
cancer of brain oligodendroglioma origin, the method comprising
measuring the relative abundance of a nucleic acid sequence
selected from the group consisting of SEQ ID NOS: 6, 7, 10, 15, 17,
26, 41, 46, 47, 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 oligodendroglioma
origin.
[0039] The invention further provides a method for classifying a
cancer of thyroid medullary origin, the method comprising measuring
the relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 6, 17-19, 24, 26, 32, 41, 42, or a
sequence having at least about 80% identity thereto in a sample
obtained from a subject; wherein the abundance of said nucleic acid
sequence is indicative of a cancer of thyroid medullary origin.
[0040] The invention further provides a method for classifying 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: 3, 6, 17-19, 24, 26, 32, 36, 41,
42, or a sequence having at least about 80% identity thereto in a
sample obtained from a subject; wherein the abundance of said
nucleic acid sequence is indicative of a cancer of lung carcinoid
origin.
[0041] The invention further provides a method for classifying 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: 3, 6, 17-19, 24,
26, 32, 36, 41, 42, or a sequence having at least about 80%
identity thereto in a sample obtained from a subject; wherein the
abundance of said nucleic acid sequence is indicative of a cancer
of lung small cell carcinoma origin.
[0042] The invention further provides a method for classifying a
cancer of colon origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 1, 3, 4, 6, 17-19, 21, 26, 29, 34,
37, 41, 42, 48, or a sequence having at least about 80% identity
thereto in a sample obtained from a subject; wherein the abundance
of said nucleic acid sequence is indicative of a cancer of colon
origin.
[0043] The invention further provides a method for classifying a
cancer of stomach origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 1, 3, 4, 6, 17-19, 21, 26, 29, 34,
37, 41, 42, 48, or a sequence having at least about 80% identity
thereto in a sample obtained from a subject; wherein the abundance
of said nucleic acid sequence is indicative of a cancer of stomach
origin.
[0044] The invention further provides a method for classifying a
cancer of pancreas origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 1, 3, 6, 17-19, 21, 26, 28, 29, 33,
37, 41, 42, or a sequence having at least about 80% identity
thereto in a sample obtained from a subject; wherein the abundance
of said nucleic acid sequence is indicative of a cancer of pancreas
origin.
[0045] The invention further provides a method for classifying a
cancer of biliary tract origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 1, 3, 6, 9, 17-19, 21, 25, 26, 28,
29, 33, 37, 41, 42, or a sequence having at least about 80%
identity thereto in a sample obtained from a subject; wherein the
abundance of said nucleic acid sequence is indicative of a cancer
of biliary tract origin.
[0046] The invention further provides a method for classifying a
cancer of prostate origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3, 6, 17-21, 26, 41, 42, or a
sequence having at least about 80% identity thereto in a sample
obtained from a subject; wherein the abundance of said nucleic acid
sequence is indicative of a cancer of prostate origin.
[0047] The invention further provides a method for classifying a
cancer of ovarian origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3, 5, 6, 11, 17-21, 26, 30, 41, 42,
or a sequence having at least about 80% identity thereto in a
sample obtained from a subject; wherein the abundance of said
nucleic acid sequence is indicative of a cancer of ovarian
origin.
[0048] The invention further provides a method for classifying a
cancer of ovarian endometrioid origin, the method comprising
measuring the relative abundance of a nucleic acid sequence
selected from the group consisting of SEQ ID NOS: 2, 3, 5, 6, 11,
17-22, 26, 30, 41, 42, or a sequence having at least about 80%
identity thereto in a sample obtained from a subject; wherein the
abundance of said nucleic acid sequence is indicative of a cancer
of ovarian endometrioid origin.
[0049] The invention further provides a method for classifying a
cancer of ovarian serous origin, the method comprising measuring
the relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 2, 3, 5, 6, 11, 17-22, 26, 30, 41,
42, or a sequence having at least about 80% identity thereto in a
sample obtained from a subject; wherein the abundance of said
nucleic acid sequence is indicative of a cancer of ovarian serous
origin.
[0050] The invention further provides a method for classifying a
cancer of breast origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3, 5, 6, 11, 17-22, 26, 30, 39, 41,
42, or a sequence having at least about 80% identity thereto in a
sample obtained from a subject; wherein the abundance of said
nucleic acid sequence is indicative of a cancer of breast
origin.
[0051] The invention further provides a method for classifying a
cancer of lung adenocarcinoma origin, the method comprising
measuring the relative abundance of a nucleic acid sequence
selected from the group consisting of SEQ ID NOS: 3, 5, 6, 8, 11,
16-22, 26, 27, 30, 37, 39, 41, 42, or a sequence having at least
about 80% identity thereto in a sample obtained from a subject;
wherein the abundance of said nucleic acid sequence is indicative
of a cancer of lung adenocarcinoma origin.
[0052] The invention further provides a method for classifying a
cancer of papillary thyroid origin, the method comprising measuring
the relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3, 5, 6, 8, 11, 16-22, 26, 27, 29,
30, 37-39, 41, 42, or a sequence having at least about 80% identity
thereto in a sample obtained from a subject; wherein the abundance
of said nucleic acid sequence is indicative of a cancer of
papillary thyroid origin.
[0053] The invention further provides a method for classifying a
cancer of follicular thyroid origin, the method comprising
measuring the relative abundance of a nucleic acid sequence
selected from the group consisting of SEQ ID NOS: 3, 5, 6, 8, 11,
16-22, 26, 27, 29, 30, 37-39, 41, 42, 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 follicular
thyroid origin.
[0054] The invention further provides a method for classifying a
cancer of thymus origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3, 5, 6, 11, 16-22, 26, 27, 29, 30,
35, 39, 41, 42, or a sequence having at least about 80% identity
thereto in a sample obtained from a subject; wherein the abundance
of said nucleic acid sequence is indicative of a cancer of thymus
origin.
[0055] The invention further provides a method for classifying a
cancer of bladder origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3-6, 11, 16-22, 26, 27, 29, 30, 35,
39, 41, 42, 44, or a sequence having at least about 80% identity
thereto in a sample obtained from a subject; wherein the abundance
of said nucleic acid sequence is indicative of a cancer of bladder
origin.
[0056] The invention further provides a method for classifying a
cancer of lung squamous origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3-6, 11, 16-23, 26, 27, 29, 30, 32,
35, 39, 41, 42, 44, or a sequence having at least about 80%
identity thereto in a sample obtained from a subject; wherein the
abundance of said nucleic acid sequence is indicative of a cancer
of lung squamous origin.
[0057] The invention further provides a method for classifying a
cancer of head and neck origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3-6, 11, 14, 16-23, 26, 27, 29, 30,
32, 35, 37, 39, 41, 42, 44, 45, or a sequence having at least about
80% identity thereto in a sample obtained from a subject; wherein
the abundance of said nucleic acid sequence is indicative of a
cancer of head and neck origin.
[0058] The invention further provides a method for classifying a
cancer of esophagus origin, the method comprising measuring the
relative abundance of a nucleic acid sequence selected from the
group consisting of SEQ ID NOS: 3-6, 11, 14, 16-23, 26, 27, 29, 30,
32, 35, 37, 39, 41, 42, 44, 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 esophagus
origin.
[0059] According to some embodiments the nucleic acid sequence
expression profile or 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.
[0060] According to some embodiments the nucleic acid amplification
method is real-time PCR. The real-time PCR method may comprise
forward and reverse primers. According to some embodiments the
forward primer comprises a sequence selected from the group
consisting of SEQ ID NOS: 50-98 and 150. According to some
embodiments the reverse primer comprises SEQ ID NO: 288.
[0061] According to additional embodiments the real-time PCR method
further comprises a probe. According to some 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: 1-49; a fragment thereof and a sequence having at least about
80% identity thereto. According to additional embodiments the probe
comprises a sequence selected from the group consisting of SEQ ID
NOS: 99-149 and 151.
[0062] According to another aspect, the present invention provides
a kit for cancer classification, said 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: SEQ ID NOS: 1-49; a fragment thereof and a sequence having at
least about 80% identity thereto.
[0063] According to additional embodiments the probe comprises a
sequence selected from the group consisting of SEQ ID NOS: 99-149
and 151.
[0064] According to certain embodiments, said cancer is selected
from the group consisting of liver cancer, biliary tract cancer,
lung cancer, bladder cancer, prostate cancer, breast cancer, colon
cancer, ovarian cancer, testicular cancer, stomach cancer, thyroid
cancer, pancreas cancer, brain cancer, head and neck cancer, kidney
cancer, melanoma, thymus cancer and esophagus cancer.
[0065] 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
[0066] FIGS. 1A-1C demonstrate the structure of the binary
decision-tree classifier, with 26 nodes (numbered, Table 3) and 27
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 left branch at node #1 continues
to node #2, otherwise it continues to node #3. A sample that
reaches node #2, is further classified to either the left branch at
node #2, and is assigned to the "liver" class, or to the right
branch at node #2, and is assigned to the "biliary tract carcinoma"
class.
[0067] 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.
[0068] Developing a different classifier for e.g. male and female
cases or for different tumor sites would inefficiently exploit
measured data and would require unwieldy numbers of samples.
Instead, exceptions were noted for several special cases: For
samples from female patients, testis or prostate origins were
excluded from the KNN database, and the right branch was
automatically taken in node 3 and node 16 in the decision-tree. For
samples from male patients, ovary origin was excluded and the right
branch taken at node 17. For samples that were indicated as
metastases to the liver, liver origin (hepatocellular carcinoma and
biliary tract carcinomas from within the liver) was excluded and
the right branch taken at node 1. For samples indicated as brain
metastases, brain origin was excluded and the right branch taken at
node 7. Additional information is thus incorporated into the
classification decision without loss of generality or need to
retrain the classifier.
[0069] FIG. 2 demonstrates 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 ("leaves") of
this node are used for training. Tumors originating from tissues at
the left branch at node #1, including tumors from the "liver" class
and the "biliary tract" class (liver-cholangio; diamonds) are
easily separated from tumors of non-liver and non-biliary tract
origins (right branch at node #2; gray squares) using the
expression levels of hsa-miR-200c (SEQ ID NO: 26) and hsa-miR-122
(SEQ ID NO: 6) (with one outlier), with a linear classifier (the
diagonal line).
[0070] FIG. 3 demonstrates binary decisions at node #5 of the
decision-tree. Tumors of epithelial origin (left branch at node #5,
marked by diamonds) are easily separated from tumors of
non-epithelial origin (right branch at node #5, marked by squares)
using the expression levels of hsa-miR-200c (SEQ ID NO: 26) and
hsa-miR-148b (SEQ ID NO: 17). The gray area (with higher levels of
hsa-miR-200c) marks the region classified as epithelial (left
branch) at this node.
[0071] FIG. 4 demonstrates binary decisions at node #7 of the
decision-tree. Tumors originating in the brain (diamonds) are
easily separated from tumors of kidney origin (squares) using the
expression levels of hsa-miR-124 (SEQ ID NO: 7) and hsa-miR-9* (SEQ
ID NO: 47).
[0072] FIG. 5 demonstrates binary decisions at node #10 of the
decision-tree. Neuroendocrine tumors originating in the lung
(diamonds) are easily separated from tumors of thyroid-medullary
origin (squares) using the expression levels of hsa-miR-200a (SEQ
ID NO: 24) and hsa-miR-222 (SEQ ID NO: 32).
[0073] FIG. 6 demonstrates binary decisions at node #12 of the
decision-tree. Tumors originating in the gastrointestinal tract
(left branch at node #12, marked by diamonds) are easily separated
from tumors of non digestive origins (right branch at node #12,
marked by squares) using the expression levels of hsa-miR-106a (SEQ
ID NO: 3) and hsa-miR-192 (SEQ ID NO: 21).
[0074] FIG. 7 demonstrates binary decisions at node #16 of the
decision-tree. Tumors originating in the prostate (left branch at
node #16, marked by diamonds) are easily separated from tumors of
other origins (right branch at node #16, marked by squares) using
the expression levels of hsa-miR-185 (SEQ ID NO: 20) and
hsa-miR-375 (SEQ ID NO: 42).
[0075] FIGS. 8A-8B demonstrate classification example. FIG. 8A
shows that the measured levels (normalized C.sub.t, inversely
proportional to log(abundance)) of hsa-miR-200c (SEQ ID NO: 26) and
hsa-miR-122 (SEQ ID NO: 6) are compared for all training set
samples, indicating the left and right branches of node #1 (circles
and stars respectively). One metastatic tumor excised from the
brain (square), from a patient that had a concomitant tumor in the
lung, and was therefore originally diagnosed as a lung cancer.
However, this sample showed an uncharacteristic high expression of
hsa-miR-122, a strong hepatic marker, and was consequently
classified as possibly originating from the liver by the microRNA
classifier. FIG. 8B shows that upon re-examination of the
metastatic brain tumor by immunohistochemistry (blinded to the
results of the microRNA classifier), this tumor was indeed found to
be negative for lung specific markers: the sample was negative for
immunohistochemical staining by both CK7 and TTF1, as well as CK20,
CEA, CA125, s-100, thyroglobulin, chromogranin, synaptophysin,
CD56, GFAP, calcitonin, and anterior pituitary hormones, while
staining positive for CAM5.5' and AE1/AE3. This staining pattern
was compatible with hepatocellular carcinoma, prompting further
staining for HEPA1 and alpha fetoprotein. The tumor stained
positive for both stains, consistent with a diagnosis of
hepatocellular carcinoma (FIG. 8B). H&E staining (upper panel)
showed that the metastasis is composed of sheets of cells with
abundant eosinophilic cytoplasm and round to oval nuclei. Among
many immunostains used to evaluate the origin of the tumor, HEPA-1
showed strong and specific immunopositivity (lower panel).
DETAILED DESCRIPTION OF THE INVENTION
[0076] 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 tissues and tumor
origins. A new microRNA-based classifier was developed for
determining tissue origin of tumors based on a surprisingly small
number of 48 microRNAs markers. The classifier uses a specific
algorithm and allows a clear interpretation of the specific
biomarkers. High confidence predictions reach 90% sensitivity and
99% specificity.
[0077] 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 cancer. The performance of the classifier using a
small number of markers highlights the utility of microRNA as
tissue-specific cancer biomarkers, and provides an effective means
for facilitating diagnosis of CUP and more generally of identifying
tumor origins of metastases.
[0078] The possibility to distinguish between different tumor
origins facilitates providing the patient with the best and most
suitable treatment.
[0079] 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 present invention provides
methods for diagnosing the presence of a specific cancer by
analyzing the levels of said microRNA molecules in biopsies, tumor
samples, cells, tissues or bodily fluids.
[0080] In the present invention, determining the levels of said
microRNA in biopsies, tumor samples, cells, tissues or bodily
fluid, is particularly useful for discriminating between different
cancers.
[0081] All the methods of the present invention may optionally
further include measuring levels of other cancer markers. Other
cancer markers, in addition to said microRNA molecules, useful in
the present invention will depend on the cancer being tested and
are known to those of skill in the art.
[0082] Assay techniques that can be used to determine levels of
gene expression, such as the nucleic acid sequence of the present
invention, in a sample derived from a patient are well known to
those of skill in the art. Such assay methods include, but are not
limited to, reverse transcriptase PCR (RT-PCR) assays, nucleic acid
microarrays and biochip analysis, immunohistochemistry assays, in
situ hybridization assays, competitive-binding assays, northern
blot analyses and ELISA assays.
[0083] According to one embodiment, the assay is based on
expression level of 48 microRNAs in RNA extracted from FFPE
metastatic tumor tissue. The test is a quantitative real time
reverse transcriptase polymerase chain reaction (qRT-PCR) test. RNA
is first polyadenylated and then reverse transcribed using
universal poly(T) adapter to create cDNA. The cDNA is amplified
using specific forward primer and universal reverse primer (with a
sequence complementary to the 5' tail of the poly(T) adapter), and
detected by specific MGB probes (see specific sequences in Table
1).
[0084] The expression levels are used to infer the sample origin
using analysis techniques such as but not limit to decision tree
classifier, logistic regression classifier, linear regression
classifier, nearest neighbor classifier (including K nearest
neighbors), neural network classifier and nearest centroid
classifier.
[0085] 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 the training set). At each
node, the expressions of one or several microRNAs are combined
together using a simple function of the form P=exp
(b0+b1*mir1+b2*mir2+b3*mir3 . . . ), where the values of b0, b1, b2
. . . and the identities of the microRNAs have been pre-determined
(using the training set). The resulting P is compared to a
threshold level PTH (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 PTH for that node.
This continues until an end-point ("leaf") of the tree is
reached.
[0086] Training the tree algorithm means determining: the tree
structure (which nodes there are and what is on each side), which
miRs are used in each node and the values of b0, b1, b2 . . . and
PTH. These were determined by a combination of machine learning,
optimization algorithm, and trial and error by experts in machine
learning and diagnostic algorithms.
[0087] In some embodiments of the invention, correlations and/or
hierarchical clustering can be used to assess the similarity of the
expression level of the nucleic acid sequences of the invention
between a specific sample and different exemplars of cancer
samples. An arbitrary threshold on the expression level of one or
more nucleic acid sequences can be set for assigning a sample or
cancer 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 for assignment is
treated as a parameter, which can be used to quantify the
confidence with which samples are assigned to each class. The
threshold for assignment 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.
[0088] In another preferred embodiment, expression level of the
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. Each such
pairwise 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.
Definitions
[0089] 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.
[0090] 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.
[0091] About
[0092] As used herein, the term "about" refers to +/-10%.
[0093] Attached
[0094] "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.
[0095] Baseline
[0096] "Baseline", as used herein, means the initial cycles of PCR,
in which there is little change in fluorescence signal.
[0097] Biological Sample
[0098] "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). 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.
[0099] Cancer
[0100] 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.
[0101] Classification
[0102] 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" is a decision tree that places categorical
variables into classes.
[0103] Complement
[0104] "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').
[0105] Ct
[0106] 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.
[0107] 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 expression levels of the microRNA.
[0108] Data Processing Routine
[0109] 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 make determination of 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.
[0110] Data Set
[0111] 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.
[0112] Data Structure
[0113] As used herein, the term "data structure" refers to a
combination of two or more data sets, applying one or more
mathematical manipulations to one or more data sets to obtain one
or more new data sets, or manipulating 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.
[0114] Detection
[0115] "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.
[0116] Differential Expression
[0117] "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.
[0118] Expression Profile
[0119] 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.
[0120] Expression Ratio
[0121] "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.
[0122] FDR
[0123] When performing multiple statistical tests, for example in
comparing the signal between 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.
[0124] Fragment
[0125] "Fragment" is used herein to indicate a non-full-length part
of a nucleic acid. Thus, a fragment is itself also a nucleic
acid.
[0126] Gene
[0127] "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.
[0128] Groove Binder/Minor Groove Binder (MGB)
[0129] "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.
[0130] Host Cell
[0131] "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.
[0132] Identity
[0133] "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.
[0134] In Situ Detection
[0135] "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.
[0136] k-Nearest Neighbor
[0137] The phrase "k-nearest neighbor" refers to a classification
method that classifies a point by calculating the distances between
the point 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).
[0138] Label
[0139] "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.
[0140] Logistic Regression
[0141] 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.
[0142] 1D/2D Threshold Classifier
[0143] "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.
[0144] Metastasis
[0145] "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.
[0146] Node
[0147] 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. A "leaf" is a node not further split, the
terminal grouping in a classification or decision tree.
[0148] Nucleic Acid
[0149] "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.
[0150] 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.
[0151] 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; O-
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.
[0152] Probe
[0153] "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.
[0154] Reference Value
[0155] 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 determine the detection of
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.
[0156] In preferred embodiments the reference value is determined
from statistical analysis of studies that compare microRNA
expression with known clinical outcomes.
[0157] Sensitivity
[0158] "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 type out of two possible
types. 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.
[0159] Specificity
[0160] "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 type out of two possible
types. The sensitivity 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.
[0161] Stringent Hybridization Conditions
[0162] "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.
[0163] Substantially Complementary
[0164] "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.
[0165] Substantially Identical
[0166] "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.
[0167] Subject
[0168] 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.
[0169] Target Nucleic Acid
[0170] "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.
[0171] 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. No. 11/384,049,
11/418,870 or 11/429,720, the contents of which are incorporated
herein.
[0172] Threshold
[0173] As used herein, the term "threshold" means the numerical
value assigned for each run, which reflects a statistically
significant point above the calculated PCR baseline.
[0174] Tissue Sample
[0175] 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.
[0176] Tumor
[0177] "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.
[0178] Variant
[0179] "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.
[0180] Wild Type
[0181] 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.
[0182] The present invention employs miRNAs for the identification,
classification and diagnosis of specific cancers and the
identification of their tissues of origin.
[0183] 1. microRNA Processing
[0184] 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.
[0185] 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.
[0186] 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.
[0187] 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.
[0188] 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.
[0189] 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).
[0190] 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 GenesDev 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).
[0191] 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.
[0192] 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.
[0193] 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.
[0194] 2. Nucleic Acids
[0195] Nucleic acids are provided herein. The nucleic acids
comprise the sequences of SEQ ID NOS: 1-288 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.
[0196] 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.
TABLE-US-00001 TABLE 1 SEQ ID NOS of miRs, forward primers and MGB
probes miR SEQ FW primer MGB probe Sanger miR name ID NO: SEQ ID
NO: SEQ ID NO: hsa-let-7b 1 50 99 hsa-let-7f 2 51 100 hsa-miR-106a
3 52 101 hsa-miR-10a 4 53 102 hsa-miR-10b 5 54 103 hsa-miR-122 6 55
104 hsa-miR-124 7 56 105 hsa-miR-125b 8 57 106 hsa-miR-126 9 58 107
hsa-miR-128 10 59 108 hsa-miR-130a 11 60 109 hsa-miR-138 12 61 110
hsa-miR-142-3p 13 62 111 hsa-miR-143 14 63 112 hsa-miR-146a 15 64
113 hsa-miR-146b-5p 16 65 114 hsa-miR-148b 17 66 115 hsa-miR-152 18
67 116 hsa-miR-15b 19 68 117 hsa-miR-185 20 69 118 hsa-miR-192 21
70 119, 120 hsa-miR-193a-3p 22 71 121 hsa-miR-19b 23 72 122
hsa-miR-200a 24 73 123 hsa-miR-200b 25 74 124 hsa-miR-200c 26 75
125, 126 hsa-miR-205 27 76 127 hsa-miR-20a 28 77 128 hsa-miR-21 29
78 129 hsa-miR-210 30 79 130 hsa-miR-221 31 80 131 hsa-miR-222 32
81 132 hsa-miR-25 33 82 133 hsa-miR-29a 34 83 134 hsa-miR-29b 35 84
135 hsa-miR-29c 36 85 136 hsa-miR-30a 37 86 137 hsa-miR-31 38 87
138 hsa-miR-342-3p 39 88 139 hsa-miR-345 40 89 140 hsa-miR-372 41
90 141 hsa-miR-375 42 91 142 hsa-miR-378 43 92 143 hsa-miR-425 44
93 144 hsa-miR-451 45 94 145 hsa-miR-497 46 95 146 hsa-miR-9* 47 96
147 hsa-mir-92b 48 97 148 hsa-miR-509-3p 49 150 151 U6 98 149
Sanger miR name: the miRBase registry name (release 9-12)
[0197] 3. Nucleic Acid Complexes
[0198] 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.
[0199] 4. Pri-miRNA
[0200] 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: 1-49 or variants thereof.
[0201] 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.
[0202] 5. Pre-miRNA
[0203] 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:
1-49 or variants thereof.
[0204] 6. miRNA
[0205] 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: 1-49 or variants thereof.
[0206] 7. Probes
[0207] 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.
[0208] 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: 1-49 or variants thereof. The probe may comprise a
sequence selected from the group consisting of SEQ ID NOS: 99-149
and 151.
[0209] 8. Biochip
[0210] 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.
[0211] 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.
[0212] 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.
[0213] 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.
[0214] 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.
[0215] 9. Diagnostics
[0216] As used herein, the term "diagnosing" refers to classifying
pathology, or a symptom, determining a severity of the pathology
(grade or stage), monitoring pathology progression, forecasting an
outcome of pathology and/or prospects of recovery.
[0217] 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.
[0218] Analyzing presence of malignant or pre-malignant cells can
be effected in vivo or ex vivo, whereby a biological sample (e.g.,
biopsy) 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.
[0219] 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.
[0220] As used herein the phrase "treatment regimen" refers to a
treatment plan that specifies the type of treatment, dosage,
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.
[0221] 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.
[0222] 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 sequence which indicate the diagnosis may differ from
those which indicate the prognosis and molecular profiling of the
condition of the cells may lead to distinctions between responsive
or refractory conditions or may be predictive of outcomes.
[0223] 10. Kits
[0224] 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.
[0225] 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.
[0226] 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.
[0227] 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
[0228] 903 tumor samples took part in the study. These included 252
that were part of a preliminary study and 651 additional
formalin-fixed paraffin-embedded (FFPE) samples. 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. 204 of the 903 samples
were used only in the validation phase, as an independent blinded
test set. The reference diagnosis of these samples from the
original clinical record was confirmed by an additional review of
pathological specimens.
2. RNA Extraction
[0229] 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
[0230] Custom microarrays (Agilent Technologies, Santa Clara,
Calif.) were produced by printing DNA oligonucleotide probes to
more than 900 human microRNAs. Each probe, printed in triplicate,
carried up to 22-nucleotide (nt) linker at the 3' end of the
microRNA's complement sequence, in addition to an amine group used
to couple the probes to coated glass slides. Twenty M of each probe
were dissolved in 2.times.SSC+0.0035% SDS and spotted in triplicate
on Schott Nexterion.RTM. Slide E-coated microarray slides using a
Genomic Solutions.RTM. BioRobotics MicroGrid II according the
MicroGrid manufacturer's directions. Fifty-four negative control
probes were designed using the sense sequences of different
microRNAs. 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, U49,
U24, Z30, U6, U48, U44), 5.8 s and 5 s ribosomal RNA) are spotted
on the array to verify RNA quality. The slides were blocked in a
solution containing 50 mM ethanolamine, 1 M Tris (pH9.0) and 0.1%
SDS for 20 min at 50.degree. C., then thoroughly rinsed with water
and spun dry.
4. Cy-Dye Labeling of miRNA for miR Array
[0231] Five 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
(Dharmacon), to the 3' end with Cy3 or Cy5. The labeling reaction
contained total RNA, spikes (0.1-20 fmoles), 300 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. The labeled RNA was mixed with 3.times. hybridization
buffer (Ambion), heated to 95.degree. C. for 3 min and then added
on top of the miR array. Slides were hybridized for 12-16 h at
42.degree. C., followed by two washes at room temperature with
1.times.SSC and 0.2% SDS and a final wash with 0.1.times.SSC.
[0232] Arrays were scanned using an Agilent Microarray Scanner
Bundle G2565BA (resolution of 10 .mu.m at 100% power). Array images
were analyzed using SpotReader software (Niles Scientific).
5. Array Signal Calculation and Normalization
[0233] Triplicate spots were combined to produce one signal for
each probe by taking the logarithmic mean of reliable spots. All
data were log-transformed (natural base) and the analysis was
performed in log-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). Data were translated back to
linear-space (by taking the exponent). Using only the training set
samples to generate the reference data vector did not affect the
results.
6. Logistic Regression
[0234] 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##
[0235] where .beta..sub.0 is the bias, M.sub.i is the expression
level (normalized, in log-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 = e .beta. 0 + .beta. 1 M 1 + e .beta. 0 + .beta. 1 M .
##EQU00002##
[0236] 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##
[0237] 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>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 3). 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.
7. Stepwise Logistic Regression and Feature Selection
[0238] 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 feature was inserted into the
algorithm, but in most cases 2-3 features were selected.
[0239] 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
about two-thirds of the samples at least once, and any one sample
had >99% chance of being left out at least once. This resulted
in an average of 2-3 features per node (4-8 in more difficult
nodes). A robust set of 2-3 features per each node was selected
(Table 3) 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
[0240] The KNN algorithm (see e.g., Ma et al., Arch Pathol Lab Med
2006; 130:465-73) calculated 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 48 microRNAs
that were used by the decision-tree). KNN algorithms with k=1; 10
were compared, and the optimal performer was selected, using
k=7.
9. qRT-PCR
[0241] Total RNA (1 .mu.g) is subjected to polyadenylation reaction
as described before (Gilad et al., PLoS ONE 2008; 3:e3148).
Briefly, RNA is incubated in the presence of poly (A) polymerase
(PAP) (Takara-2180A), MnCl2, and ATP for 1 h at 37.degree. C.
Reverse transcription is performed on the total RNA. An oligodT
primer harboring a consensus sequence (complementary to the reverse
primer, oligodT starch, an N nucleotide (a mixture of all A, C, and
G) and V nucleotide (mixture of four nucleotides) was used for the
reverse transcription reaction. The primer was first annealed to
the polyA-RNA and then subjected to a reverse transcription
reaction of SuperScript II RT (Invitrogen). The cDNA was then
amplified by a real-time PCR reaction, using a microRNA-specific
forward primer, TaqMan probe and universal reverse primer that is
complementary to the 3' sequence of the oligo dT tail. The
reactions were incubated for 10 min at 95.degree. C., followed by
42 cycles of 95.degree. C. for 15 s and 60.degree. C. for 1 min.
qRT-PCR was performed using probes for the 104 candidate microRNAs,
of which 5 were tested with two different forward primers, and for
U6 snoRNA.
10. Feature Selection and Training
[0242] The training samples were kept with average C.sub.t below 36
and at least 30 microRNAs detected (C.sub.t<38). Each sample was
normalized by subtracting from the C.sub.t of each microRNA the
average C.sub.t of all microRNAs of the sample, and adding back a
scaling constant (the average C.sub.t over the entire sample set).
Feature selection and classifier training were using the scaled
C.sub.t as the input signal. The feature selection resulted in a
set of 48 microRNAs. The decision-tree (FIG. 1) used logistic
regression on combinations of two-to-three microRNAs in each node
to make binary decisions. The KNN was based on comparing the
expression of all 48 microRNAs in each sample to all other samples
in the training database. The decision-tree and KNN each return a
predicted tissue of origin and histological type where applicable.
The classifier returns the two different predictions or a single
consensus prediction if the predictions concur. When the
decision-tree and KNN predict different histological types of the
same tissue of origin, the tissue of origin is returned as a
consensus prediction with no histological type indicated.
11. Test Protocol
[0243] RNA was extracted in batches together with a negative
control. The negative control was a no-RNA sample that served to
detect potential contaminations, and should not give any signal in
the PCR reaction. The extracted RNA, together with a positive
control sample, underwent cDNA preparation and 48 microRNAs were
measured by qRT-PCR in duplicates in one 96-well plate per sample.
The positive control was a specific RNA sample that should meet
defined C.sub.t ranges in the assay. Quality assessment of each
well was based on the fluorescence amplification curve, using
thresholds on the maximal fluorescence and the linear slope as a
function of the measured C.sub.t. For each microRNA,
C.sub.t.sup.miR was calculated by taking the average C.sub.t of the
two repeats. Quality assessment for each sample was based on the
number and identity of expressed microRNAs (C.sub.t<38) and the
average C.sub.t of the measured microRNAs. C.sub.t.sup.miR values
for each sample were normalized by rescaling as described above.
The rescaled values were used as input to the classifier that was
trained using qRT-PCR data (as described above).
Example 1
Samples and Profiling
[0244] A discovery process that profiled hundreds of samples on the
array platforms was performed to identify candidate biomarkers. A
training set of .about.400 FFPE samples was used. RNA was extracted
from these samples and qRT-PCR was preformed. An assay was
constructed using 48 microRNAs (Table 3; FIGS. 1-7), to
differentiate between 26 classes representing 18 tissue origins. An
alternative assay was constructed, which does not identify bladder
as an origin, i.e., differentiates between 25 classes representing
17 tissue origins.
[0245] A validation set of 255 new FFPE tumor samples was used to
assess the performance of the assay, representing 26 different
tumor origins or "classes" (see Table 2 for a summary of samples).
About half of the samples in the set were metastatic tumors to
different sites (e.g., lung, bone, brain and liver). Tumor
percentage was at least 50% for all samples in the set.
TABLE-US-00002 TABLE 2 Cancer types, classes and histology Class
Cancer types and histological classifications 1 bladder
transitional cell carcinoma 2 biliary tract cholangiocarcinoma,
gallbladder adenocarcinoma 3 brain-astrocytoma astrocytic tumor;
astrocytic tumor, anaplastic astrocytoma; astrocytic tumor,
glioblastoma multiforme 4 brain- oligodendroglial tumor, anaplastic
oligodendroglioma oligodendroglioma; oligodendroglial tumor,
oligodendroglioma 5 breast adenocarcinoma; invasive ductal
carcinoma 6 colon adenocarcinoma 7 esophagus-squamous squamous cell
carcinoma 8 esophagus-stomach esophagus adenocarcinoma; stomach
adenocarcinoma 9 head & neck squamous cell carcinoma of the
larynx, pharynx and nose 10 kidney renal cell carcinoma; clear cell
carcinoma 11 liver hepatocellular carcinoma 12 lung-carcinoid
neuroendocrine, carcinoid 13 lung-squamous NSCLC, squamous cell
carcinoma 14 lung-adeno-large non-small, adenocarcinoma; non-small,
large cell carcinoma 15 lung-small neuroendocrine, small 16
melanoma malignant melanoma 17 ovary-serous ovary serous
adenocarcinoma 18 ovary-endometrioid ovary endometrioid
adenocarcinoma 19 pancreas adenocarcinoma 20 prostate
adenocarcinoma 21 testis-seminoma GCT; seminoma 22
testis-non-seminoma GCT; non-seminoma 23 thymus thymoma - type B2;
thymoma - type B3 24 thyroid-follicular follicular carcinoma 25
thyroid-medullary neuroendocrine; medullary 26 thyroid-papillary
papillary carcinoma; tall cell
Example 2
Decision-Tree Classification Algorithm
[0246] A tumor classifier was built using the microRNA expression
levels by applying a binary tree classification scheme (FIG. 1).
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 3). 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
liver-cholangio tumors and tumors of non-liver origin, and these
are therefore separated at node #1 (FIG. 2) into separate branches
(FIG. 1).
[0247] 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. 2-7 and Methods). Since gene
expression classifiers have an inherent redundancy in selecting the
gene features, we used bootstrapping 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 48 microRNAs for the full classifier (Table 3).
This approach provides a systematic process for identifying new
biomarkers for differential expression.
Example 3
Defining High-Confidence Classifications
[0248] In clinical practice it is often useful to assess
information of different degrees of confidence.sup.17,18. In the
diagnosis of tumor origin in particular, a short list of highly
probable possibilities is a practical option when no definite
diagnosis can be made. Since the decision-tree and the KNN
algorithms are designed differently and trained independently,
improved accuracy and greater confidence can be obtained by
combining and comparing their classifications. When the two
classifiers agree, the diagnosis is considered high-confidence and
a single origin is identified. When the two disagree, the
classification is made with low-confidence and two origins are
suggested. Sensitivity of the union refers to the percentage in
which at least one of the classifiers (Tree and KNN) was
correct.
Example 4
Performance of the Test in Blinded Validation
[0249] The test performance was assessed using an independent set
of 204 validation samples. These archival samples included primary
as well as metastatic tumor samples, preserved as FFPE blocks,
whose original clinical diagnosis ("reference diagnosis") was one
of the origins on which the classifier was trained. The samples
were processed by personnel who were blinded to the original
reference diagnosis for these samples, and classifications were
automatically generated by dedicated software. 16 of the 204
samples (8%) failed QA criteria. For 188 samples (92%), including
87 metastatic tumor samples (46% of the samples), the test was
completed successfully and produced tissue-of-origin predictions.
For 159 of these samples (84%), the reference diagnosis for tissue
of origin was predicted by at least one of the two classifiers
(Table 4). For 124 samples (66%), the two classifiers agreed,
generating a consensus prediction for a single tissue-of-origin.
For these single-prediction cases, the sensitivity (positive
agreement) was 90% (111/124 of the classifications agreed with the
reference diagnosis), and it exceeded 90% for most tissue-types.
Specificity (negative agreement) in this group ranged from 94% to
100%.
[0250] FFPE sections from 73 of the validation samples were
processed independently and blindly in a second laboratory. Data
and classifications for these samples were compared between the two
laboratories. The mean correlation for the qRT-PCR signals was
0.979 (4 samples had correlation coefficients between 0.91 and
0.95, all other correlations were greater than 0.95). The two labs
disagreed on only 4 samples. For another 8, they had one of two
answers in common and for the remaining 61, classifications matched
perfectly between the two laboratories, demonstrating the precision
of the test.
TABLE-US-00003 TABLE 3 Nodes of the decision-tree and microRNAs (#
SEQ ID NO.) used in each node Left Node Right Node Num Or Node Num
Node Node Node Node Node Node All Classes Num Class Or Class miR 1
miR 2 miR 3 Beta 0 Node Beta 1 Beta 2 Node Beta 3 Right 1 2 3 hsa-
hsa- -- 9.11E+01 4.42E+00 -8.39E+00 NaN biliary tract miR- miR-
carcinoma, liver 200c 122 (#26) (#6) 2 liver biliary tract hsa-
hsa- -- -3.10E+03 6.76E+01 2.48E+01 NaN liver carcinoma miR- miR-
200b 126 (#25) (#9) 3 4 5 hsa- -- -- 5.34E+02 -1.56E+01 NaN NaN
testis-non-seminoma, miR- testis-seminoma 372 (#41) 4 testis-non-
testis- hsa- hsa- hsa- -6.13E+02 -2.10E+01 -1.59E+01 5.68E+01
testis-non-seminoma seminoma seminoma miR- miR- miR- 451 221 92b
(#45) (#31) (#48) 5 9 6 hsa- hsa- -- 1.18E+02 1.63E+00 -5.26E+00
NaN biliary tract miR- miR- carcinoma, bladder, 148b 200c breast,
colon, (#17) (#26) esophagus-squamous, head_neck, lung-adeno large,
lung-carcinoid, lung-small_cell, lung- squamous, ovary-
endometrioid, ovary- serous, pancreas, prostate, stomach/esophagus-
adeno, thymus, thyroid-follicular, thyroid-medullary,
thyroid-papillary 6 melanoma 7 hsa- hsa- -- -6.61E+02 3.58E+01
-1.54E+01 NaN melanoma miR- miR- 497 146a (#46) (#15) 7 8 kidney
hsa- hsa- -- 6.66E+02 -1.02E+01 -8.88E+00 NaN brain-astrocytoma,
miR- miR- brain- 9* 124 oligodendroglioma (#47) (#7) 8 brain-
brain- hsa- hsa- -- -3.99E+03 7.04E+01 5.75E+01 NaN
brain-astrocytoma astrocytoma oligodendro- miR- miR- glioma 497 128
(#46) (#10) 9 10 12 hsa- hsa- hsa- 2.56E+01 -1.20E+00 1.29E+00
-1.22E+00 lung-carcinoid, lung- miR- miR- miR- small cell, thyroid-
15b 152 375 medullary (#19) (#18) (#42) 10 11 thyroid- hsa- hsa- --
-3.52E+02 2.97E+01 -1.89E+01 NaN lung-carcinoid, lung- medullary
miR- miR- small cell 222 200a (#32) (#24) 11 lung- lung-small hsa-
hsa- -- 2.53E+02 2.36E+01 -3.12E+01 NaN lung-carcinoid carcinoid
cell miR- miR- 106a 29c (#3) (#36) 12 13 16 hsa- hsa- -- 7.25E+01
3.73E-01 -2.38E+00 NaN Biliary tract miR- miR- carcinoma, colon,
106a 192 pancreas, (#3) (#21) stomach/esophagus- adeno 13 14 15
hsa- hsa- hsa- -1.40E+02 1.90E-01 3.33E+00 1.18E+00 colon, miR-
let-7b miR- stomach/esophagus- 21 (#1) 30a adeno (#29) (#37) 14
colon Stomach hsa- hsa- hsa- 9.31E+02 -1.18E+01 1.32E+01 -3.37E+01
colon esophagus- miR- miR- miR- adeno 10a 92b 29a- (#4) (#48) fw18
(#34) 15 pancreas biliary tract hsa- hsa- hsa- -2.06E+02 9.37E+00
-6.45E+00 3.10E+00 pancreas carcinoma miR- miR- miR- 25 200c 20a-
(#33) (#26) fw18 (#28) 16 prostate 17 hsa- hsa- -- 2.68E+02
8.84E+00 -2.10E+01 NaN prostate miR- miR- 185 375 (#20) (#42) 17 18
19 hsa- hsa- hsa- 1.35E+02 -1.53E+00 -1.63E+00 -8.83E-01
ovary-endometrioid, miR- miR- miR- ovary-serous 10b- 130a 210 fw18
(#11) (#30) (#5) 18 ovary- ovary- hsa- hsa- hsa- -3.81E+02 3.64E+00
4.26E+00 3.38E+00 ovary-endometrioid endometrioid serous miR- miR-
let-7f 148b 193a- (#2) (#17) 3p (#22) 19 20 breast hsa- hsa- --
-1.32E+02 2.26E+00 1.66E+00 NaN bladder, esophagus- miR- miR-
squamous, head/neck, 193a- 342-3p lung-adeno large, 3p (#39)
lung-squamous, (#22) thymus, thyroid- follicular, thyroid-
papillary 20 23 21 hsa- hsa- -- -2.02E+01 -1.22E+00 1.82E+00 NaN
bladder, esophagus- miR- miR- squamous, head/neck, 205 146b-5p
lung-squamous, (#27) (#16) thymus 21 lung-adeno 22 hsa- hsa- --
-3.03E+03 3.93E+01 6.02E+01 NaN lung-adeno large large miR- miR-
125b 30a (#8) (#37) 22 thyroid- thyroid- hsa- hsa- -- -1.53E+03
2.78E+01 2.17E+01 NaN thyroid-follicular follicular papillary miR-
miR- 31 21 (#38) (#29) 23 24 thymus hsa- hsa- -- -9.24E+01 5.39E+00
-2.98E+00 NaN bladder, esophagus- miR- miR- squamous, head/neck,
29b 21 lung-squamous (#35) (#29) 24 25 bladder hsa- hsa- hsa-
-9.03E+01 2.70E+00 1.79E+00 -1.73E+00 esophagus-squamous, miR- miR-
miR- head/neck, lung- 425 10a 375 squamous (#44) (#4) (#42) 25
lung- 26 hsa- hsa- hsa- 2.52E+02 -2.10E+00 -3.19E+00 -2.50E+00
lung-squamous squamous miR- miR- miR- 10a 19b 222 (#4) (#23) (#32)
26 esophagus- head/neck hsa- hsa- hsa- -1.32E+02 -1.75E+01 4.47E+00
1.53E+01 esophagus-squamous squamous miR- miR- miR- 143 451 30a
(#14) (#45) (#37) Node Num The number of the node (1-26) Left Node
Num Or Left branch - the node number or the class reached Class
Right Node Num Or Right branch - the node number or the class
reached Class Node miR1 miRs used in node - #1 Node miR2 miRs used
in node - #2 (could be empty) Node miR3 miRs used in node - #3
(could be empty) Node Beta 0 The value of the beta0 (intercept)
Node Beta 1 The value of the beta1, corresponding to nodeMir1 Node
Beta 2 The value of the beta2, corresponding to nodeMir2 - could be
NaN (empty) Node Beta 3 The value of the beta3, corresponding to
nodeMir3 - could be NaN (empty) Node All Classes A list of all the
classes that are on the left branch Left Node All Classes A list of
all the classes that are on the right branch Right
TABLE-US-00004 TABLE 4 Performance of the test in blinded
validation Successful Sensitivity Specificity Fraction in
Sensitivity Specificity samples in union of union of high of high
of high Class test set prediction prediction confidence confidence
confidence Biliary tract 6 66.67 93.96 33.33 100 98.36 Brain 10 100
100 80 100 100 Breast 33 66.67 93.55 45.45 53.33 100 Colon 9 88.89
94.41 66.67 83.33 99.15 Esophagus 1 100 98.4 0 NaN 100 Neck &
Head 3 100 92.43 100 100 97.52 Kidney 8 87.5 99.44 62.5 80 100
Liver 8 100 99.44 100 100 100 Lung 23 91.3 84.85 86.96 95 94.23
Melanocyte 7 85.71 97.79 85.71 83.33 100 Ovary 13 84.62 100 38.46
100 100 Pancreas 6 50 97.8 16.67 100 99.19 Prostate 19 89.47 99.41
57.89 100 100 Stomach or 5 40 98.91 40 50 100 esophagus Testis 7
100 100 100 100 100 Thymus 6 83.33 97.8 83.33 80 100 Thyroid 24 100
98.17 83.33 100 100 Total 188 84.57 96.91 65.96 89.52 99.34
Example 5
Classification Example
[0251] One of the training-set samples originally diagnosed in the
clinical setting as a metastatic tumor to the brain originating
from the lung, was classified by the tree (in leave-one-out
cross-validation) as originating from the liver. This
classification was traced back to node #1, the branching point
where lung and liver origins diverge (FIG. 1). This node uses
hsa-miR-122 (SEQ ID NO: 6), together with hsa-miR-200c (SEQ ID NO:
26). The expression of these microRNAs in this sample, in
particular the very high expression of hsa-miR-122 (FIG. 8A), are
strong indicators of a possible hepatic origin of this sample. Upon
re-examination of the clinical record, it was found that this
sample was originally classified as a lung metastasis based on the
fact that the patient had a known mass in the lung. This
disagreement between the original clinical diagnosis and our test
was followed up by blinded pathological review. Indeed, the
sample's immunohistochemical staining pattern was incompatible with
lung adenocarcinoma origin, but was consistent with a diagnosis of
hepatocellular carcinoma (FIG. 8B). Thus, the test could suggest an
alternative diagnosis for this patient, namely a primary
hepatocellular carcinoma with metastatic spread to both lung and
brain.
Example 6
[0252] Variant microRNAs
[0253] For some of the microRNAs in Table 3, other variant
microRNAs having a similar seed sequence (identical nucleotides
2-8) are known in the human genome (see Table 5), and are therefore
considered to target a very similar set of (mRNA-coding) genes (via
the RISC machinery). These microRNAs with identical seed sequence
may be substituted for the indicated miRs.
TABLE-US-00005 TABLE 5 microRNAs with identical seed sequence miRs
with SEQ Indicated miRs Seed same seed miR sequence ID NO:
hsa-let-7b GAGGTAG hsa-let-7d AGAGGTAGTAGGTTGCATAGTT 152 GAGGTAG
hsa-let-7e TGAGGTAGGAGGTTGTATAGTT 153 GAGGTAG hsa-miR-98
TGAGGTAGTAAGTTGTATTGTT 154 GAGGTAG hsa-let-7f
TGAGGTAGTAGATTGTATAGTT 2 GAGGTAG hsa-let-7a TGAGGTAGTAGGTTGTATAGTT
155 GAGGTAG hsa-let-7c TGAGGTAGTAGGTTGTATGGTT 156 GAGGTAG
hsa-let-7g TGAGGTAGTAGTTTGTACAGTT 157 GAGGTAG hsa-let-7i
TGAGGTAGTAGTTTGTGCTGTT 158 hsa-let-7f GAGGTAG hsa-let-7d
AGAGGTAGTAGGTTGCATAGTT 152 GAGGTAG hsa-let-7e
TGAGGTAGGAGGTTGTATAGTT 153 GAGGTAG hsa-miR-98
TGAGGTAGTAAGTTGTATTGTT 154 GAGGTAG hsa-let-7c
TGAGGTAGTAGGTTGTATGGTT 156 GAGGTAG hsa-let-7b
TGAGGTAGTAGGTTGTGTGGTT 1 GAGGTAG hsa-let-7g TGAGGTAGTAGTTTGTACAGTT
157 GAGGTAG hsa-let-7i TGAGGTAGTAGTTTGTGCTGTT 158 hsa-miR-106a
AAAGTGC hsa-miR-519d CAAAGTGCCTCCCTTTAGAGTG 159 AAAGTGC hsa-miR-20b
CAAAGTGCTCATAGTGCAGGTAG 160 AAAGTGC hsa-miR-93
CAAAGTGCTGTTCGTGCAGGTAG 161 AAAGTGC hsa-miR-17
CAAAGTGCTTACAGTGCAGGTAG 162 AAAGTGC hsa-miR-526b*
GAAAGTGCTTCCTTTTAGAGGC 163 AAAGTGC hsa-miR-106b
TAAAGTGCTGACAGTGCAGAT 164 AAAGTGC hsa-miR-20a
TAAAGTGCTTATAGTGCAGGTAG 28 hsa-miR-10a ACCCTGT hsa-miR-10b
TACCCTGTAGAACCGAATTTGTG 165 hsa-miR-10b ACCCTGT hsa-miR-10a
TACCCTGTAGATCCGAATTTGTG 4 hsa-miR-124 AAGGCAC hsa-miR-506
TAAGGCACCCTTCTGAGTAGA 166 hsa-miR-125b CCCTGAG hsa-miR-125a-5p
TCCCTGAGACCCTTTAACCTGTGA 167 hsa-miR-130a AGTGCAA hsa-miR-301a
CAGTGCAATAGTATTGTCAAAGC 168 AGTGCAA hsa-miR-301b
CAGTGCAATGATATTGTCAAAGC 169 AGTGCAA hsa-miR-130b
CAGTGCAATGATGAAAGGGCAT 170 AGTGCAA hsa-miR-454
TAGTGCAATATTGCTTATAGGGT 171 hsa-miR-146a GAGAACT hsa-miR-146b-5p
TGAGAACTGAATTCCATAGGCT 16 hsa-miR-146b-5p GAGAACT hsa-miR-146a
TGAGAACTGAATTCCATGGGTT 15 hsa-miR-148b CAGTGCA hsa-miR-148a
TCAGTGCACTACAGAACTTTGT 172 CAGTGCA hsa-miR-152
TCAGTGCATGACAGAACTTGG 18 hsa-miR-152 CAGTGCA hsa-miR-148a
TCAGTGCACTACAGAACTTTGT 172 CAGTGCA hsa-miR-148b
TCAGTGCATCACAGAACTTTGT 17 hsa-miR-15b AGCAGCA hsa-miR-424
CAGCAGCAATTCATGTTTTGAA 173 AGCAGCA hsa-miR-497
CAGCAGCACACTGTGGTTTGT 46 AGCAGCA hsa-miR-195 TAGCAGCACAGAAATATTGGC
174 AGCAGCA hsa-miR-15a TAGCAGCACATAATGGTTTGTG 175 AGCAGCA
hsa-miR-16 TAGCAGCACGTAAATATTGGCG 176 hsa-miR-192 TGACCTA
hsa-miR-215 ATGACCTATGAATTGACAGAC 177 hsa-miR-193a-3p ACTGGCC
hsa-miR-193b AACTGGCCCTCAAAGTCCCGCT 178 hsa-miR-19b GTGCAAA
hsa-miR-19a TGTGCAAATCTATGCAAAACTGA 179 hsa-miR-200a AACACTG
hsa-miR-141 TAACACTGTCTGGTAAAGATGG 180 hsa-miR-200b AATACTG
hsa-miR-200c TAATACTGCCGGGTAATGATGGA 26 hsa-miR-200b AATACTG
hsa-miR-429 TAATACTGTCTGGTAAAACCGT 181 hsa-miR-200c AATACTG
hsa-miR-200b TAATACTGCCTGGTAATGATGA 25 AATACTG hsa-miR-429
TAATACTGTCTGGTAAAACCGT 181 hsa-miR-20a AAAGTGC hsa-miR-106a
AAAAGTGCTTACAGTGCAGGTAG 3 AAAGTGC hsa-miR-519d
CAAAGTGCCTCCCTTTAGAGTG 159 AAAGTGC hsa-miR-20b
CAAAGTGCTCATAGTGCAGGTAG 160 AAAGTGC hsa-miR-93
CAAAGTGCTGTTCGTGCAGGTAG 161 AAAGTGC hsa-miR-17
CAAAGTGCTTACAGTGCAGGTAG 162 AAAGTGC hsa-miR-526b*
GAAAGTGCTTCCTTTTAGAGGC 163 AAAGTGC hsa-miR-106b
TAAAGTGCTGACAGTGCAGAT 164 hsa-miR-21 AGCTTAT hsa-miR-590-5p
GAGCTTATTCATAAAAGTGCAG 182 hsa-miR-221 GCTACAT hsa-miR-222
AGCTACATCTGGCTACTGGGT 32 hsa-miR-222 GCTACAT hsa-miR-221
AGCTACATTGTCTGCTGGGTTTC 31 hsa-miR-25 ATTGCAC hsa-miR-363
AATTGCACGGTATCCATCTGTA 184 ATTGCAC hsa-miR-367
AATTGCACTTTAGCAATGGTGA 185 ATTGCAC hsa-miR-32
TATTGCACATTACTAAGTTGCA 186 ATTGCAC hsa-miR-92b
TATTGCACTCGTCCCGGCCTCC 48 ATTGCAC hsa-miR-92a
TATTGCACTTGTCCCGGCCTGT 187 hsa-miR-29a AGCACCA hsa-miR-29b
TAGCACCATTTGAAATCAGTGTT 35 AGCACCA hsa-miR-29c
TAGCACCATTTGAAATCGGTTA 36 hsa-miR-29b AGCACCA hsa-miR-29a
TAGCACCATCTGAAATCGGTTA 34 AGCACCA hsa-miR-29c
TAGCACCATTTGAAATCGGTTA 36 hsa-miR-29c AGCACCA hsa-miR-29a
TAGCACCATCTGAAATCGGTTA 34 AGCACCA hsa-miR-29b
TAGCACCATTTGAAATCAGTGTT 35 hsa-miR-30a GTAAACA hsa-miR-30d
TGTAAACATCCCCGACTGGAAG 188 GTAAACA hsa-miR-30b
TGTAAACATCCTACACTCAGCT 189 GTAAACA hsa-miR-30c
TGTAAACATCCTACACTCTCAGC 190 GTAAACA hsa-miR-30e
TGTAAACATCCTTGACTGGAAG 191 hsa-miR-372 AAGTGCT hsa-miR-520a-3p
AAAGTGCTTCCCTTTGGACTGT 192 AAGTGCT hsa-miR-520b
AAAGTGCTTCCTTTTAGAGGG 193 AAGTGCT hsa-miR-520c-3p
AAAGTGCTTCCTTTTAGAGGGT 194 AAGTGCT hsa-miR-520e
AAAGTGCTTCCTTTTTGAGGG 195 AAGTGCT hsa-miR-520e-3p
AAAGTGCTTCTCTTTGGTGGGT 196 AAGTGCT hsa-miR-373
GAAGTGCTTCGATTTTGGGGTGT 197 AAGTGCT hsa-miR-302e TAAGTGCTTCCATGCTT
198 AAGTGCT hsa-miR-302c TAAGTGCTTCCATGTTTCAGTGG 199 AAGTGCT
hsa-miR-302d TAAGTGCTTCCATGTTTGAGTGT 200 AAGTGCT hsa-miR-202b
TAAGTGCTTCCATGTTTTAGTAG 201 AAGTGCT hsa-miR-302a
TAAGTGCTTCCATGTTTTGGTGA 202 hsa-miR-378 CTGGACT hsa-miR-422a
ACTGGACTTAGGGTCAGAAGGC 203 hsa-miR-497 AGCAGCA hsa-miR-424
CAGCAGCAATTCATGTTTTGAA 173 AGCAGCA hsa-miR-195
TAGCAGCACAGAAATATTGGC 174 AGCAGCA hsa-miR-15a
TAGCAGCACATAATGGTTTGTG 175 AGCAGCA hsa-miR-15b
TAGCAGCACATCATGGTTTACA 19 AGCAGCA hsa-miR-16 TAGCAGCACGTAAATATTGGCG
176 hsa-miR-92b ATTGCAC hsa-miR-363 AATTGCACGGTATCCATCTGTA 184
ATTGCAC hsa-miR-367 AATTGCACTTTAGCAATGGTGA 185 ATTGCAC hsa-miR-25
CATTGCACTTGTCTCGGTCTGA 33 ATTGCAC hsa-miR-32 TATTGCACATTACTAAGTTGCA
186 ATTGCAC hsa-miR-92a TATTGCACTTGTCCCGGCCTGT 187
[0254] For some of the microRNAs in Table 3, other microRNAs that
are known in the human genome are located in close proximity on the
genome (genomic cluster) (see Table 6), and are co-transcribed with
the indicated miRs. These microRNAs from nearly the same genomic
location may be substituted for the indicated miRs.
TABLE-US-00006 TABLE 6 microRNAs within the same genomic cluster
miRs within the SEQ Indicated miRs same genomic cluster miR
sequence ID NO: hsa-let-7b hsa-let-7a TGAGGTAGTAGGTTGTATAGTT 155
hsa-let-7a* CTATACAATCTACTGTCTTTC 204 hsa-let-7b*
CTATACAACCTACTGCCTTCCC 205 hsa-let-7f hsa-let-7a*
CTATACAATCTACTGTCTTTC 204 hsa-let-7a TGAGGTAGTAGGTTGTATAGTT 155
hsa-let-7d AGAGGTAGTAGGTTGCATAGTT 152 hsa-let-7d*
CTATACGACCTGCTGCCTTTCT 206 hsa-let-7f-1* CTATACAATCTATTGCCTTCCC 207
hsa-1et-7f-2* CTATACAGTCTACTGTCTTTCC 208 hsa-miR-98
TGAGGTAGTAAGTTGTATTGTT 154 hsa-miR-106a hsa-miR-19b-2*
AGTTTTGCAGGTTTGCATTTCA 209 hsa-miR-20b CAAAGTGCTCATAGTGCAGGTAG 160
hsa-miR-20b* ACTGTAGTATGGGCACTTCCAG 210 hsa-miR-363
AATTGCACGGTATCCATCTGTA 184 hsa-miR-363* CGGGTGGATCACGATGCAATTT 211
hsa-miR-92a TATTGCACTTGTCCCGGCCTGT 187 hsa-miR-92a-2*
GGGTGGGGATTTGTTGCATTAC 212 hsa-miR-106a* CTGCAATGTAAGCACTTCTTAC 213
hsa-miR-18b TAAGGTGCATCTAGTGCAGTTAG 214 hsa-miR-18b*
TGCCCTAAATGCCCCTTCTGGC 215 hsa-miR-19b TGTGCAAATCCATGCAAAACTGA 23
hsa-miR-10a hsa-miR-10a* CAAATTCGTATCTAGGGGAATA 216 hsa-miR-10b
hsa-miR-10b* ACAGATTCGATTCTAGGGGAAT 217 hsa-miR-122 hsa-miR-122*
AACGCCATTATCACACTAAATA 218 hsa-miR-124 hsa-miR-124*
CGTGTTCACAGCGGACCTTGAT 219 hsa-miR-125b hsa-miR-125b-1*
ACGGGTTAGGCTCTTGGGAGCT 220 hsa-miR-125b-2* TCACAAGTCAGGCTCTTGGGAC
221 hsa-miR-99a AACCCGTAGATCCGATCTTGTG 222 hsa-miR-99a*
CAAGCTCGCTTCTATGGGTCTG 223 hsa-miR-100 AACCCGTAGATCCGAACTTGTG 224
hsa-miR-100* CAAGCTTGTATCTATAGGTATG 225 hsa-let-7a
TGAGGTAGTAGGTTGTATAGTT 155 hsa-let-7c TGAGGTAGTAGGTTGTATGGTT 156
hsa-let-7c* TAGAGTTACACCCTGGGAGTTA 226 hsa-miR-126 hsa-miR-126*
CATTATTACTTTTGGTACGCG 227 hsa-miR-130a hsa-miR-130a*
TTCACATTGTGCTACTGTCTGC 228 hsa-miR-138 hsa-miR-138-1*
GCTACTTCACAACACCAGGGCC 229 hsa-miR-138-2* GCTATTTCACGACACCAGGGTT
230 hsa-miR-142-3p hsa-miR-142-5p CATAAAGTAGAAAGCACTACT 231
hsa-miR-143 hsa-miR-143* GGTGCAGTGCTGCATCTCTGGT 232 hsa-miR-145
GTCCAGTTTTCCCAGGAATCCCT 233 hsa-miR-145* GGATTCCTGGAAATACTGTTCT 234
hsa-miR-146a hsa-miR-146a* CCTCTGAAATTCAGTTCTTCAG 235
hsa-miR-146b-5p hsa-miR-146b-3p TGCCCTGTGGACTCAGTTCTGG 236
hsa-miR-148b hsa-miR-148b* AAGTTCTGTTATACACTCAGGC 237 hsa-miR-15b
hsa-miR-15b* CGAATCATTATTTGCTGCTCTA 238 hsa-miR-16
TAGCAGCACGTAAATATTGGCG 176 hsa-miR-16-2* CCAATATTACTGTGCTGCTTTA 239
hsa-miR-185 hsa-miR-185* AGGGGCTGGCTTTCCTCTGGTC 240 hsa-miR-1306
ACGTTGGCTCTGGTGGTG 241 hsa-miR-192 hsa-miR-192*
CTGCCAATTCCATAGGTCACAG 242 hsa-miR-194 TGTAACAGCAACTCCATGTGGA 243
hsa-miR-194* CCAGTGGGGCTGCTGTTATCTG 244 hsa-miR-193a-3p
hsa-miR-193a-5p TGGGTCTTTGCGGGCGAGATGA 245 hsa-miR-365
TAATGCCCCTAAAAATCCTTAT 246 hsa-miR-19b hsa-miR-19a
TGTGCAAATCTATGCAAAACTGA 179 hsa-miR-19a* AGTTTTGCATAGTTGCACTACA 247
hsa-miR-18a TAAGGTGCATCTAGTGCAGATAG 248 hsa-miR-18a*
ACTGCCCTAAGTGCTCCTTCTGG 249 hsa-miR-18b TAAGGTGCATCTAGTGCAGTTAG 214
hsa-miR-18b* TGCCCTAAATGCCCCTTCTGGC 215 hsa-miR-17
CAAAGTGCTTACAGTGCAGGTAG 162 hsa-miR-17* ACTGCAGTGAAGGCACTTGTAG 250
hsa-miR-106a AAAAGTGCTTACAGTGCAGGTAG 3 hsa-miR-106a*
CTGCAATGTAAGCACTTCTTAC 213 hsa-miR-20a* ACTGCATTATGAGCACTTAAAG 251
hsa-miR-20b CAAAGTGCTCATAGTGCAGGTAG 160 hsa-miR-20b*
ACTGTAGTATGGGCACTTCCAG 210 hsa-miR-363 AATTGCACGGTATCCATCTGTA 184
hsa-miR-363* CGGGTGGATCACGATGCAATTT 211 hsa-miR-92a
TATTGCACTTGTCCCGGCCTGT 187 hsa-miR-92a-1* AGGTTGGGATCGGTTGCAATGCT
252 hsa-miR-92a-2* GGGTGGGGATTTGTTGCATTAC 212 hsa-miR-19b-1*
AGTTTTGCAGGTTTGCATCCAGC 253 hsa-miR-19b-2* AGTTTTGCAGGTTTGCATTTCA
209 hsa-miR-20a TAAAGTGCTTATAGTGCAGGTAG 28 hsa-miR-200a
hsa-miR-200b* CATCTTACTGGGCAGCATTGGA 254 hsa-miR-429
TAATACTGTCTGGTAAAACCGT 181 hsa-miR-200a* CATCTTACCGGACAGTGCTGGA 255
hsa-miR-200b TAATACTGCCTGGTAATGATGA 25 hsa-miR-200b hsa-miR-200a
TAACACTGTCTGGTAACGATGTT 24 hsa-miR-200a* CATCTTACCGGACAGTGCTGGA 255
hsa-miR-200b* CATCTTACTGGGCAGCATTGGA 254 hsa-miR-429
TAATACTGTCTGGTAAAACCGT 181 hsa-miR-200c hsa-miR-200c*
CGTCTTACCCAGCAGTGTTTGG 256 hsa-miR-141 TAACACTGTCTGGTAAAGATGG 180
hsa-miR-141* CATCTTCCAGTACAGTGTTGGA 257 hsa-miR-20a hsa-miR-17*
ACTGCAGTGAAGGCACTTGTAG 250 hsa-miR-17 CAAAGTGCTTACAGTGCAGGTAG 162
hsa-miR-18a* ACTGCCCTAAGTGCTCCTTCTGG 249 hsa-miR-18a
TAAGGTGCATCTAGTGCAGATAG 248 hsa-miR-19a* AGTTTTGCATAGTTGCACTACA 247
hsa-miR-19a TGTGCAAATCTATGCAAAACTGA 179 hsa-miR-20a*
ACTGCATTATGAGCACTTAAAG 251 hsa-miR-92a TATTGCACTTGTCCCGGCCTGT 187
hsa-miR-92a-1* AGGTTGGGATCGGTTGCAATGCT 252 hsa-miR-19b-1*
AGTTTTGCAGGTTTGCATCCAGC 253 hsa-miR-19b TGTGCAAATCCATGCAAAACTGA 23
hsa-miR-21 hsa-miR-21* CAACACCAGTCGATGGGCTGT 258 hsa-miR-221
hsa-miR-221* ACCTGGCATACAATGTAGATTT 259 hsa-miR-222
AGCTACATCTGGCTACTGGGT 183 hsa-miR-222* CTCAGTAGCCAGTGTAGATCCT 260
hsa-miR-222 hsa-miR-221* ACCTGGCATACAATGTAGATTT 259 hsa-miR-222*
CTCAGTAGCCAGTGTAGATCCT 260 hsa-miR-221 AGCTACATTGTCTGCTGGGTTTC 31
hsa-miR-25 hsa-miR-25* AGGCGGAGACTTGGGCAATTG 261 hsa-miR-93
CAAAGTGCTGTTCGTGCAGGTAG 161 hsa-miR-93* ACTGCTGAGCTAGCACTTCCCG 262
hsa-miR-106b TAAAGTGCTGACAGTGCAGAT 164 hsa-miR-106b*
CCGCACTGTGGGTACTTGCTGC 263 hsa-miR-29a hsa-miR-29a*
ACTGATTTCTTTTGGTGTTCAG 264 hsa-miR-29b TAGCACCATTTGAAATCAGTGTT 35
hsa-miR-29b-1* GCTGGTTTCATATGGTGGTTTAGA 265 hsa-miR-29b
hsa-miR-29a* ACTGATTTCTTTTGGTGTTCAG 264 hsa-miR-29b-1*
GCTGGTTTCATATGGTGGTTTAGA 265 hsa-miR-29b-2* CTGGTTTCACATGGTGGCTTAG
266 hsa-miR-29c TAGCACCATTTGAAATCGGTTA 36 hsa-miR-29a
TAGCACCATCTGAAATCGGTTA 34 hsa-miR-29c* TGACCGATTTCTCCTGGTGTTC 267
hsa-miR-29c hsa-miR-29b-2* CTGGTTTCACATGGTGGCTTAG 266 hsa-miR-29c*
TGACCGATTTCTCCTGGTGTTC 267 hsa-miR-29b TAGCACCATTTGAAATCAGTGTT
35
hsa-miR-30a hsa-miR-30a* CTTTCAGTCGGATGTTTGCAGC 268 hsa-miR-30c
TGTAAACATCCTACACTCTCAGC 190 hsa-miR-30c-2* CTGGGAGAAGGCTGTTTACTCT
269 hsa-miR-31 hsa-miR-31* TGCTATGCCAACATATTGCCAT 270
hsa-miR-342-3p hsa-miR-342-5p AGGGGTGCTATCTGTGATTGA 271 hsa-miR-372
hsa-miR-371-3p AAGTGCCGCCATCTTTTGAGTGT 272 hsa-miR-371-5p
ACTCAAACTGTGGGGGCACT 273 hsa-miR-373 GAAGTGCTTCGATTTTGGGGTGT 197
hsa-miR-373* ACTCAAAATGGGGGCGCTTTCC 274 hsa-miR-378 hsa-miR-378*
CTCCTGACTCCAGGTCCTGTGT 275 hsa-miR-425 hsa-miR-425*
ATCGGGAATGTCGTGTCCGCCC 276 hsa-miR-191 CAACGGAATCCCAAAAGCAGCTG 277
hsa-miR-191* GCTGCGCTTGGATTTCGTCCCC 278 hsa-miR-451 hsa-miR-144
TACAGTATAGATGATGTACT 279 hsa-miR-144* GGATATCATCATATACTGTAAG 280
hsa-miR-497 hsa-miR-195 TAGCAGCACAGAAATATTGGC 174 hsa-miR-195*
CCAATATTGGCTGTGCTGCTCC 281 hsa-miR-497* CAAACCACACTGTGGTGTTAGA 282
hsa-miR-9* hsa-miR-9 TCTTTGGTTATCTAGCTGTATGA 283 hsa-miR-92b
hsa-miR-92b* AGGGACGGGACGCGGTGCAGTG 284
[0255] For some of the microRNAs in Table 3, other microRNAs that
are known in the human genome have similar sequences (less than 6
mismatches in the sequence) (see Table 7), and may therefore also
be captured by probes with the same design. These microRNAs with
similar overall sequence may be substituted for the indicated
miRs.
TABLE-US-00007 TABLE 7 microRNAs with similar sequence miRs in
Indicated miRs sequence cluster Sequence SEQ ID NO: hsa-let-7b
hsa-let-7a TGAGGTAGTAGGTTGTATAGTT 155 hsa-let-7e
TGAGGTAGGAGGTTGTATAGTT 153 hsa-let-7c TGAGGTAGTAGGTTGTATGGTT 156
hsa-let-7f TGAGGTAGTAGATTGTATAGTT 2 hsa-let-7d
AGAGGTAGTAGGTTGCATAGTT 152 hsa-miR-1827 TGAGGCAGTAGATTGAAT 285
hsa-let-7g TGAGGTAGTAGTTTGTACAGTT 157 hsa-miR-98
TGAGGTAGTAAGTTGTATTGTT 154 hsa-let-7f hsa-let-7b
TGAGGTAGTAGGTTGTGTGGTT 1 hsa-let-7c TGAGGTAGTAGGTTGTATGGTT 156
hsa-miR-1827 TGAGGCAGTAGATTGAAT 285 hsa-let-7g
TGAGGTAGTAGTTTGTACAGTT 157 hsa-miR-98 TGAGGTAGTAAGTTGTATTGTT 154
hsa-let-7d AGAGGTAGTAGGTTGCATAGTT 152 hsa-let-7e
TGAGGTAGGAGGTTGTATAGTT 153 hsa-let-7a TGAGGTAGTAGGTTGTATAGTT 155
hsa-miR-106a hsa-miR-17 CAAAGTGCTTACAGTGCAGGTAG 162 hsa-miR-93
CAAAGTGCTGTTCGTGCAGGTAG 161 hsa-miR-106b TAAAGTGCTGACAGTGCAGAT 164
hsa-miR-20b CAAAGTGCTCATAGTGCAGGTAG 160 hsa-miR-20a
TAAAGTGCTTATAGTGCAGGTAG 28 hsa-miR-10a hsa-miR-10b
TACCCTGTAGAACCGAATTTGTG 165 hsa-miR-10b hsa-miR-10 a
TACCCTGTAGATCCGAATTTGTG 4 hsa-miR-130a hsa-miR-130b
CAGTGCAATGATGAAAGGGCAT 170 hsa-miR-146a hsa-miR-146b-5p
TGAGAACTGAATTCCATAGGCT 16 hsa-miR-146b-5p hsa-miR-146a
TGAGAACTGAATTCCATGGGTT 15 hsa-miR-148b hsa-miR-148a
TCAGTGCACTACAGAACTTTGT 172 hsa-miR-148b hsa-miR-152
TCAGTGCATGACAGAACTTGG 18 hsa-miR-152 hsa-miR-148b
TCAGTGCATCACAGAACTTTGT 17 hsa-miR-148a TCAGTGCACTACAGAACTTTGT 172
hsa-miR-15b hsa-miR-15a TAGCAGCACATAATGGTTTGTG 175 hsa-miR-192
hsa-miR-215 ATGACCTATGAATTGACAGAC 177 hsa-miR-193a-3p hsa-miR-193b
AACTGGCCCTCAAAGTCCCGCT 178 hsa-miR-19b hsa-miR-19a
TGTGCAAATCTATGCAAAACTGA 179 hsa-miR-200a hsa-miR-141
TAACACTGTCTGGTAAAGATGG 180 hsa-miR-200b hsa-miR-200c
TAATACTGCCGGGTAATGATGGA 26 hsa-miR-200c hsa-miR-200b
TAATACTGCCTGGTAATGATGA 25 hsa-miR-20a hsa-miR-106b
TAAAGTGCTGACAGTGCAGAT 164 hsa-miR-20b CAAAGTGCTCATAGTGCAGGTAG 160
hsa-miR-93 CAAAGTGCTGTTCGTGCAGGTAG 161 hsa-miR-17
CAAAGTGCTTACAGTGCAGGTAG 162 hsa-miR-106a AAAAGTGCTTACAGTGCAGGTAG 3
hsa-miR-29a hsa-miR-29c TAGCACCATTTGAAATCGGTTA 36 hsa-miR-29b
TAGCACCATTTGAAATCAGTGTT 35 hsa-miR-29b hsa-miR-29a
TAGCACCATCTGAAATCGGTTA 34 hsa-miR-29c TAGCACCATTTGAAATCGGTTA 36
hsa-miR-29c hsa-miR-29b TAGCACCATTTGAAATCAGTGTT 35 hsa-miR-29a
TAGCACCATCTGAAATCGGTTA 34 hsa-miR-30a hsa-miR-30d
TGTAAACATCCCCGACTGGAAG 188 hsa-miR-30e TGTAAACATCCTTGACTGGAAG 191
hsa-miR-378 hsa-miR-422a ACTGGACTTAGGGTCAGAAGGC 203 hsa-miR-92b
hsa-miR-92a TATTGCACTTGTCCCGGCCTGT 187
[0256] 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.
[0257] 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
[0258] 1. Bentwich, I. et al. Identification of hundreds of
conserved and nonconserved human microRNAs. Nat Genet (2005).
[0259] 2. Farh, K. K. et al. The Widespread Impact of Mammalian
MicroRNAs on mRNA Repression and Evolution. Science (2005). [0260]
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). [0261] 4. He, L.
et al. A microRNA polycistron as a potential human oncogene. Nature
435, 828-33 (2005). [0262] 5. Baskerville, S. & Bartel, D. P.
Microarray profiling of microRNAs reveals frequent coexpression
with neighboring miRNAs and host genes. Rna 11, 241-7 (2005).
[0263] 6. Landgraf, P. et al. A Mammalian microRNA Expression Atlas
Based on Small RNA Library Sequencing. Cell 129, 1401-14 (2007).
[0264] 7. Volinia, S. et al. A microRNA expression signature of
human solid tumors defines cancer gene targets. Proc Natl Acad Sci
USA (2006). [0265] 8. Lu, J. et al. MicroRNA expression profiles
classify human cancers. Nature 435, 834-8 (2005). [0266] 9.
Varadhachary, G. R., Abbruzzese, J. L. & Lenzi, R. Diagnostic
strategies for unknown primary cancer. Cancer 100, 1776-85 (2004).
[0267] 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). [0268] 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). [0269] 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).
[0270] 13. Blaszyk, H., Hartmann, A. & Bjornsson, J. Cancer of
unknown primary: clinicopathologic correlations. Apmis 111, 1089-94
(2003). [0271] 14. Bloom, G. et al. Multi-platform, multi-site,
microarray-based human tumor classification. Am J Pathol 164, 9-16
(2004). [0272] 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).
[0273] 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).
[0274] 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). [0275] 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).
[0276] 19. Raver-Shapira, N. et al. Transcriptional Activation of
miR-34a Contributes to p53-Mediated Apoptosis. Mol Cell (2007).
[0277] 20. Xiao, C. et al. MiR-150 Controls B Cell Differentiation
by Targeting the Transcription Factor c-Myb. Cell 131, 146-59
(2007).
[0278] The contents of U.S. patent application Ser. No. 14/320,113,
filed Jun. 30, 2014; U.S. patent application Ser. No. 13/167,489,
filed Jun. 23, 2011; International Application No. PCT/IL09/01212,
filed Dec. 23, 2009; U.S. Provisional Application No. 61/140,642,
filed Dec. 24, 2008; U.S. patent application Ser. No. 12/532,940,
filed Sep. 24, 2009; International Application No. PCT/IL08/00396,
filed Mar. 20, 2008; U.S. Provisional Application No. 60/907,266,
filed Mar. 27, 2007; U.S. Provisional Application No. 60/929,244,
filed Jun. 19, 2007; and, U.S. Provisional Application No.
61/024,565, filed Jan. 30, 2008, are herein incorporated by
reference in their entirety for all purposes.
Sequence CWU 1
1
288122RNAHomo Sapiens 1ugagguagua gguugugugg uu 22222RNAHomo
Sapiens 2ugagguagua gauuguauag uu 22323RNAHomo Sapiens 3aaaagugcuu
acagugcagg uag 23423RNAHomo Sapiens 4uacccuguag auccgaauuu gug
23522RNAHomo Sapiens 5uacccuguag aaccgaauuu gu 22622RNAHomo Sapiens
6uggaguguga caaugguguu ug 22720RNAHomo Sapiens 7uaaggcacgc
ggugaaugcc 20822RNAHomo Sapiens 8ucccugagac ccuaacuugu ga
22922RNAHomo Sapiens 9ucguaccgug aguaauaaug cg 221021RNAHomo
Sapiens 10ucacagugaa ccggucucuu u 211122RNAHomo Sapiens
11cagugcaaug uuaaaagggc au 221223RNAHomo Sapiens 12agcugguguu
gugaaucagg ccg 231323RNAHomo Sapiens 13uguaguguuu ccuacuuuau gga
231421RNAHomo Sapiens 14ugagaugaag cacuguagcu c 211522RNAHomo
Sapiens 15ugagaacuga auuccauggg uu 221622RNAHomo Sapiens
16ugagaacuga auuccauagg cu 221722RNAHomo Sapiens 17ucagugcauc
acagaacuuu gu 221821RNAHomo Sapiens 18ucagugcaug acagaacuug g
211922RNAHomo Sapiens 19uagcagcaca ucaugguuua ca 222022RNAHomo
Sapiens 20uggagagaaa ggcaguuccu ga 222121RNAHomo Sapiens
21cugaccuaug aauugacagc c 212222RNAHomo Sapiens 22aacuggccua
caaaguccca gu 222323RNAHomo Sapiens 23ugugcaaauc caugcaaaac uga
232423RNAHomo Sapiens 24uaacacuguc ugguaacgau guu 232522RNAHomo
Sapiens 25uaauacugcc ugguaaugau ga 222623RNAHomo Sapiens
26uaauacugcc ggguaaugau gga 232722RNAHomo Sapiens 27uccuucauuc
caccggaguc ug 222823RNAHomo Sapiens 28uaaagugcuu auagugcagg uag
232922RNAHomo Sapiens 29uagcuuauca gacugauguu ga 223022RNAHomo
Sapiens 30cugugcgugu gacagcggcu ga 223123RNAHomo Sapiens
31agcuacauug ucugcugggu uuc 233221RNAHomo Sapiens 32agcuacaucu
ggcuacuggg u 213322RNAHomo Sapiens 33cauugcacuu gucucggucu ga
223422RNAHomo Sapiens 34uagcaccauc ugaaaucggu ua 223523RNAHomo
Sapiens 35uagcaccauu ugaaaucagu guu 233622RNAHomo Sapiens
36uagcaccauu ugaaaucggu ua 223722RNAHomo Sapiens 37uguaaacauc
cucgacugga ag 223821RNAHomo Sapiens 38aggcaagaug cuggcauagc u
213923RNAHomo Sapiens 39ucucacacag aaaucgcacc cgu 234022RNAHomo
Sapiens 40gcugacuccu aguccagggc uc 224123RNAHomo Sapiens
41aaagugcugc gacauuugag cgu 234222RNAHomo Sapiens 42uuuguucguu
cggcucgcgu ga 224321RNAHomo Sapiens 43acuggacuug gagucagaag g
214423RNAHomo Sapiens 44aaugacacga ucacucccgu uga 234521RNAHomo
Sapiens 45aaaccguuac cauuacugag u 214621RNAHomo Sapiens
46cagcagcaca cugugguuug u 214722RNAHomo Sapiens 47auaaagcuag
auaaccgaaa gu 224822RNAHomo Sapiens 48uauugcacuc gucccggccu cc
224922RNAHomo Sapiens 49ugauugguac gucugugggu ag
225028DNAArtificial SequenceSynthetic 50cagtcatttg gctgaggtag
taggttgt 285128DNAArtificial SequenceSynthetic 51cagtcatttg
ggtgaggtag tagattgt 285229DNAArtificial SequenceSynthetic
52cagtcatttg gaaaagtgct tacagtgca 295328DNAArtificial
SequenceSynthetic 53cagtcatttg gctaccctgt agatccga
285430DNAArtificial SequenceSynthetic 54cagtcatttg gctaccctgt
agaaccgaat 305528DNAArtificial SequenceSynthetic 55cagtcatttg
ggtggagtgt gacaatgg 285628DNAArtificial SequenceSynthetic
56cagtcatttg gctaaggcac gcggtgaa 285728DNAArtificial
SequenceSynthetic 57cagtcatttg ggtccctgag accctaac
285828DNAArtificial SequenceSynthetic 58cagtcatttg ggtcgtaccg
tgagtaat 285928DNAArtificial SequenceSynthetic 59cagtcatttg
gctcacagtg aaccggtc 286028DNAArtificial SequenceSynthetic
60cagtcatttg ggcagtgcaa tgttaaaa 286128DNAArtificial
SequenceSynthetic 61cagtcatttg gcagctggtg ttgtgaat
286228DNAArtificial SequenceSynthetic 62cagtcatttg ggtgtagtgt
ttcctact 286328DNAArtificial SequenceSynthetic 63cagtcatttg
gctgagatga agcactgt 286428DNAArtificial SequenceSynthetic
64cagtcatttg gctgagaact gaattcca 286528DNAArtificial
SequenceSynthetic 65cagtcatttg gctgagaact gaattcca
286628DNAArtificial SequenceSynthetic 66cagtcatttg gctcagtgca
tcacagaa 286728DNAArtificial SequenceSynthetic 67cagtcatttg
gctcagtgca tgacagaa 286828DNAArtificial SequenceSynthetic
68cagtcatttg gctagcagca catcatgg 286928DNAArtificial
SequenceSynthetic 69cagtcatttg gctggagaga aaggcagt
287028DNAArtificial SequenceSynthetic 70cagtcatttg ggctgaccta
tgaattga 287128DNAArtificial SequenceSynthetic 71cagtcatttg
ggaactggcc tacaaagt 287228DNAArtificial SequenceSynthetic
72cagtcatttg gctgtgcaaa tccatgca 287328DNAArtificial
SequenceSynthetic 73cagtcatttg ggtaacactg tctggtaa
287428DNAArtificial SequenceSynthetic 74cagtcatttg ggtaatactg
cctggtaa 287528DNAArtificial SequenceSynthetic 75cagtcatttg
ggtaatactg ccgggtaa 287628DNAArtificial SequenceSynthetic
76cagtcatttg gctccttcat tccaccgg 287729DNAArtificial
SequenceSynthetic 77cagtcatttg gtaaagtgct tatagtgca
297828DNAArtificial SequenceSynthetic 78cagtcatttg gctagcttat
cagactga 287928DNAArtificial SequenceSynthetic 79cagtcatttg
ggctgtgcgt gtgacagc 288028DNAArtificial SequenceSynthetic
80cagtcatttg ggagctacat tgtctgct 288128DNAArtificial
SequenceSynthetic 81cagtcatttg ggagctacat ctggctac
288228DNAArtificial SequenceSynthetic 82cagtcatttg gccattgcac
ttgtctcg 288329DNAArtificial SequenceSynthetic 83cagtcatttg
gtagcaccat ctgaaatcg 298428DNAArtificial SequenceSynthetic
84cagtcatttg gctagcacca tttgaaat 288528DNAArtificial
SequenceSynthetic 85cagtcatttg gctagcacca tttgaaat
288628DNAArtificial SequenceSynthetic 86cagtcatttg ggtgtaaaca
tcctcgac 288728DNAArtificial SequenceSynthetic 87cagtcatttg
gcaggcaaga tgctggca 288828DNAArtificial SequenceSynthetic
88cagtcatttg ggtctcacac agaaatcg 288928DNAArtificial
SequenceSynthetic 89cagtcatttg gcgctgactc ctagtcca
289028DNAArtificial SequenceSynthetic 90cagtcatttg ggaaagtgct
gcgacatt 289128DNAArtificial SequenceSynthetic 91cagtcatttg
ggtttgttcg ttcggctc 289228DNAArtificial SequenceSynthetic
92cagtcatttg gcactggact tggagtca 289328DNAArtificial
SequenceSynthetic 93cagtcatttg gcaatgacac gatcactc
289428DNAArtificial SequenceSynthetic 94cagtcatttg ggaaaccgtt
accattac 289528DNAArtificial SequenceSynthetic 95cagtcatttg
gccagcagca cactgtgg 289628DNAArtificial SequenceSynthetic
96cagtcatttg gcataaagct agataacc 289728DNAArtificial
SequenceSynthetic 97cagtcatttg ggtattgcac tcgtcccg
289821DNAArtificial SequenceSynthetic 98gcaaggatga cacgcaaatt c
219923DNAArtificial SequenceSynthetic 99ccgttttttt tttttaacca cac
2310023DNAArtificial SequenceSynthetic 100ccgttttttt tttttaacta tac
2310123DNAArtificial SequenceSynthetic 101ccgttttttt tttttctacc tgc
2310222DNAArtificial SequenceSynthetic 102cgtttttttt ttttcacaaa tt
2210322DNAArtificial SequenceSynthetic 103cgtttttttt ttttacaaat tc
2210422DNAArtificial SequenceSynthetic 104cgtttttttt ttttcaaaca cc
2210523DNAArtificial SequenceSynthetic 105ccgttttttt tttttggcat tca
2310623DNAArtificial SequenceSynthetic 106ccgttttttt tttttcacaa gtt
2310723DNAArtificial SequenceSynthetic 107ccgttttttt tttttcgcat tat
2310823DNAArtificial SequenceSynthetic 108ccgttttttt tttttaaaga gac
2310923DNAArtificial SequenceSynthetic 109ccgttttttt tttttatgcc ctt
2311022DNAArtificial SequenceSynthetic 110cgtttttttt ttttcggcct ga
2211123DNAArtificial SequenceSynthetic 111ccgttttttt tttttccata aag
2311223DNAArtificial SequenceSynthetic 112ccgttttttt tttttgagct aca
2311323DNAArtificial SequenceSynthetic 113ccgttttttt tttttaaccc atg
2311423DNAArtificial SequenceSynthetic 114ccgttttttt tttttagcct atg
2311523DNAArtificial SequenceSynthetic 115ccgttttttt tttttacaaa gtt
2311622DNAArtificial SequenceSynthetic 116cgtttttttt ttttccaagt tc
2211723DNAArtificial SequenceSynthetic 117ccgttttttt tttttgtaaa cca
2311822DNAArtificial SequenceSynthetic 118cgtttttttt ttttcaggaa ct
2211923DNAArtificial SequenceSynthetic 119ccgttttttt tttttggctg tca
2312022DNAArtificial SequenceSynthetic 120cgtttttttt ttttggctgt ca
2212123DNAArtificial SequenceSynthetic 121ccgttttttt tttttactgg gac
2312222DNAArtificial SequenceSynthetic 122cgtttttttt ttttcagttt tg
2212323DNAArtificial SequenceSynthetic 123ccgttttttt tttttaacat cgt
2312423DNAArtificial SequenceSynthetic 124ccgttttttt tttttcatca tta
2312523DNAArtificial SequenceSynthetic 125ccgttttttt tttttccatc att
2312622DNAArtificial SequenceSynthetic 126cgtttttttt ttttccatca tt
2212722DNAArtificial SequenceSynthetic 127cgtttttttt ttttcagact cc
2212823DNAArtificial SequenceSynthetic 128ccgttttttt tttttctacc tgc
2312923DNAArtificial SequenceSynthetic 129ccgttttttt tttttcaaca tca
2313022DNAArtificial SequenceSynthetic 130cgtttttttt ttttcagccg ct
2213122DNAArtificial SequenceSynthetic 131cgtttttttt ttttgaaacc ca
2213225DNAArtificial SequenceSynthetic 132atccgttttt tttttttacc
cagta 2513323DNAArtificial SequenceSynthetic 133ccgttttttt
tttttcagac cga 2313423DNAArtificial SequenceSynthetic 134ccgttttttt
tttttaaccg att 2313524DNAArtificial SequenceSynthetic 135tccgtttttt
ttttttaaca ctga 2413623DNAArtificial SequenceSynthetic
136ccgttttttt tttttaaccg att 2313723DNAArtificial SequenceSynthetic
137ccgttttttt tttttcttcc agt 2313823DNAArtificial SequenceSynthetic
138ccgttttttt tttttagcta tgc 2313923DNAArtificial SequenceSynthetic
139ccgttttttt tttttacggg tgc 2314022DNAArtificial SequenceSynthetic
140cgtttttttt ttttgagccc tg 2214123DNAArtificial SequenceSynthetic
141ccgttttttt tttttacgct caa 2314223DNAArtificial SequenceSynthetic
142ccgttttttt tttttcacgc gag 2314323DNAArtificial SequenceSynthetic
143ccgttttttt tttttccttc tga 2314422DNAArtificial SequenceSynthetic
144cgtttttttt ttttcaacgg ga 2214524DNAArtificial SequenceSynthetic
145tccgtttttt ttttttactc agta 2414623DNAArtificial
SequenceSynthetic 146ccgttttttt tttttacaaa cca 2314723DNAArtificial
SequenceSynthetic 147ccgttttttt tttttacttt cgg 2314822DNAArtificial
SequenceSynthetic 148cgtttttttt ttttggaggc cg 2214918DNAArtificial
SequenceSynthetic 149aatatggaac gcttcacg 1815028DNAArtificial
SequenceSynthetic 150cagtcatttg gctgattggt acgtctgt
2815123DNAArtificial SequenceSynthetic 151ccgttttttt tttttctacc cac
2315222RNAHomo Sapiens 152agagguagua gguugcauag uu 2215322RNAHomo
Sapiens 153ugagguagga gguuguauag uu 2215422RNAHomo Sapiens
154ugagguagua aguuguauug uu 2215522RNAHomo Sapiens 155ugagguagua
gguuguauag uu 2215622RNAHomo Sapiens 156ugagguagua gguuguaugg uu
2215722RNAHomo Sapiens 157ugagguagua guuuguacag uu 2215822RNAHomo
Sapiens 158ugagguagua guuugugcug uu 2215922RNAHomo Sapiens
159caaagugccu cccuuuagag ug 2216023RNAHomo Sapiens 160caaagugcuc
auagugcagg uag 2316123RNAHomo Sapiens 161caaagugcug uucgugcagg uag
2316223RNAHomo Sapiens 162caaagugcuu acagugcagg uag 2316322RNAHomo
Sapiens 163gaaagugcuu ccuuuuagag gc 2216421RNAHomo Sapiens
164uaaagugcug acagugcaga u 2116523RNAHomo Sapiens 165uacccuguag
aaccgaauuu gug 2316621RNAHomo Sapiens 166uaaggcaccc uucugaguag a
2116724RNAHomo Sapiens 167ucccugagac ccuuuaaccu guga 2416823RNAHomo
Sapiens 168cagugcaaua guauugucaa agc 2316923RNAHomo Sapiens
169cagugcaaug auauugucaa agc 2317022RNAHomo Sapiens
170cagugcaaug augaaagggc au 2217123RNAHomo Sapiens 171uagugcaaua
uugcuuauag ggu 2317222RNAHomo Sapiens 172ucagugcacu acagaacuuu gu
2217322RNAHomo Sapiens 173cagcagcaau ucauguuuug aa 2217421RNAHomo
Sapiens 174uagcagcaca gaaauauugg c 2117522RNAHomo Sapiens
175uagcagcaca uaaugguuug ug 2217622RNAHomo Sapiens 176uagcagcacg
uaaauauugg cg 2217721RNAHomo Sapiens 177augaccuaug aauugacaga c
2117822RNAHomo Sapiens 178aacuggcccu caaagucccg cu 2217923RNAHomo
Sapiens 179ugugcaaauc uaugcaaaac uga 2318022RNAHomo Sapiens
180uaacacuguc ugguaaagau gg 2218122RNAHomo Sapiens 181uaauacuguc
ugguaaaacc gu 2218222RNAHomo Sapiens 182gagcuuauuc auaaaagugc ag
2218321RNAHomo Sapiens 183agcuacaucu ggcuacuggg u 2118422RNAHomo
Sapiens 184aauugcacgg uauccaucug ua 2218522RNAHomo Sapiens
185aauugcacuu uagcaauggu ga 2218622RNAHomo Sapiens 186uauugcacau
uacuaaguug ca 2218722RNAHomo Sapiens 187uauugcacuu gucccggccu gu
2218822RNAHomo Sapiens 188uguaaacauc cccgacugga ag 2218922RNAHomo
Sapiens 189uguaaacauc cuacacucag cu 2219023RNAHomo Sapiens
190uguaaacauc cuacacucuc agc 2319122RNAHomo Sapiens 191uguaaacauc
cuugacugga ag 2219222RNAHomo Sapiens 192aaagugcuuc ccuuuggacu gu
2219321RNAHomo Sapiens 193aaagugcuuc cuuuuagagg g 2119422RNAHomo
Sapiens 194aaagugcuuc cuuuuagagg gu 2219521RNAHomo Sapiens
195aaagugcuuc cuuuuugagg g 2119622RNAHomo Sapiens 196aaagugcuuc
ucuuuggugg gu 2219723RNAHomo Sapiens 197gaagugcuuc gauuuugggg ugu
2319817RNAHomo Sapiens 198uaagugcuuc caugcuu 1719923RNAHomo Sapiens
199uaagugcuuc cauguuucag ugg 2320023RNAHomo Sapiens 200uaagugcuuc
cauguuugag ugu 2320123RNAHomo Sapiens 201uaagugcuuc cauguuuuag uag
2320223RNAHomo Sapiens 202uaagugcuuc cauguuuugg uga 2320322RNAHomo
Sapiens 203acuggacuua gggucagaag gc 2220421RNAHomo Sapiens
204cuauacaauc uacugucuuu c 2120522RNAHomo Sapiens 205cuauacaacc
uacugccuuc cc 2220622RNAHomo Sapiens 206cuauacgacc ugcugccuuu cu
2220722RNAHomo Sapiens 207cuauacaauc uauugccuuc cc 2220822RNAHomo
Sapiens 208cuauacaguc uacugucuuu cc 2220922RNAHomo Sapiens
209aguuuugcag guuugcauuu ca 2221022RNAHomo Sapiens 210acuguaguau
gggcacuucc ag 2221122RNAHomo Sapiens 211cggguggauc acgaugcaau uu
2221222RNAHomo Sapiens 212ggguggggau uuguugcauu ac 2221322RNAHomo
Sapiens 213cugcaaugua agcacuucuu ac 2221423RNAHomo Sapiens
214uaaggugcau cuagugcagu uag 2321522RNAHomo Sapiens 215ugcccuaaau
gccccuucug gc 2221622RNAHomo Sapiens 216caaauucgua ucuaggggaa ua
2221722RNAHomo Sapiens 217acagauucga uucuagggga au 2221822RNAHomo
Sapiens 218aacgccauua ucacacuaaa ua 2221922RNAHomo Sapiens
219cguguucaca gcggaccuug au 2222022RNAHomo Sapiens 220acggguuagg
cucuugggag cu 2222122RNAHomo Sapiens 221ucacaaguca ggcucuuggg ac
2222222RNAHomo Sapiens 222aacccguaga uccgaucuug ug 2222322RNAHomo
Sapiens 223caagcucgcu ucuauggguc ug 2222422RNAHomo Sapiens
224aacccguaga uccgaacuug ug 2222522RNAHomo Sapiens 225caagcuugua
ucuauaggua ug 2222622RNAHomo Sapiens 226uagaguuaca cccugggagu ua
2222721RNAHomo Sapiens 227cauuauuacu uuugguacgc g 2122822RNAHomo
Sapiens 228uucacauugu gcuacugucu gc 2222922RNAHomo Sapiens
229gcuacuucac aacaccaggg cc 2223022RNAHomo Sapiens 230gcuauuucac
gacaccaggg uu 2223121RNAHomo Sapiens 231cauaaaguag aaagcacuac u
2123222RNAHomo Sapiens 232ggugcagugc ugcaucucug gu 2223323RNAHomo
Sapiens 233guccaguuuu cccaggaauc ccu 2323422RNAHomo Sapiens
234ggauuccugg aaauacuguu cu 2223522RNAHomo Sapiens 235ccucugaaau
ucaguucuuc ag 2223622RNAHomo Sapiens 236ugcccugugg acucaguucu gg
2223722RNAHomo Sapiens 237aaguucuguu auacacucag gc 2223822RNAHomo
Sapiens 238cgaaucauua uuugcugcuc ua 2223922RNAHomo Sapiens
239ccaauauuac ugugcugcuu ua 2224022RNAHomo Sapiens 240aggggcuggc
uuuccucugg uc 2224118RNAHomo Sapiens 241acguuggcuc ugguggug
1824222RNAHomo Sapiens 242cugccaauuc cauaggucac ag 2224322RNAHomo
Sapiens 243uguaacagca acuccaugug ga 2224422RNAHomo Sapiens
244ccaguggggc ugcuguuauc ug 2224522RNAHomo Sapiens 245ugggucuuug
cgggcgagau ga 2224622RNAHomo Sapiens 246uaaugccccu aaaaauccuu au
2224722RNAHomo Sapiens 247aguuuugcau aguugcacua ca 2224823RNAHomo
Sapiens 248uaaggugcau cuagugcaga uag 2324923RNAHomo Sapiens
249acugcccuaa gugcuccuuc ugg 2325022RNAHomo Sapiens 250acugcaguga
aggcacuugu ag 2225122RNAHomo Sapiens 251acugcauuau gagcacuuaa ag
2225223RNAHomo Sapiens 252agguugggau cgguugcaau gcu 2325323RNAHomo
Sapiens 253aguuuugcag guuugcaucc agc 2325422RNAHomo Sapiens
254caucuuacug ggcagcauug ga 2225522RNAHomo Sapiens 255caucuuaccg
gacagugcug ga 2225622RNAHomo Sapiens 256cgucuuaccc agcaguguuu gg
2225722RNAHomo Sapiens 257caucuuccag uacaguguug ga 2225821RNAHomo
Sapiens 258caacaccagu cgaugggcug u 2125922RNAHomo Sapiens
259accuggcaua caauguagau uu 2226022RNAHomo Sapiens 260cucaguagcc
aguguagauc cu 2226121RNAHomo Sapiens 261aggcggagac uugggcaauu g
2126222RNAHomo Sapiens 262acugcugagc uagcacuucc cg 2226322RNAHomo
Sapiens 263ccgcacugug gguacuugcu gc 2226422RNAHomo Sapiens
264acugauuucu uuugguguuc ag 2226524RNAHomo Sapiens 265gcugguuuca
uauggugguu uaga 2426622RNAHomo Sapiens 266cugguuucac augguggcuu ag
2226722RNAHomo Sapiens 267ugaccgauuu cuccuggugu uc 2226822RNAHomo
Sapiens 268cuuucagucg gauguuugca gc 2226922RNAHomo Sapiens
269cugggagaag gcuguuuacu cu 2227022RNAHomo Sapiens 270ugcuaugcca
acauauugcc au 2227121RNAHomo Sapiens 271aggggugcua ucugugauug a
2127223RNAHomo Sapiens 272aagugccgcc aucuuuugag ugu 2327320RNAHomo
Sapiens 273acucaaacug ugggggcacu 2027422RNAHomo Sapiens
274acucaaaaug ggggcgcuuu cc 2227522RNAHomo Sapiens 275cuccugacuc
cagguccugu gu 2227622RNAHomo Sapiens 276aucgggaaug ucguguccgc cc
2227723RNAHomo Sapiens 277caacggaauc ccaaaagcag cug 2327822RNAHomo
Sapiens 278gcugcgcuug gauuucgucc cc 2227920RNAHomo Sapiens
279uacaguauag augauguacu 2028022RNAHomo Sapiens 280ggauaucauc
auauacugua ag 2228122RNAHomo Sapiens 281ccaauauugg cugugcugcu cc
2228222RNAHomo Sapiens 282caaaccacac ugugguguua ga 2228323RNAHomo
Sapiens 283ucuuugguua ucuagcugua uga 2328422RNAHomo Sapiens
284agggacggga cgcggugcag ug 2228518RNAHomo Sapiens 285ugaggcagua
gauugaau 1828623RNAHomo Sapiens 286uacccuguag aaccgaauuu gug
2328746DNAArtificial SequenceSyntheticmisc_feature(46)..(46)n is a,
c, g or tmisc_feature(45)..(45)v is a, c or g 287gcgagcacag
aattaatacg actcactatc ggtttttttt ttttvn 4628822DNAArtificial
SequenceSynthetic 288gcgagcacag aattaatacg ac 22
* * * * *