U.S. patent application number 13/446820 was filed with the patent office on 2013-07-11 for systems and methods for de-risking patient treatment.
This patent application is currently assigned to Molecular Health. The applicant listed for this patent is Stephan Brock, David Jackson, Theodoros Soldatos, Guillaume Taglang, Alexander Zien. Invention is credited to Stephan Brock, David Jackson, Theodoros Soldatos, Guillaume Taglang, Alexander Zien.
Application Number | 20130179187 13/446820 |
Document ID | / |
Family ID | 48744491 |
Filed Date | 2013-07-11 |
United States Patent
Application |
20130179187 |
Kind Code |
A1 |
Jackson; David ; et
al. |
July 11, 2013 |
SYSTEMS AND METHODS FOR DE-RISKING PATIENT TREATMENT
Abstract
The present disclosure describes systems and methods for
de-risking patient treatment by identifying medications or
combinations of medications to be contraindicated for a specific
indication. An analyzer executed by a processor of a computing
device from a user may receive an identification of an indication
(e.g. the subject of a clinical trial, or the diagnosis of a
patient visiting a physician's office). The analyzer may retrieve,
from an adverse event database, medication and co-medication
information of patients that experienced a side effect
corresponding to the indication. The analyzer may sort the
retrieved medication and co-medication information to generate an
ordered list of medications consumed by patients that experienced
the side effect, and identify a first medication of the ordered
list. A display module executed by the computing device may
display, to the user, the first medication of the ordered list for
contraindication from the clinical trial.
Inventors: |
Jackson; David; (Heidelberg,
DE) ; Soldatos; Theodoros; (Heidelberg, DE) ;
Taglang; Guillaume; (Heidelberg, DE) ; Zien;
Alexander; (Heidelberg, DE) ; Brock; Stephan;
(Heidelberg, DE) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Jackson; David
Soldatos; Theodoros
Taglang; Guillaume
Zien; Alexander
Brock; Stephan |
Heidelberg
Heidelberg
Heidelberg
Heidelberg
Heidelberg |
|
DE
DE
DE
DE
DE |
|
|
Assignee: |
Molecular Health
|
Family ID: |
48744491 |
Appl. No.: |
13/446820 |
Filed: |
April 13, 2012 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61584164 |
Jan 6, 2012 |
|
|
|
61605625 |
Mar 1, 2012 |
|
|
|
Current U.S.
Class: |
705/3 |
Current CPC
Class: |
G16H 20/10 20180101;
G16H 50/20 20180101; G16B 20/00 20190201; G16H 70/40 20180101; G16H
50/50 20180101 |
Class at
Publication: |
705/3 |
International
Class: |
G06Q 50/24 20120101
G06Q050/24 |
Claims
1. A method for identifying a medication for contraindication from
an indication, comprising: receiving, by an analyzer executed by a
processor of a computing device from a user, an identification of
an indication of a first patient; retrieving, by the analyzer from
an adverse event database, medication and co-medication information
of patients that experienced a side effect corresponding to the
indication; sorting the retrieved medication and co-medication
information, by the analyzer, to generate an ordered list of
medications consumed by patients that experienced the side effect;
identifying, by the analyzer, a first medication of the ordered
list; and displaying, by a display module executed by the computing
device to the user, the first medication of the ordered list for
contraindication for the first patient.
2. The method of claim 1, further comprising determining, by the
analyzer, that an organ is associated with the indication; and
wherein retrieving medication and co-medication information of
patients that experienced the side effect further comprises
extracting a subset from the retrieved information of medications
and co-medications identified as affecting the organ.
3. The method of claim 1, further comprising identifying, by the
analyzer, a molecular interaction associated with the side effect;
and identifying the first medication, responsive to the first
medication identified in a medication information database as
affecting the identified molecular interaction.
4. The method of claim 1, wherein identifying a first medication
comprises determining a proportional reporting ratio of the first
medication to all medications in the ordered list, and identifying
the first medication, responsive to the proportional reporting
ratio being above a predetermined threshold.
5. The method of claim 1, wherein sorting the retrieved medication
and co-medication information comprises scoring each medication in
the list responsive to its frequency of appearance in the retrieved
medication and co-medication information, and sorting the list by
score.
6. The method of claim 1, further comprising identifying, by the
analyzer, a combination of a second medication and third medication
appearing together in the retrieved medication and co-medication
information at a statistical rate above a predetermined threshold;
and displaying, by the display module, the combination of the
second medication and third medication for contraindication from
the clinical trial.
7. A system for identifying a medication for contraindication from
an indication, comprising: a computing device comprising a
processor executing an analyzer configured for receiving, from a
user, an identification of an indication of a first patient,
retrieving, from an adverse event database, medication and
co-medication information of patients that experienced a side
effect corresponding to the indication, sorting the retrieved
medication and co-medication information to generate an ordered
list of medications consumed by patients that experienced the side
effect, and identifying, by the analyzer, a first medication of the
ordered list; and a display module configured for displaying the
first medication of the ordered list for contraindication from the
first patient.
8. The system of claim 7, wherein the analyzer is further
configured for determining that an organ is associated with the
indication; and wherein retrieving medication and co-medication
information of patients that experienced the side effect further
comprises extracting a subset from the retrieved information of
medications and co-medications identified as affecting the
organ.
9. The system of claim 7, wherein the analyzer is further
configured for identifying a molecular interaction associated with
the side effect; and identifying the first medication, responsive
to the first medication identified in a medication information
database as affecting the identified molecular interaction.
10. The system of claim 7, wherein the analyzer is further
configured for determining a proportional reporting ratio of the
first medication to all medications in the ordered list, and
identifying the first medication, responsive to the proportional
reporting ratio being above a predetermined threshold.
11. The system of claim 7, wherein the analyzer is further
configured for scoring each medication in the list responsive to
its frequency of appearance in the retrieved medication and
co-medication information, and sorting the list by score.
12. The system of claim 7, wherein the analyzer is further
configured for identifying a combination of a second medication and
third medication appearing together in the retrieved medication and
co-medication information at a statistical rate above a
predetermined threshold; and the display module is further
configured for displaying the combination of the second medication
and third medication for contraindication from the clinical
trial.
13. A computer readable storage device comprising computer-readable
instructions for identifying a medication for contraindication from
an indication, comprising: instructions for receiving, by an
analyzer executed by a processor of a computing device from a user,
an identification of an indication of a first patient; instructions
for retrieving, by the analyzer from an adverse event database,
medication and co-medication information of patients that
experienced a side effect corresponding to the indication;
instructions for sorting the retrieved medication and co-medication
information, by the analyzer, to generate an ordered list of
medications consumed by patients that experienced the side effect;
instructions for identifying, by the analyzer, a first medication
of the ordered list; and instructions for displaying, by a display
module executed by the computing device to the user, the first
medication of the ordered list for contraindication from the first
patient.
14. The computer readable storage device of claim 13, further
comprising instructions for determining, by the analyzer, that an
organ is associated with the indication; and wherein retrieving
medication and co-medication information of patients that
experienced the side effect further comprises extracting a subset
from the retrieved information of medications and co-medications
identified as affecting the organ.
15. The computer readable storage device of claim 13, further
comprising instructions for identifying, by the analyzer, a
molecular interaction associated with the side effect; and
identifying the first medication, responsive to the first
medication identified in a medication information database as
affecting the identified molecular interaction.
16. The computer readable storage device of claim 13, further
comprising instructions for determining a proportional reporting
ratio of the first medication to all medications in the ordered
list, and identifying the first medication, responsive to the
proportional reporting ratio being above a predetermined
threshold.
17. The computer readable storage device of claim 13, further
comprising instructions for scoring each medication in the list
responsive to its frequency of appearance in the retrieved
medication and co-medication information, and sorting the list by
score.
18. The computer readable storage device of claim 13, further
comprising instructions for identifying, by the analyzer, a
combination of a second medication and third medication appearing
together in the retrieved medication and co-medication information
at a statistical rate above a predetermined threshold; and
displaying, by the display module, the combination of the second
medication and third medication for contraindication from the
clinical trial.
Description
RELATED APPLICATIONS
[0001] The present application claims priority to and the benefit
of U.S. Provisional Patent Application No. 61/584,164, entitled
"Translating Clinico-Molecular Data Into Safer, More Effective Drug
Choices," filed Jan. 6, 2012, and U.S. Provisional Patent
Application No. 61/605,625, entitled "Systems and Methods for
Analysis of Adverse Event Data," filed Mar. 1, 2012, each of which
are hereby incorporated by reference in their entirety.
FIELD OF THE INVENTION
[0002] The present disclosure relates to systems and methods for
bioinformatics and data processing. In particular, the present
disclosure relates to methods and systems for de-risking patient
treatment or clinical trials.
BACKGROUND OF THE INVENTION
[0003] Adverse event data from adverse event reporting systems
(AERS) such as those maintained by the U.S. Food and Drug
Association may be useful in statistically identifying potential
drug hazards. However, analysis of such data is typically limited
to simple univariate analysis, such as rates of adverse events
associated with a medication. Such analysis may fail to examine
other factors and associations between medications or relationships
between molecular entities associated with the medications, such as
target (and off-target) proteins, enzymes, transporters, pathways,
drug classes, or other information.
BRIEF SUMMARY OF THE INVENTION
[0004] In one aspect, the present disclosure is directed to systems
and methods for analysis of adverse event data. Adverse event data
may be integrated with data regarding drug targets, classes of
drugs or therapeutic categories, indications, target proteins,
metabolizing enzymes or pathways, and may be analyzed on a
molecular basis. Deciphering the molecular basis of such adverse
responses is not only paramount to the protection of patient
well-being and the development of safer drugs, but it also presents
a unique opportunity to dissect disease systems in search of novel
predictive biomarkers, drug targets and efficacious combination
therapies.
[0005] In another aspect, the present disclosure is directed to
systems and methods for identifying treatment strategies based on
integrating drug molecular data and patient genome sequencing data
with critical clinical information about the patient. Disaggregated
data may be combined and translated into evidence-based treatment
strategies for marketed and clinical stage therapies.
[0006] In still another aspect, the present disclosure is directed
to systems and methods for clinical trial design based on
integrated molecular data regarding adverse events, drug targets,
classes of drugs or therapeutic categories, indications, target
proteins, metabolizing enzymes or pathways, and may be analyzed on
a molecular basis. Through analysis of adverse events at the level
of drug target proteins, pathways, or metabolizing enzymes, trials
may be designed to focus on specific adverse events while reducing
false positives or negatives through drug interaction at the
protein, pathway, or enzyme level. In some embodiments, adverse
events for new drugs in development may be predicted through
analysis of adverse event data for drugs with similar molecular
interactions or targets.
[0007] Accordingly, in some embodiments, the systems and methods
discussed herein may allow: [0008] Integration of all
patient-specific clinical information and molecular testing results
into a single decision support framework; [0009] Automated patient
genome analysis and functional prioritization of variants; [0010]
Conversion and visualizations of clinical data and patient-specific
therapeutic system models; [0011] Conversion of clinical data into
an easy-to-view representation of a patient's treatment history;
[0012] Identification of off-target safety, resistance, or other
clinical effects (e.g. improved response, lower death rate, etc.)
via analysis of the molecular basis of adverse events; [0013]
Safety signal detection and analysis of potentially causative
molecular mechanisms; [0014] Analysis of adverse events data for
drugs, drug classes, targets, or pathways; [0015] Integration of
adverse event reports with relevant clinical and molecular
knowledge; and [0016] Capturing of proprietary outcomes data,
permitting novel insights into clinical trial and adverse drug
event management program results.
[0017] In one aspect, the present disclosure is directed to systems
and methods for analysis of adverse event data. Adverse event data
may be integrated with data regarding drug targets, classes of
drugs or therapeutic categories, indications, target proteins,
metabolizing enzymes or pathways, and may be analyzed on a
molecular basis. Deciphering the molecular basis of such adverse
responses is not only paramount to the protection of patient
well-being and the development of safer drugs, but it also presents
a unique opportunity to dissect disease systems in search of novel
biomarkers, drug targets and efficacious combination therapies.
Adverse event information may be combined with clinico-molecular
knowledge about drug activity within a patient. A user, drug
manufacturer, patient, or medical service provider may explore and
analyze adverse event information from both statistical and
molecular perspectives. In some embodiments, the system may
comprise analytical and visualization tools supporting the
expedited detection and validation of drug-related safety
science.
[0018] In another aspect, the present disclosure is directed to
systems and methods for identifying treatment strategies based on
integrating drug molecular data and patient genome sequencing data
with critical clinical information about the patient. Disaggregated
data may be combined and translated into evidence-based treatment
strategies for marketed and clinical stage therapies.
[0019] In still another aspect, the present disclosure is directed
to systems and methods for clinical trial design based on
integrated molecular data regarding adverse events, drug targets,
classes of drugs or therapeutic categories, indications, target
proteins, metabolizing enzymes or pathways, and may be analyzed on
a molecular basis. Through analysis of adverse events at the level
of drug target proteins, pathways, or metabolizing enzymes, trials
may be designed to avoid specific adverse events while reducing
false positives or negatives through drug interaction at the
protein, pathway, or enzyme level. In some embodiments, adverse
events for new drugs in development may be predicted through
analysis of adverse event data for drugs with similar metabolic
interactions or targets.
[0020] In yet another aspect, the present disclosure describes
systems and methods for reducing false signals in planned clinical
trials by identifying medications to be contraindicated for a
cohort. For example, in many instances, a disease and a side effect
may differ only due to the side effect being drug-induced.
Accordingly, the side effect may be thought of as a drug-induced
disease. For manufacturers and researchers developing new
pharmaceuticals, it may be important during trials to avoid
including patients taking other drugs that may induce the same side
effect as the disease in question. Furthermore, it may be desirable
to screen all patient co-medications for drug interactions at many
levels, including on a molecular basis.
[0021] In one aspect, the present disclosure is directed to a
method for identifying a medication for contraindication from an
indication, such as a disease that is the subject of a clinical
trial of another medication, or a diagnosis of a patient by a
physician. The method includes receiving, by an analyzer executed
by a processor of a computing device from a user, an identification
of an indication of a first patient, such as an indication that is
the subject of a clinical trial or an indication identified by a
physician as experienced by the first patient. The method also
includes retrieving, by the analyzer from an adverse event
database, medication and co-medication information of patients that
experienced a side effect corresponding to the indication. The
method further includes sorting the retrieved medication and
co-medication information, by the analyzer, to generate an ordered
list of medications consumed by patients that experienced the side
effect. The method also includes identifying, by the analyzer, a
first medication of the ordered list. The method also includes
displaying, by a display module executed by the computing device to
the user, a first medication of the ordered list for
contraindication from the patient or from the clinical trial.
[0022] In one embodiment, the method includes determining, by the
analyzer, that an organ is associated with the indication; and
retrieving medication and co-medication information of patients
that experienced the side effect further comprises extracting a
subset from the retrieved information of medications and
co-medications identified as affecting the organ. In another
embodiment, the method includes identifying, by the analyzer, a
molecular interaction associated with the side effect; and
identifying the first medication, responsive to the first
medication identified in a medication information database as
affecting the identified molecular interaction.
[0023] In still another embodiment, the method includes determining
a proportional reporting ratio of the first medication to all
medications in the ordered list, and identifying the first
medication, responsive to the proportional reporting ratio being
above a predetermined threshold. In yet still another embodiment,
the method includes scoring each medication in the list responsive
to its frequency of appearance in the retrieved medication and
co-medication information, and sorting the list by score. In some
embodiments, the method includes identifying, by the analyzer, a
combination of a second medication and third medication appearing
together in the retrieved medication and co-medication information
at a statistical rate above a predetermined threshold; and
displaying, by the display module, the combination of the second
medication and third medication for contraindication from the
patient or from the clinical trial.
[0024] In another aspect, the present disclosure is directed to a
system for identifying a medication for contraindication from an
indication, such as a disease that is the subject of a clinical
trial of another medication, or a diagnosis of a patient by a
physician. The system includes a computing device comprising a
processor executing an analyzer and a display module. The analyzer
is configured for receiving, from a user, an identification of an
indication of a first patient, such as an indication that is the
subject of a clinical trial or an indication identified by a
physician as experienced by the first patient. The analyzer is
further configured for retrieving, from an adverse event database,
medication and co-medication information of patients that
experienced a side effect corresponding to the indication. The
analyzer is also configured for sorting the retrieved medication
and co-medication information to generate an ordered list of
medications consumed by patients that experienced the side effect,
and identifying a first medication of the ordered list. The display
module is configured for displaying a first medication of the
ordered list for contraindication from the patient or from the
clinical trial.
[0025] In some embodiments, the analyzer is further configured for
determining that an organ is associated with the indication; and
retrieving medication and co-medication information of patients
that experienced the side effect further comprises extracting a
subset from the retrieved information of medications and
co-medications identified as affecting the organ. In other
embodiments, the analyzer is further configured for identifying a
molecular interaction associated with the side effect; and
identifying the first medication, responsive to the first
medication identified in a medication information database as
affecting the identified molecular interaction.
[0026] In one embodiment, the analyzer is further configured for
determining a proportional reporting ratio of the first medication
to all medications in the ordered list, and identifying the first
medication, responsive to the proportional reporting ratio being
above a predetermined threshold. In another embodiment, the
analyzer is further configured for scoring each medication in the
list responsive to its frequency of appearance in the retrieved
medication and co-medication information, and sorting the list by
score. In still another embodiment, the analyzer is further
configured for identifying a combination of a second medication and
third medication appearing together in the retrieved medication and
co-medication information at a statistical rate above a
predetermined threshold; and the display module is further
configured for displaying the combination of the second medication
and third medication for contraindication from the clinical
trial.
[0027] In still another aspect, the present disclosure is directed
to a computer readable storage device comprising computer-readable
instructions for identifying a medication for contraindication from
an indication, such as a disease that is the subject of a clinical
trial of another medication, or a diagnosis of a patient by a
physician. The storage device includes instructions for receiving,
by an analyzer executed by a processor of a computing device from a
user, an identification of an indication of a first patient, such
as an indication that is the subject of a clinical trial or an
indication identified by a physician as experienced by the first
patient. The storage device also includes instructions for
retrieving, by the analyzer from an adverse event database,
medication and co-medication information of patients that
experienced a side effect corresponding to the indication. The
storage device further includes instructions for sorting the
retrieved medication and co-medication information, by the
analyzer, to generate an ordered list of medications consumed by
patients that experienced the side effect. The storage device also
includes instructions for identifying, by the analyzer, a first
medication of the ordered list. The storage device also includes
instructions for displaying, by a display module executed by the
computing device to the user, a first medication of the ordered
list for contraindication from the clinical trial.
[0028] In some embodiments, the storage device includes
instructions for determining, by the analyzer, that an organ is
associated with the indication; and extracting a subset from the
retrieved information of medications and co-medications identified
as affecting the organ. In other embodiments, the storage device
includes instructions for identifying, by the analyzer, a molecular
interaction associated with the side effect; and identifying the
first medication, responsive to the first medication identified in
a medication information database as affecting the identified
molecular interaction.
[0029] In one embodiment, the storage device includes instructions
for determining a proportional reporting ratio of the first
medication to all medications in the ordered list, and identifying
the first medication, responsive to the proportional reporting
ratio being above a predetermined threshold. In another embodiment,
the storage device includes instructions for scoring each
medication in the list responsive to its frequency of appearance in
the retrieved medication and co-medication information, and sorting
the list by score. In still another embodiment, the storage device
includes instructions for identifying, by the analyzer, a
combination of a second medication and third medication appearing
together in the retrieved medication and co-medication information
at a statistical rate above a predetermined threshold; and
displaying, by the display module, the combination of the second
medication and third medication for contraindication from the
clinical trial.
[0030] The details of various embodiments of the invention are set
forth in the accompanying drawings and the description below.
BRIEF DESCRIPTION OF THE FIGURES
[0031] The foregoing and other objects, aspects, features, and
advantages of the disclosure will become more apparent and better
understood by referring to the following description taken in
conjunction with the accompanying drawings, in which:
[0032] FIG. 1A is a block diagram depicting relationships between
data provided by embodiments of an adverse event reporting
system;
[0033] FIG. 1B is a block diagram depicting relationships between
molecular entities in an embodiment of a multivariate analysis
system;
[0034] FIG. 2A is a block diagram depicting an embodiment of a
network environment comprising local machines in communication with
remote machines;
[0035] FIGS. 2B-2E are block diagrams depicting embodiments of
computers useful in connection with the methods and systems
described herein;
[0036] FIG. 3A is a block diagram of an embodiment of a system for
multivariate analysis of adverse event data;
[0037] FIG. 3B is a diagram of an example embodiment of a global
molecular entity graph;
[0038] FIG. 3C is a diagram of an example embodiment of extracted
subgraphs;
[0039] FIG. 4A is a diagram of an embodiment of method for
identifying molecular entities responsible for adverse event
differences between similar indications;
[0040] FIG. 4B is a flow chart of an embodiment of method for
identifying molecular entities responsible for adverse event
differences between similar indications;
[0041] FIG. 4C is a flow chart of an embodiment of a method for
retrieving an ordered list of medications for an indication and
adverse event;
[0042] FIG. 5A is a diagram of another embodiment of a global
molecular entity graph;
[0043] FIG. 5B is a flow diagram of an embodiment of a method for
extracting an indication-specific model from a global molecular
entity graph;
[0044] FIG. 5C is another diagram of another embodiment of a global
molecular entity graph;
[0045] FIG. 5D is a flow diagram of an embodiment of a method for
examining side effects associated with activating a pathway vs.
inactivating the pathway;
[0046] FIG. 6A is a diagram of a method of utilizing side effect
profile dissimilarities to identify likely unknown targets of a
medication;
[0047] FIG. 6B is a flow chart of an embodiment of a method for
identifying unknown likely targets of a first medication via
comparison of adverse event data;
[0048] FIG. 7A-7C are screenshots of an example of embodiments of a
molecular entity dependency graph that provides intuitive
identification of redundancies and molecular interactions between
medications in a patient's prescription load;
[0049] FIG. 8 is a flow chart of an embodiment of a method for
personalized de-risking of medications based on genomic information
of a patient and adverse event data of combination therapies;
[0050] FIG. 9 is a flow chart of an embodiment of a method for
identifying a medication for contraindication from a clinical trial
of another medication;
[0051] FIG. 10A is a Venn diagram of an example of an embodiment of
defining cohorts within adverse event data and extracting
difference profiles for a cohort;
[0052] FIG. 10B is a flow chart of an embodiment of a method for
identifying potential combination therapies for research via
adverse event data;
[0053] FIG. 11A is a graph of an example of a region of an example
embodiment of a global molecular entity graph or molecular entity
network comprising a plurality of molecular entities 1106 connected
via functional links;
[0054] FIG. 11B is a flow chart of an embodiment of a method for
generating a predicted side effect profile for a medication
targeting a novel target;
[0055] FIG. 12A is a block diagram of an embodiment of a process
for using genomic information to identify protein targets
responsible for adverse events;
[0056] FIG. 12B is a flow chart of an embodiment of a method of
identifying genetic variants associated with adverse events;
[0057] FIGS. 13A-13Y are screenshots of an example embodiment of an
interface for analyzing adverse event data; and
[0058] FIGS. 14A-14C are screenshots of an example embodiment of
comparison of side effect profiles for molecular entities.
[0059] The features and advantages of the present invention will
become more apparent from the detailed description set forth below
when taken in conjunction with the drawings, in which like
reference characters identify corresponding elements throughout. In
the drawings, like reference numbers generally indicate identical,
functionally similar, and/or structurally similar elements.
DETAILED DESCRIPTION OF THE INVENTION
[0060] Adverse events are a common and, for the most part,
unavoidable consequence of therapeutic intervention. The
identification of novel adverse events is critical to the
protection of patient well-being and the healthcare system that
supports them. From the induction of avoidable and sometimes fatal
side effects to the billions of dollars in associated medical
costs, adverse events (AE's) remain a critical issue for all
stakeholders in the healthcare system.
[0061] Data about adverse events are provided by clinicians,
researchers, and manufacturers to spontaneous reporting systems,
such as the U.S. Food and Drug Administration's Adverse Event
Reporting System (AERS). After a manual review of each submission
the data are made publically available on quarterly basis via the
online AERS data files. All reports contain information surrounding
the treatment, side effects, and patient
characteristics/demographics. Drug information is further qualified
as to whether the drug is suspected as the primary or secondary
cause of the adverse event or whether it was concomitant. However,
there are a number of considerations that limit the usefulness of
the AERS data for pharmacovigilance purposes. Traditional methods
of Adverse Drug Reaction (ADR) detection have often relied on the
manual review of drug-specific cases by clinical pharmacologists.
However, the increasing size and complexity of SRS databases, and
limitations in human resources have led to demands for more
efficient methods of ADR detection. Additionally, AERS data is
frequently difficult to use, with misspellings, abbreviations, and
inconsistent synonyms used. Furthermore, as adverse event reporting
systems focus on adverse events and drugs, detailed molecular
information is absent. For example, referring briefly to FIG. 1A,
adverse event data typically includes identifications of drugs
prescribed to a patient 102; indications 104, or diseases or
symptoms for which the drug or drugs was prescribed; reactions or
side effects 106; and outcomes 108. For example, an outcome 108 may
comprise prolonged hospitalization, short term hospitalization, or
death. Accordingly, while the data may be useful for identifying
drug-drug interactions, or performing univariate analysis, such as
the statistical percentage of patients taking a drug that had a
particular outcome when experiencing an adverse event, the data may
be limited in utility on its own.
[0062] The systems and methods discussed herein provide for
multivariate analysis of molecular entities involved with adverse
events. Referring briefly to FIG. 1B and in contradistinction from
FIG. 1A, a multivariate analyzer 120 may utilize links between not
just drugs 102, indications 104, reactions 106, and outcomes 108,
but molecular entities such as pathways 110, protein targets 112,
metabolizing enzymes or transporters 114. Drugs 102 may also be
associated with a drug class 116. This enables investigation of the
relationship between, say, a particular side effect or reaction 106
and a protein target 112, or other entity types such as protein
domains, gene ontology terms for biological processes, and other
biological, chemical, or clinical descriptors. Deciphering the
molecular basis of such adverse responses is not only paramount to
the protection of patient well-being and the development of safer
drugs, but it also presents a unique opportunity to dissect disease
systems in search of novel predictive biomarkers, drug targets and
efficacious combination therapies.
[0063] Prior to discussing specifics of methods and systems
utilizing multivariate analysis of adverse event data, it may be
helpful to briefly define a few terms as used herein. The following
definitions are not intended to be limiting, but may comprise
alternate definitions commonly utilized by those of ordinary skill
in the art. Accordingly, context may clarify whether, for example,
the term indication refers to a symptom or disease, a flag in a
database, or a selection by a user. Additionally, the following
list of definitions is not intended to be exhaustive, but rather
discuss a few key terms that may be helpful to those of skill in
the art.
[0064] Adverse event: In pharmacology, an adverse event may refer
to any unexpected or dangerous reaction to a drug. An unwanted
effect caused by the administration of a drug. The onset of the
adverse reaction may be sudden or develop over time. Also
interchangeably called: adverse drug event (ADE), adverse drug
reaction (ADR), adverse effect or adverse reaction.
[0065] Absorption, Distribution, Metabolism, Excretion (ADME):
Refers to the standard pharmacokinetic mechanism of a drug (see
Pharmacokinetics).
[0066] AERS--Adverse Event Reporting System: The Adverse Event
Reporting System (AERS) is a computerized information database
designed to support the FDA's post-marketing safety surveillance
program for all approved drug and therapeutic biologic products.
The FDA uses AERS to monitor for new adverse events and medication
errors that might occur with these marketed products.
[0067] Bioavailability: Also referred to as availability, this is
the amount of a drug that is absorbed into circulation after
administration of a specific dosage.
[0068] Challenge-dechallenge-rechallenge (CDR): This is a medical
testing protocol in which a medicine (or drug) is administered
(challenge), withdrawn (dechallenge), then re-administered
(rechallenge), while being monitored for adverse effects
(reactions) at each stage.
[0069] Contingency table (or matrix): Also referred to as cross
tabulation or cross tab. A contingency table is often used to
record and analyze the relation between two or more categorical
variables. It displays the (multivariate) frequency distribution of
the variables in a matrix format.
[0070] Drug interaction: A drug interaction is a situation in which
a substance affects the activity of a drug, i.e. the effects are
increased or decreased, or they produce a new effect that neither
produces on its own. However, interactions may also exist between
drugs & foods (drug-food interactions), as well as drugs &
herbs (drug-herb interactions). These may occur out of accidental
misuse or due to lack of knowledge about the active ingredients
involved in the relevant substances or the underlying molecular
mechanisms.
[0071] Entity Coverage/Co-Entity Coverage: The Entity Coverage is
an estimate that refers to the significance with which a first
entity (E1) is related with a second entity (E2) in a data set. It
is the calculated from the number of data entries containing E1 and
E2 divided by the overall number of data entries containing E1. The
Co-Entity Coverage is the calculated from the number of data
entries containing E1 and E2 divided by the overall number of data
entries containing E2. This method gives thus an indication for the
significance of entity relations in subsets of data.
[0072] Gamma Poisson Shrinker: Advanced method for
Pharmacovigilance Signal Detection. In contrast to simple methods
that focus on a specific AE-drug-combination at a time (encoded in
2*2 contingency tables), it can directly use contingency tables
that range over all drugs and AEs.
[0073] Idiosyncratic response: An abnormal response from a drug
that is specific to the person having the response.
[0074] Indication (or `drug use`): In medicine, an indication is a
valid reason to use a certain test, medication, procedure, or
surgery. An indication may thus refer to a disease, a symptom, or
diagnosis. The opposite of indication is contraindication.
[0075] Metabolizing enzyme: A protein that metabolizes a
medication; the enzyme may help transforming a pro-drug to its
pharmacologically active chemical compound form or it may play a
role in its degradation.
[0076] Molecular mechanism: The flow of events that take place in
the molecular level when a drug is administered. The molecular
mechanisms can be highly complex due to the variety of
participating components (e.g., drugs, organs, cells, proteins,
etc.), systems (e.g., pathways, disease networks, etc.), entity
interrelations (e.g., drug-target, drug-metabolizing enzyme,
carriers, transporters, overlapping systems and pathways, etc.),
and molecular aberrations (e.g., mutations, radiation damage,
etc.). Components of the molecular mechanism, such as protein
targets, pathways, transporters, drugs, or drug classes may be
referred to variously as molecular entities or biomolecular
entities.
[0077] Side effect: Any unintended effect of a pharmaceutical
product occurring at a dose normally used in man, which is related
to the pharmacological properties of the drug. A side effect may
frequently correspond to an indication. For example, nausea may be
a side effect of a first drug, but may be an indication to be
treated by a second drug. A negative side effect may also be
referred to as an adverse event.
[0078] Prior to discussing specifics of methods and systems for
multivariate analysis of adverse event data, it may be helpful to
briefly discuss embodiments of networks and computing devices that
may be utilized in various embodiments of these methods and
systems. Referring now to FIG. 2A, an embodiment of a network
environment is depicted. In brief overview, the network environment
comprises one or more local machines 202a-202n (also generally
referred to as local machine(s) 202, client(s) 202, client node(s)
202, client machine(s) 202, client computer(s) 202, client
device(s) 202, endpoint(s) 202, or endpoint node(s) 202) in
communication with one or more remote machines 206a-206n (also
generally referred to as server(s) 206 or remote machine(s) 206)
via one or more networks 204. In some embodiments, a local machine
202 has the capacity to function as both a client node seeking
access to resources provided by a server and as a server providing
access to hosted resources for other clients 202a-202n.
[0079] Although FIG. 2A shows a network 204 between the local
machines 202 and the remote machines 206, the local machines 202
and the remote machines 206 may be on the same network 204. The
network 204 can be a local-area network (LAN), such as a company
Intranet, a metropolitan area network (MAN), or a wide area network
(WAN), such as the Internet or the World Wide Web. In some
embodiments, there are multiple networks 204 between the local
machines 202 and the remote machines 206. In one of these
embodiments, a network 204' (not shown) may be a private network
and a network 204 may be a public network. In another of these
embodiments, a network 204 may be a private network and a network
204' a public network. In still another embodiment, networks 204
and 204' may both be private networks. In yet another embodiment,
networks 204 and 204' may both be public networks.
[0080] The network 204 may be any type and/or form of network and
may include any of the following: a point to point network, a
broadcast network, a wide area network, a local area network, a
telecommunications network, a data communication network, a
computer network, an ATM (Asynchronous Transfer Mode) network, a
SONET (Synchronous Optical Network) network, a SDH (Synchronous
Digital Hierarchy) network, a wireless network and a wireline
network. In some embodiments, the network 204 may comprise a
wireless link, such as an infrared channel or satellite band. The
topology of the network 204 may be a bus, star, or ring network
topology. The network 204 may be of any such network topology as
known to those ordinarily skilled in the art capable of supporting
the operations described herein. The network may comprise mobile
telephone networks utilizing any protocol or protocols used to
communicate among mobile devices, including AMPS, TDMA, CDMA, GSM,
GPRS or UMTS. In some embodiments, different types of data may be
transmitted via different protocols. In other embodiments, the same
types of data may be transmitted via different protocols.
[0081] In some embodiments, the system may include multiple,
logically-grouped remote machines 206. In one of these embodiments,
the logical group of remote machines may be referred to as a server
farm 38. In another of these embodiments, the remote machines 206
may be geographically dispersed. In other embodiments, a server
farm 38 may be administered as a single entity. In still other
embodiments, the server farm 38 comprises a plurality of server
farms 38. The remote machines 206 within each server farm 38 can be
heterogeneous--one or more of the remote machines 206 can operate
according to one type of operating system platform (e.g., WINDOWS
NT, WINDOWS 2003, WINDOWS 2008, WINDOWS 7 and WINDOWS Server 2008
R2, all of which are manufactured by Microsoft Corp. of Redmond,
Wash.), while one or more of the other remote machines 206 can
operate on according to another type of operating system platform
(e.g., Unix or Linux).
[0082] The remote machines 206 of each server farm 38 do not need
to be physically proximate to another remote machine 206 in the
same server farm 38. Thus, the group of remote machines 206
logically grouped as a server farm 38 may be interconnected using a
wide-area network (WAN) connection or a metropolitan-area network
(MAN) connection. For example, a server farm 38 may include remote
machines 206 physically located in different continents or
different regions of a continent, country, state, city, campus, or
room. Data transmission speeds between remote machines 206 in the
server farm 38 can be increased if the remote machines 206 are
connected using a local-area network (LAN) connection or some form
of direct connection.
[0083] A remote machine 206 may be a file server, application
server, web server, proxy server, appliance, network appliance,
gateway, application gateway, gateway server, virtualization
server, deployment server, SSL VPN server, or firewall. In some
embodiments, a remote machine 206 provides a remote authentication
dial-in user service, and is referred to as a RADIUS server. In
other embodiments, a remote machine 206 may have the capacity to
function as either an application server or as a master application
server. In still other embodiments, a remote machine 206 is a blade
server. In yet other embodiments, a remote machine 206 executes a
virtual machine providing, to a user or client computer 202, access
to a computing environment.
[0084] In one embodiment, a remote machine 206 may include an
Active Directory. The remote machine 206 may be an application
acceleration appliance. For embodiments in which the remote machine
206 is an application acceleration appliance, the remote machine
206 may provide functionality including firewall functionality,
application firewall functionality, or load balancing
functionality. In some embodiments, the remote machine 206
comprises an appliance such as one of the line of appliances
manufactured by the Citrix Application Networking Group, of San
Jose, Calif., or Silver Peak Systems, Inc., of Mountain View,
Calif., or of Riverbed Technology, Inc., of San Francisco, Calif.,
or of F5 Networks, Inc., of Seattle, Wash., or of Juniper Networks,
Inc., of Sunnyvale, Calif.
[0085] In some embodiments, a remote machine 206 executes an
application on behalf of a user of a local machine 202. In other
embodiments, a remote machine 206 executes a virtual machine, which
provides an execution session within which applications execute on
behalf of a user of a local machine 202. In one of these
embodiments, the execution session is a hosted desktop session. In
another of these embodiments, the execution session provides access
to a computing environment, which may comprise one or more of: an
application, a plurality of applications, a desktop application,
and a desktop session in which one or more applications may
execute.
[0086] In some embodiments, a local machine 202 communicates with a
remote machine 206. In one embodiment, the local machine 202
communicates directly with one of the remote machines 206 in a
server farm 38. In another embodiment, the local machine 202
executes a program neighborhood application to communicate with a
remote machine 206 in a server farm 38. In still another
embodiment, the remote machine 206 provides the functionality of a
master node. In some embodiments, the local machine 202
communicates with the remote machine 206 in the server farm 38
through a network 204. Over the network 204, the local machine 202
can, for example, request execution of various applications hosted
by the remote machines 206a-206n in the server farm 38 and receive
output of the results of the application execution for display. In
some embodiments, only a master node provides the functionality
required to identify and provide address information associated
with a remote machine 206b hosting a requested application.
[0087] In one embodiment, the remote machine 206 provides the
functionality of a web server. In another embodiment, the remote
machine 206a receives requests from the local machine 202, forwards
the requests to a second remote machine 206b and responds to the
request by the local machine 202 with a response to the request
from the remote machine 206b. In still another embodiment, the
remote machine 206a acquires an enumeration of applications
available to the local machine 202 and address information
associated with a remote machine 206b hosting an application
identified by the enumeration of applications. In yet another
embodiment, the remote machine 206 presents the response to the
request to the local machine 202 using a web interface. In one
embodiment, the local machine 202 communicates directly with the
remote machine 206 to access the identified application. In another
embodiment, the local machine 202 receives output data, such as
display data, generated by an execution of the identified
application on the remote machine 206.
[0088] In some embodiments, the remote machine 206 or a server farm
38 may be running one or more applications, such as an application
providing a thin-client computing or remote display presentation
application. In one embodiment, the remote machine 206 or server
farm 38 executes as an application any portion of the CITRIX ACCESS
SUITE by Citrix Systems, Inc., such as the METAFRAME or CITRIX
PRESENTATION SERVER products, any of the following products
manufactured by Citrix Systems, Inc.: CITRIX XENAPP, CITRIX
XENDESKTOP, CITRIX ACCESS GATEWAY, and/or any of the MICROSOFT
WINDOWS Terminal Services manufactured by the Microsoft
Corporation. In another embodiment, the application is an ICA
client, developed by Citrix Systems, Inc. of Fort Lauderdale, Fla.
In still another embodiment, the remote machine 206 may run an
application, which, for example, may be an application server
providing email services such as MICROSOFT EXCHANGE manufactured by
the Microsoft Corporation of Redmond, Wash., a web or Internet
server, or a desktop sharing server, or a collaboration server. In
yet another embodiment, any of the applications may comprise any
type of hosted service or products, such as GOTOMEETING provided by
Citrix Online Division, Inc. of Santa Barbara, Calif., WEBEX
provided by WebEx, Inc. of Santa Clara, Calif., or Microsoft Office
LIVE MEETING provided by Microsoft Corporation of Redmond,
Wash.
[0089] A local machine 202 may execute, operate or otherwise
provide an application, which can be any type and/or form of
software, program, or executable instructions such as any type
and/or form of web browser, web-based client, client-server
application, a thin-client computing client, an ActiveX control, or
a Java applet, or any other type and/or form of executable
instructions capable of executing on local machine 202. In some
embodiments, the application may be a server-based or a
remote-based application executed on behalf of the local machine
202 on a remote machine 206. In other embodiments, the remote
machine 206 may display output to the local machine 202 using any
thin-client protocol, presentation layer protocol, or
remote-display protocol, such as the Independent Computing
Architecture (ICA) protocol manufactured by Citrix Systems, Inc. of
Ft. Lauderdale, Fla.; the Remote Desktop Protocol (RDP)
manufactured by the Microsoft Corporation of Redmond, Wash.; the
X11 protocol; the Virtual Network Computing (VNC) protocol,
manufactured by AT&T Bell Labs; the SPICE protocol,
manufactured by Qumranet, Inc., of Sunnyvale, Calif., USA, and of
Raanana, Israel; the Net2Display protocol, manufactured by VESA, of
Milpitas, Calif.; the PC-over-IP protocol, manufactured by Teradici
Corporation, of Burnaby, B.C.; the TCX protocol, manufactured by
Wyse Technology, Inc., of San Jose, Calif.; the THINC protocol
developed by Columbia University in the City of New York, of New
York, N.Y.; or the Virtual-D protocols manufactured by Desktone,
Inc., of Chelmsford, Mass. The application can use any type of
protocol and it can be, for example, an HTTP client, an FTP client,
an Oscar client, or a Telnet client. In still other embodiments,
the application comprises any type of software related to voice
over Internet protocol (VoIP) communications, such as a soft IP
telephone. In further embodiments, the application comprises any
application related to real-time data communications, such as
applications for streaming video and/or audio.
[0090] The local machine 202 and remote machine 206 may be deployed
as and/or executed on any type and form of computing device, such
as a computer, network device or appliance capable of communicating
on any type and form of network and performing the operations
described herein. FIGS. 2B and 2C depict block diagrams of a
computing device 200 useful for practicing an embodiment of the
local machine 202 or a remote machine 206. As shown in FIGS. 2B and
2C, each computing device 200 includes a central processing unit
221, and a main memory unit 222. As shown in FIG. 2B, a computing
device 200 may include a storage device 228, an installation device
216, a network interface 218, an I/O controller 223, display
devices 224a-n, a keyboard 226 and a pointing device 227, such as a
mouse. The storage device 228 may include, without limitation, an
operating system, software, and a client agent 220. As shown in
FIG. 2C, each computing device 200 may also include additional
optional elements, such as a memory port 203, a bridge 270, one or
more input/output devices 230a-230n (generally referred to using
reference numeral 230), and a cache memory 240 in communication
with the central processing unit 221.
[0091] The central processing unit 221 is any logic circuitry that
responds to and processes instructions fetched from the main memory
unit 222. In many embodiments, the central processing unit 221 is
provided by a microprocessor unit, such as: those manufactured by
Intel Corporation of Mountain View, Calif.; those manufactured by
Motorola Corporation of Schaumburg, Ill.; those manufactured by
Transmeta Corporation of Santa Clara, Calif.; the RS/6000
processor, those manufactured by International Business Machines of
White Plains, N.Y.; or those manufactured by Advanced Micro Devices
of Sunnyvale, Calif. The computing device 200 may be based on any
of these processors, or any other processor capable of operating as
described herein.
[0092] Main memory unit 222 may be one or more memory chips capable
of storing data and allowing any storage location to be directly
accessed by the microprocessor 221, such as Static random access
memory (SRAM), Burst SRAM or SynchBurst SRAM (BSRAM), Dynamic
random access memory (DRAM), Fast Page Mode DRAM (FPM DRAM),
Enhanced DRAM (EDRAM), Extended Data Output RAM (EDO RAM), Extended
Data Output DRAM (EDO DRAM), Burst Extended Data Output DRAM (BEDO
DRAM), Enhanced DRAM (EDRAM), synchronous DRAM (SDRAM), JEDEC SRAM,
PC 100 SDRAM, Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM
(ESDRAM), SyncLink DRAM (SLDRAM), Direct Rambus DRAM (DRDRAM), or
Ferroelectric RAM (FRAM). The main memory 222 may be based on any
of the above described memory chips, or any other available memory
chips capable of operating as described herein. In the embodiment
shown in FIG. 2B, the processor 221 communicates with main memory
222 via a system bus 250 (described in more detail below). FIG. 2C
depicts an embodiment of a computing device 200 in which the
processor communicates directly with main memory 222 via a memory
port 203. For example, in FIG. 2C the main memory 222 may be
DRDRAM.
[0093] FIG. 2C depicts an embodiment in which the main processor
221 communicates directly with cache memory 240 via a secondary
bus, sometimes referred to as a backside bus. In other embodiments,
the main processor 221 communicates with cache memory 240 using the
system bus 250. Cache memory 240 typically has a faster response
time than main memory 222 and is typically provided by SRAM, BSRAM,
or EDRAM. In the embodiment shown in FIG. 2B, the processor 221
communicates with various I/O devices 230 via a local system bus
250. Various buses may be used to connect the central processing
unit 221 to any of the I/O devices 230, including a VESA VL bus, an
ISA bus, an EISA bus, a MicroChannel Architecture (MCA) bus, a PCI
bus, a PCI-X bus, a PCI-Express bus, or a NuBus. For embodiments in
which the I/O device is a video display 224, the processor 221 may
use an Advanced Graphics Port (AGP) to communicate with the display
224. FIG. 2C depicts an embodiment of a computer 200 in which the
main processor 221 communicates directly with I/O device 230b via
HYPERTRANSPORT, RAPIDIO, or INFINIBAND communications technology.
FIG. 2C also depicts an embodiment in which local busses and direct
communication are mixed: the processor 221 communicates with I/O
device 230a using a local interconnect bus while communicating with
I/O device 230b directly.
[0094] A wide variety of I/O devices 230a-230n may be present in
the computing device 200. Input devices include keyboards, mice,
trackpads, trackballs, microphones, and drawing tablets. Output
devices include video displays, speakers, inkjet printers, laser
printers, and dye-sublimation printers. An I/O controller 223, as
shown in FIG. 2B, may control the I/O devices. The I/O controller
may control one or more I/O devices such as a keyboard 226 and a
pointing device 227, e.g., a mouse or optical pen. Furthermore, an
I/O device may also provide storage and/or an installation medium
216 for the computing device 200. In still other embodiments, the
computing device 200 may provide USB connections (not shown) to
receive handheld USB storage devices such as the USB Flash Drive
line of devices manufactured by Twintech Industry, Inc. of Los
Alamitos, Calif.
[0095] Referring again to FIG. 2B, the computing device 200 may
support any suitable installation device 216, such as a floppy disk
drive for receiving floppy disks such as 3.5-inch, 5.25-inch disks
or ZIP disks, a CD-ROM drive, a CD-R/RW drive, a DVD-ROM drive,
tape drives of various formats, USB device, hard-drive or any other
device suitable for installing software and programs. The computing
device 200 may further comprise a storage device, such as one or
more hard disk drives or redundant arrays of independent disks, for
storing an operating system and other related software, and for
storing application software programs such as any program related
to the client agent 220. Optionally, any of the installation
devices 216 could also be used as the storage device. Additionally,
the operating system and the software can be run from a bootable
medium, for example, a bootable CD, such as KNOPPIX, a bootable CD
for GNU/Linux that is available as a GNU/Linux distribution from
knoppix.net.
[0096] Furthermore, the computing device 200 may include a network
interface 218 to interface to the network 204 through a variety of
connections including, but not limited to, standard telephone
lines, LAN or WAN links (e.g., 802.11, T1, T3, 56 kb, X.25, SNA,
DECNET), broadband connections (e.g., ISDN, Frame Relay, ATM,
Gigabit Ethernet, Ethernet-over-SONET), wireless connections, or
some combination of any or all of the above. Connections can be
established using a variety of communication protocols (e.g.,
TCP/IP, IPX, SPX, NetBIOS, Ethernet, ARCNET, SONET, SDH, Fiber
Distributed Data Interface (FDDI), RS232, IEEE 802.11, IEEE
802.11a, IEEE 802.11b, IEEE 802.11g, CDMA, GSM, WiMax and direct
asynchronous connections). In one embodiment, the computing device
200 communicates with other computing devices 200' via any type
and/or form of gateway or tunneling protocol such as Secure Socket
Layer (SSL) or Transport Layer Security (TLS), or the Citrix
Gateway Protocol manufactured by Citrix Systems, Inc. of Ft.
Lauderdale, Fla. The network interface 218 may comprise a built-in
network adapter, network interface card, PCMCIA network card, card
bus network adapter, wireless network adapter, USB network adapter,
modem or any other device suitable for interfacing the computing
device 200 to any type of network capable of communication and
performing the operations described herein.
[0097] In some embodiments, the computing device 200 may comprise
or be connected to multiple display devices 224a-224n, which each
may be of the same or different type and/or form. As such, any of
the I/O devices 230a-230n and/or the I/O controller 223 may
comprise any type and/or form of suitable hardware, software, or
combination of hardware and software to support, enable or provide
for the connection and use of multiple display devices 224a-224n by
the computing device 200. For example, the computing device 200 may
include any type and/or form of video adapter, video card, driver,
and/or library to interface, communicate, connect or otherwise use
the display devices 224a-224n. In one embodiment, a video adapter
may comprise multiple connectors to interface to multiple display
devices 224a-224n. In other embodiments, the computing device 200
may include multiple video adapters, with each video adapter
connected to one or more of the display devices 224a-224n. In some
embodiments, any portion of the operating system of the computing
device 200 may be configured for using multiple displays 224a-224n.
In other embodiments, one or more of the display devices 224a-224n
may be provided by one or more other computing devices, such as
computing devices 200a and 200b connected to the computing device
200, for example, via a network. These embodiments may include any
type of software designed and constructed to use another computer's
display device as a second display device 224a for the computing
device 200. One ordinarily skilled in the art will recognize and
appreciate the various ways and embodiments that a computing device
200 may be configured to have multiple display devices
224a-224n.
[0098] In further embodiments, an I/O device 230 may be a bridge
between the system bus 250 and an external communication bus, such
as a USB bus, an Apple Desktop Bus, an RS-232 serial connection, a
SCSI bus, a FireWire bus, a FireWire 800 bus, an Ethernet bus, an
AppleTalk bus, a Gigabit Ethernet bus, an Asynchronous Transfer
Mode bus, a HIPPI bus, a Super HIPPI bus, a SerialPlus bus, a
SCI/LAMP bus, a FibreChannel bus, or a Serial Attached small
computer system interface bus, or any other type and form of
communication bus.
[0099] A computing device 200 of the sort depicted in FIGS. 2B and
2C typically operates under the control of operating systems, which
control scheduling of tasks and access to system resources. The
computing device 200 can be running any operating system such as
any of the versions of the MICROSOFT WINDOWS operating systems, the
different releases of the Unix and Linux operating systems, any
version of the MAC OS for Macintosh computers, any embedded
operating system, any real-time operating system, any open source
operating system, any proprietary operating system, any operating
systems for mobile computing devices, or any other operating system
capable of running on the computing device and performing the
operations described herein. Typical operating systems include, but
are not limited to: WINDOWS 3.x, WINDOWS 95, WINDOWS 98, WINDOWS
2000, WINDOWS NT 3.51, WINDOWS NT 4.0, WINDOWS 7, WINDOWS CE,
WINDOWS XP, and WINDOWS VISTA, all of which are manufactured by
Microsoft Corporation of Redmond, Wash.; MAC OS, manufactured by
Apple Inc., of Cupertino, Calif.; OS/2, manufactured by
International Business Machines of Armonk, N.Y.; and Linux, a
freely-available operating system distributed by Caldera Corp. of
Salt Lake City, Utah, or any type and/or form of a Unix operating
system, among others.
[0100] The computing device 200 can be any workstation, desktop
computer, laptop or notebook computer, server, handheld computer,
mobile telephone or other portable telecommunication device, media
playing device, a gaming system, mobile computing device, or any
other type and/or form of computing, telecommunications or media
device that is capable of communication and that has sufficient
processor power and memory capacity to perform the operations
described herein. In some embodiments, the computing device 200 may
have different processors, operating systems, and input devices
consistent with the device. For example, in one embodiment, the
computing device 200 is a TREO 180, 270, 600, 650, 680, 700p,
700w/wx, 750, 755p, 800w, Centro, or Pro smart phone manufactured
by Palm, Inc. In some of these embodiments, the TREO smart phone is
operated under the control of the PalmOS operating system and
includes a stylus input device as well as a five-way navigator
device.
[0101] In other embodiments the computing device 200 is a mobile
device, such as a JAVA-enabled cellular telephone or personal
digital assistant (PDA), such as the i55sr, i58sr, i85s, i88s,
i90c, i95cl, i335, i365, i570, I576, i580, i615, i760, i836, i850,
i870, i880, i920, i930, ic502, ic602, ic902, i776 or the im1100,
all of which are manufactured by Motorola Corp. of Schaumburg,
Ill., the 6035 or the 7135, manufactured by Kyocera of Kyoto,
Japan, or the i300 or i330, manufactured by Samsung Electronics
Co., Ltd., of Seoul, Korea. In some embodiments, the computing
device 200 is a mobile device manufactured by Nokia of Finland, or
by Sony Ericsson Mobile Communications AB of Lund, Sweden.
[0102] In still other embodiments, the computing device 200 is a
Blackberry handheld or smart phone, such as the devices
manufactured by Research In Motion Limited, including the
Blackberry 7100 series, 8700 series, 7700 series, 7200 series, the
Blackberry 7520, the Blackberry PEARL 8100, the 8700 series, the
8800 series, the Blackberry Storm, Blackberry Bold, Blackberry
Curve 8900, and the Blackberry Pearl Flip. In yet other
embodiments, the computing device 200 is a smart phone, Pocket PC,
Pocket PC Phone, or other handheld mobile device supporting
Microsoft Windows Mobile Software. Moreover, the computing device
200 can be any workstation, desktop computer, laptop or notebook
computer, server, handheld computer, mobile telephone, any other
computer, or other form of computing or telecommunications device
that is capable of communication and that has sufficient processor
power and memory capacity to perform the operations described
herein.
[0103] In some embodiments, the computing device 200 comprises a
combination of devices, such as a mobile phone combined with a
digital audio player or portable media player. In one of these
embodiments, the computing device 200 is a Motorola RAZR or
Motorola ROKR line of combination digital audio players and mobile
phones. In another of these embodiments, the computing device 200
is a device in the iPhone line of smartphones, manufactured by
Apple Inc., of Cupertino, Calif. In still other embodiments, the
computing device 200 may comprise a tablet computer, such as an
iPad tablet computer manufactured by Apple, Inc., or any other type
and form of tablet computer.
[0104] In one embodiment, a computing device 202a may request
resources from a remote machine 206, while providing the
functionality of a remote machine 206 to a client 202b. In such an
embodiment, the computing device 202a may be referred to as a
client with respect to data received from the remote machine 206
(which may be referred to as a server) and the computing device
202a may be referred to as a server with respect to the second
client 202b. In another embodiment, the client 202 may request
resources from the remote machine 206 on behalf of a user of the
client 202.
[0105] As shown in FIG. 2D, the computing device 200 may comprise
multiple processors and may provide functionality for simultaneous
execution of instructions or for simultaneous execution of one
instruction on more than one piece of data. In some embodiments,
the computing device 200 may comprise a parallel processor with one
or more cores. In one of these embodiments, the computing device
200 is a shared memory parallel device, with multiple processors
and/or multiple processor cores, accessing all available memory as
a single global address space. In another of these embodiments, the
computing device 200 is a distributed memory parallel device with
multiple processors each accessing local memory only. In still
another of these embodiments, the computing device 200 has both
some memory which is shared and some memory which can only be
accessed by particular processors or subsets of processors. In
still even another of these embodiments, the computing device 200,
such as a multicore microprocessor, combines two or more
independent processors into a single package, often a single
integrated circuit (IC). In yet another of these embodiments, the
computing device 200 includes a chip having a CELL BROADBAND ENGINE
architecture and including a Power processor element and a
plurality of synergistic processing elements, the Power processor
element and the plurality of synergistic processing elements linked
together by an internal high speed bus, which may be referred to as
an element interconnect bus.
[0106] In some embodiments, the processors provide functionality
for execution of a single instruction simultaneously on multiple
pieces of data (SIMD). In other embodiments, the processors provide
functionality for execution of multiple instructions simultaneously
on multiple pieces of data (MIMD). In still other embodiments, the
processor may use any combination of SIMD and MIMD cores in a
single device.
[0107] In some embodiments, the computing device 200 may comprise a
graphics processing unit. In one of these embodiments, depicted in
FIG. 2E, the computing device 200 includes at least one central
processing unit 221 and at least one graphics processing unit. In
another of these embodiments, the computing device 200 includes at
least one parallel processing unit and at least one graphics
processing unit. In still another of these embodiments, the
computing device 200 includes a plurality of processing units of
any type, one of the plurality of processing units comprising a
graphics processing unit.
[0108] In one embodiment, a resource may be a program, an
application, a document, a file, a plurality of applications, a
plurality of files, an executable program file, a desktop
environment, a computing environment, or other resource made
available to a user of the local computing device 202. The resource
may be delivered to the local computing device 202 via a plurality
of access methods including, but not limited to, conventional
installation directly on the local computing device 202, delivery
to the local computing device 202 via a method for application
streaming, delivery to the local computing device 202 of output
data generated by an execution of the resource on a third computing
device 206b and communicated to the local computing device 202 via
a presentation layer protocol, delivery to the local computing
device 202 of output data generated by an execution of the resource
via a virtual machine executing on a remote computing device 206,
or execution from a removable storage device connected to the local
computing device 202, such as a USB device, or via a virtual
machine executing on the local computing device 202 and generating
output data. In some embodiments, the local computing device 202
transmits output data generated by the execution of the resource to
another client computing device 202b.
[0109] In some embodiments, a user of a local computing device 202
connects to a remote computing device 206 and views a display on
the local computing device 202 of a local version of a remote
desktop environment, comprising a plurality of data objects,
generated on the remote computing device 206. In one of these
embodiments, at least one resource is provided to the user by the
remote computing device 206 (or by a second remote computing device
206b) and displayed in the remote desktop environment. However,
there may be resources that the user executes on the local
computing device 202, either by choice, or due to a policy or
technological requirement. In another of these embodiments, the
user of the local computing device 202 would prefer an integrated
desktop environment providing access to all of the resources
available to the user, instead of separate desktop environments for
resources provided by separate machines. For example, a user may
find navigating between multiple graphical displays confusing and
difficult to use productively. Or, a user may wish to use the data
generated by one application provided by one machine in conjunction
with another resource provided by a different machine. In still
another of these embodiments, requests for execution of a resource,
windowing moves, application minimize/maximize, resizing windows,
and termination of executing resources may be controlled by
interacting with a remote desktop environment that integrates the
display of the remote resources and of the local resources. In yet
another of these embodiments, an application or other resource
accessible via an integrated desktop environment--including those
resources executed on the local computing device 202 and those
executed on the remote computing device 206--is shown in a single
desktop environment.
[0110] In one embodiment, data objects from a remote computing
device 206 are integrated into a desktop environment generated by
the local computing device 202. In another embodiment, the remote
computing device 206 maintains the integrated desktop. In still
another embodiment, the local computing device 202 maintains the
integrated desktop.
[0111] In some embodiments, a single remote desktop environment 204
is displayed. In one of these embodiments, the remote desktop
environment 204 is displayed as a full-screen desktop. In other
embodiments, a plurality of remote desktop environments 204 is
displayed. In one of these embodiments, one or more of the remote
desktop environments are displayed in non-full-screen mode on one
or more display devices 224. In another of these embodiments, the
remote desktop environments are displayed in full-screen mode on
individual display devices. In still another of these embodiments,
one or more of the remote desktop environments are displayed in
full-screen mode on one or more display devices 224.
[0112] Referring now to FIG. 3A, illustrated is a block diagram of
a system for multivariate analysis of adverse event data. In brief
overview, a client 300 may comprise an application 302 and, in some
embodiments, genomic information 303. In some embodiments, a client
300 may communicate with a server 304 via any type of network, such
as those discussed herein. Although shown as a separate
client-server system, in many embodiments, a client 300 and server
304 may be on the same physical machine. In other embodiments,
server 304 may be executed by a virtual machine provided by a cloud
computing environment. For example, server 304 may comprise a
hosted service or cloud service, providing scalability and ease of
management. In some embodiments, a medical literature server 340
and/or an adverse event data server 342 may also communicate with a
server 304. In other embodiments not shown, a second client 300 may
be used to gather data from a medical literature server 340 and/or
an adverse event data server 342 and processed or transferred to
server 304. In some embodiments, a server 304 may comprise an
input/output interface 306, a security module 308, and/or a display
module 310. Server 304 may also comprise one or more databases or
data stores, including an adverse event database 312, a medication
information database 314, a literature database 316, and a variant
database 318. Server 304 may, in some embodiments, comprise an
analyzer 320 and/or a parser 322. In some embodiments, server 304
may comprise a global molecular entity graph 324.
[0113] Still referring to FIG. 3A and in more detail, in some
embodiments, a client 300 may comprise a computing device of any
type, such as a desktop computer, portable computer, smart phone,
tablet computer, or any other type of computing device. Client 300
may execute an application 302 for accessing server 304. In some
embodiments, application 302 may comprise a web browser, while in
other embodiments, application 302 may comprise a dedicated
application for communicating with server 304.
[0114] In some embodiments, client 300 may store, include, or
otherwise access genomic information 303. Genomic information 303
may comprise genetic data about a patient. For example, in some
embodiments, genomic information 303 may comprise a list of genetic
variants or mutations of the patient, a full or partial genetic
sequence, or any similar information. In some embodiments, genomic
information 303 may be utilized for generating personalized drug
efficacy or risk information or identifying potential drug
interactions. Although shown on client 300, in many embodiments,
genomic information 303 may be stored externally to client 300,
obtained from a third party or stored on a second server or network
storage device, or otherwise be supplied to server 304.
[0115] Server 304 may comprise a computing device of any type, such
as a desktop computer, portable computer, rackmount server,
workstation, or any other type of computing device. In some
embodiments, server 304 may comprise a virtual machine executed by
a cloud service, a plurality of servers forming a grid or server
farm 38 and acting as a single server 304, or any other type of
server. Although shown with components 306-324 as part of server
304, in many embodiments, one or more of components 306-324 may be
external to server 304, on a second server (not illustrated), on an
external storage device, or otherwise accessible to server 304.
[0116] In some embodiments, server 304 may execute an input/output
interface 306. Input/output interface 306 may comprise an
application, service, daemon, routine, or other executable logic
for communicating with one or more clients 300 or other servers,
medical literature servers 340, and/or adverse event data servers
342. In some embodiments, input/output interface 306 may comprise a
web server or web page executed by a web server. Input/output
interface 306 may provide an interface allowing a user to provide
queries, make selections or identifications of drugs, indications,
targets, pathways, or other molecular entities, define cohorts for
analysis, or perform other functions. In some embodiments,
input/output interface 306 may provide data tables, graphics, or
other output views to the user. In many embodiments, input/output
interface 306 may communicate via a network with application 302,
while in other embodiments in which client 300 and server 304
comprise the same computing device, application 302 may be executed
on server 304 and may communicate with input/output interface 306
via an API.
[0117] In some embodiments, server 304 may execute a security
module 308. Security module 308 may comprise an application,
service, daemon, routine, or other executable logic for receiving
user credentials or login information and/or computing device
credentials, such as a network address, operating system version or
other identification, and processing the credentials to allow or
deny access to server 304. Security module 308 may, in some
embodiments, comprise a user and password database or similar
features to control access to functions of server 304.
[0118] In some embodiments, server 304 may execute a display module
310. Display module 310 may comprise an application, service,
daemon, routine, or other executable logic for generating graphic
displays for presentation by input/output interface 306 and/or
application 302 to a user. In some embodiments, display module 310
may generate graphs, tables, radial graphs, charts, biological
network diagrams, or other graphical entities. In some embodiments,
input/output interface 306 and display module 310 may be provided
as part of a web server or application, while in other embodiments,
these services may comprise separate executable modules.
[0119] Server 304 may include an adverse event database 312 and/or
a medication information database 314. In some embodiments, adverse
event database 312 and/or medication information database 314 may
be stored on server 304, while in other embodiments, adverse event
database 312 and/or medication information database 314 may be
stored on a data storage server, external storage device, within a
cloud storage system, or otherwise accessible to parser 322 and/or
analyzer 320. An adverse event database 312 may comprise a
database, flat file, data array, or other data file for storing
molecular data regarding adverse events. Similarly, a medication
information database 314 may comprise a database, flat file, data
array, or other data file for storing molecular entity information
for one or more drugs. As discussed above in connection with FIG.
1B, stored data may comprise identifications of one or more drugs
102, indications 104, reactions 106, outcomes 108, pathways 110,
targets 112, metabolizing enzymes or transporters 114, and drug
classes 116. In many embodiments, adverse event data may comprise
demographic information of a patient, trial participant, or other
person that experienced the adverse event. In many embodiments,
adverse event data 102-108 from adverse event reporting systems may
be combined and linked with molecular entity data 110-116 in the
adverse event database 312 and/or medication information database
314. In some embodiments, molecular entity data 110-116 for a drug
may be retrieved from pharmaceutical manufacturer literature,
research literature or white papers, or other literature from one
or more medical literature servers 340. In many embodiments,
adverse event database 312 and medication information database 314
may comprise a single database, while in other embodiments,
databases 312-314 may be linked to allow associations between
entities and adverse event data. In some embodiments, associations
may be one-to-one, such as a single outcome for a single patient,
while in other embodiments, associations may be one-to-many, such
as a plurality of prescribed and co-prescribed drugs for the
patient, or many-to-many, such as a plurality of indications
associated with each of a plurality of drugs. Accordingly, a
adverse event/molecular entity database comprising adverse event
database 312 and medication information database 314 may comprise a
multi-dimensional database allowing associations between adverse
events and biological information. Such a database may be used for
novel univariate analyses, such as generating an ordered list of
metabolizing enzymes most frequently associated with a specified
side effect (by numbers of adverse event reports for the side
effect or reaction including a drug, the drug associated with the
metabolizing enzyme in medical literature). Similarly, such a
database may be used for multivariate analyses, such as comparing
reported side effects of all drugs targeting a first protein with
side effects of all drugs targeting a second protein.
[0120] In some embodiments, medication information database 314 may
comprise or be associated with a literature database 316.
Literature database 316 may comprise a database, data array, flat
file, or other data comprising one or more items of literature
about one or molecular entities. Literature database 316 may
comprise white papers, research papers, theses, dissertations,
abstracts of literature, publicly available literature, proprietary
manufacturer literature, research data, or other literature. In
some embodiments, literature database 316 may comprise medication
information, which may be extracted to generate a medication
information database 314. In some embodiments, a server 304 may
retrieve or receive literature from one or more medical literature
servers 340. For example, in one embodiment, server 304 may
retrieve abstracts or full papers from the PubMed database provided
by the National Institutes of Health of Bethesda, Md. Such papers
or abstracts may be parsed to identify drug names, drug classes,
protein targets, metabolizing enzymes, transporters, gene variants
or wild types, or other molecular entities. Once identified, the
entities and associations between identified entities may be added
to literature database 316, medication information database 314,
adverse event database 312, or a combined multi-dimensional
molecular data database.
[0121] In some embodiments, adverse event database 312 may further
comprise identification of patient genetic variants or mutations,
or may be associated with a variant database 318. A variant
database may comprise a database, data file, flat file, data array,
or other file comprising a full genetic sequence for one or more
patients, clinical trial participants, or other persons, or may
comprise a partial sequence, or may comprise an identification of
one or more variants or mutated gene sequences for a patient,
participant, or person. In some embodiments, a variant database may
further comprise identifications of one or more proteins
corresponding to a variant, in which expression or activation of
the protein is affected by the mutation. For example, in one such
embodiment, a database may comprise an identification of a variant
and an identification of a protein activated by the wild type
corresponding to the variant. By linking variant identifications,
protein activation or deactivation, and drug target proteins, a
user may identify potential decreased efficacy of a drug or high
risk biological interactions.
[0122] In some embodiments, a server 304 may comprise an analyzer
or analysis module 320. Analyzer 320 may comprise an application,
service, daemon, routine, or other executable logic for performing
univariate or multivariate analysis. In some embodiments, analyzer
320 may identify associated entities from a database, such as
reactions associated with a target protein, or outcomes associated
with a genetic variant. In many embodiments, analyzer 320 may
generate one or more lists of associated entities based on an input
or requested first entity. Such lists may be ordered, for example,
by a percentage of total associations or by number of associations
in the database. Accordingly, for a query of adverse reactions
associated with a first drug, analyzer 320 may return an ordered
list indicating that, for example, of all reported adverse
reactions associated with the first drug, nausea occurs in 60% of
cases, fatigue occurs in 50% of cases, and a rash occurs in 40% of
cases. Due to the possibility of patients experiencing multiple
adverse events, totals may exceed 100%. Similarly, for a query of
targets associated with an adverse reaction such as fatigue,
analyzer 320 may return a list of molecular targets ordered by
proportional reporting ratio (PRR), such as dihydroorotase having a
PRR of 32.91, DNA polymerase i having a PRR of 16.45, and
cytochrome b having a PRR of 8.22. Such proportional reporting
rations may be determined based on a proportion of reactions to the
molecular entity compared to the same proportion for all such
entities in the database. In some embodiments, analyzer 320 may
further comprise functionality for performing multivariate analyses
and comparisons. For example, analyzer 320 may comprise logic for
extracting subsets of statistical data of adverse events associated
experienced by an identified first cohort of patients or trial
participants and an identified second cohort, and comparing the two
subsets to identify adverse event differences between the cohorts.
Phenotype or genotype distinctions between the cohorts may then be
identified as the likely cause or mitigation of adverse events.
[0123] In some embodiments, server 304 may comprise a parser 322.
Parser 322 may comprise an application, service, daemon, routine,
or other executable logic for reading and interpreting medical
literature obtained from a medical literature server 340 or stored
in a literature database 316. Reading and interpreting medical
literature may comprise scanning literature for identifications of
one or more molecular entities. Inclusion of identifications of a
plurality of entities within a single item of literature may
indicate an association between those entities. Such associations
may then be incorporated into a medication information database 314
and/or adverse event database 312. For example, parser 322 may scan
medical literature and identify that the terms "headache" and
"aspirin" frequently appear in the same items of literature.
Accordingly, parser 322 may identify the indication "headache" as
related to the drug "aspirin" in a medication information database
314. Similarly, in some embodiments, parser 322 may identify
associations within literature between drugs, targets,
transporters, metabolizing enzymes, drug classes, genetic variants,
side effects, indications, reactions, outcomes, patient demographic
information, or any other such information. Parser 322 may scan
white papers, abstracts, articles, theses, research documents,
manufacturer literature, or any other type of document for
associations between molecular entities. In some embodiments,
parser 322 may score the identified associations responsive to one
or more factors, such as frequency, proximity, and secondary
citations. For example, parser 322 may give a low association score
to two molecular entities that appear in only a single item of
literature once. However, parser 322 may give a higher association
score to the two molecular entities, if they appear in close
proximity to each other within the literature, such as in the same
sentence or paragraph. In some embodiments, parser 322 may give a
higher association score to associations between two entities that
appear in a plurality of items of literature than an association
between two entities that appears repeatedly in only a single item
of literature. In such embodiments, parser 322 may thus identify
associations that are commonly understood by researchers, rather
than unconfirmed or proposed associations. In some embodiments,
parser 322 may further identify secondary items of literature that
cite a first item of literature, and give a higher score to
associations identified within the first item of literature.
Frequently cited literature thus may become more authoritative
regarding associations.
[0124] In some embodiments, server 304 may comprise a global
molecular entity graph 324. Global molecular entity graph 324 may
comprise a graph, database, or other data file for identifying a
plurality of molecular entities and relationships between entities.
Global molecular entity graph 324 may comprise a system-wide
representation of some or all biological systems within the human
body. For example, referring briefly to FIG. 3B, illustrated is a
diagram of an example embodiment of a global molecular entity graph
324. The graph may comprise a plurality of molecular entities 350,
such as proteins, enzymes, transporters, or other entities, and
each entity 350 may be associated with one or more other entities
350 via a relationship 352. In some embodiments, a global molecular
entity graph 324 may be used by an analyzer 320 to extract
subgraphs 354, which may comprise portions of the molecular entity
graph important to a particular entity. For example, a subgraph 354
may comprise all entities and relationships between entities
associated with a first identified entity, such as a drug target.
In some embodiments, multiple subgraphs 354 may be extracted and
compared to identify common entities and/or relationships between
the subgraphs. For example, referring briefly to FIG. 3C,
illustrated is a diagram of an example embodiment of two extracted
subgraphs, 354a and 354b, intersected to identify an intersection
subgraph 354c. A first subgraph 354a may be extracted for a first
drug target (P1), and a second subgraph 354b extracted for a second
drug target (P2). The intersection subgraph 354c may identify one
or more molecular entities 350 affected by each of P1 and P2. These
dual-affected entities may be causes of adverse effects experienced
when drugs targeting P1 and P2 are taken simultaneously, but not
experienced when drugs targeting P1 and P2 are taken separately. By
using multivariate analysis of adverse event data and extracting
subgraphs for identified entities with disparate adverse event
data, server 304 may be able to identify one or more molecular
entities associated with a particular side effect, even when the
association would be normally hidden in univariate analyses.
[0125] Returning to FIG. 3A, in some embodiments, server 304 may
communicate with a medical literature server 340 and/or an adverse
event data server 342. Medical literature server 340 may comprise
any server, database, online storage system, cloud storage device,
offline storage system, computing device, or other device for
storing medical literature, including research documents, theses,
white papers, manufacturer data, or other literature. In some
embodiments, server 304 may access medical literature server 340 to
retrieve documents to fill literature database 316, medication
information database 314, variant database 318, or for parsing one
or more items of literature via parser 322 as discussed above.
Similarly, adverse event data server 342 may comprise any server,
database, online storage system, cloud storage device, offline
storage system, computing device, or other device for storing
adverse event data, such as the Adverse Event Reporting System
provided by the U.S. Food & Drug Administration. In some
embodiments, server 304 may access an adverse event data server 342
to retrieve records to fill an adverse event database 312 or for
parsing by parser 322 or analysis by analyzer 320, as discussed
above.
[0126] In some embodiments, a safety profile, sometimes referred to
as an adverse event profile or side effect profile, may comprise a
list of all adverse event reports associated with a molecular
entity, such as all adverse event reports for a prescribed or
co-prescribed medication. In other embodiments, a safety profile
may comprise a statistical table of adverse event reports
associated with a molecular entity, such as a table identifying
frequency of occurrence of one or more adverse events with patients
or trial participants consuming a specified drug. A molecular
entity multivariate analysis system may be used to compare the
safety profiles of a plurality of molecular entities, allowing
identification of entities responsible for adverse event
differences between safety profiles. For example, in some
embodiments, a safety profile for a first drug or medication may be
compared to a safety profile for a second drug or medication.
Similarly, safety profiles may be generated based on molecular
entities associated with adverse event reports. For example, a
patient that experienced an adverse event may have been prescribed
a first drug. The first drug may be known to target a first
protein. Accordingly, by correlating this information with the
adverse event report, a safety profile for the protein may be
generated. Thus, in some embodiments, a safety profile for a
protein target may be compared to a safety profile for a second
protein target.
[0127] Similarly, safety profiles may be generated and compared for
indications themselves. Such safety profiles may comprise a list of
medications prescribed or co-prescribed to patients identified as
being treated for the indication. In one embodiment, such a list
may be ordered by percentage of patients prescribed or
co-prescribed the medication, while in another embodiment, such a
list may be ordered by percentage of patients prescribed or
co-prescribed the medication who experienced an adverse event, or a
particular outcome or outcomes. Accordingly, in some embodiments, a
multivariate analysis system may be able to determine if two
similar indications, such as depression and post-partum depression,
have a different prioritization of drugs responsible for adverse
events. Although discussed primarily in terms of similar
indications, in many embodiments, any two or more indications may
compared, allowing complex analysis of similarities between
apparently diverse indications. For example, and referring briefly
to FIG. 4A, illustrated is a block diagram of an embodiment of a
method for identifying molecular entities responsible for adverse
event differences between indications. A multivariate analysis
system may retrieve a safety profile for a first indication 402
from adverse event data 400, and may generate a list of medications
404A-404n ordered by percentage of medication-consumers
experiencing an adverse event 406A-406n. In some embodiments, the
list may be ordered by percentage of medication-consumers
experiencing any adverse event, while in other embodiments, the
list may be narrowed to include only percentages of
medication-consumers experiencing a specific adverse event.
Similarly, the multivariate analysis system may retrieve a second
safety profile for a second indication 402', and may generate a
list of medications 404A-404n ordered by percentage of
medication-consumers experiencing an adverse event 406A'-406n'. In
some embodiments, safety profiles may include different medications
404A-404N, although in most embodiments, a medication 404A-404n may
appear in both safety profiles. Additionally, medications may
appear in different priorities in each ordered list, such as
medication 404C and medication 404F in the example lists of FIG.
4A. Differences in order may be due to physiological specificities
of either indication and their differential effect on drug
pharmacokinetics or dynamics. Accordingly, through analysis of the
different molecular entities (e.g. entities 408A-408D) associated
with a medication appearing in a first position in one safety
profile for a first indication and in a second, different position
in another safety profile for a second indication (e.g. medication
6 404F), molecular entities affected differently by each indication
may be immediately identified. In many embodiments, such second
indication may comprise an indication similar to the first. This
may provide opportunities for more targeted therapies for one or
both indications. Furthermore, when safety profiles for each of the
indication are narrowed by a specific adverse event, differences
between each safety profile may identify potentially unknown
interactions between molecular entities associated with the
indication and molecular entities associated with the adverse
event. For example, if a large percentage of patients with a first
indication taking a first medication experience a specific adverse
event, but a small percentage of patents with a second indication
taking the first medication experience the specific adverse event,
this may indicate differences between each indications interaction
with the molecular entities responsible for the adverse event.
Although shown ordered by percentage in FIG. 4A, in many
embodiments, each list may be in any order, with comparisons
performed on percentage values associated with each medication as
opposed to order.
[0128] Referring now to FIG. 4B, illustrated is a flow chart of an
embodiment of a method for identifying molecular entities
responsible for adverse event differences between indications. In
brief overview, a multivariate analyzer such as analyzer 320 of a
computing device 304 may receive an identification of a first
indication at step 422. The analyzer may receive an identification
of a second indication at step 424. In many embodiments, the second
indication may be similar to the first indication. At step 426, in
some embodiments as discussed above, the analyzer may receive an
identification of an adverse event. At step 428, the analyzer may
retrieve from an adverse event database a first list of medications
prescribed to patients for the first indication, the list
comprising percentages of patients prescribed each medication who
experienced an adverse event. In some embodiments, the list may be
limited to adverse event data for the identified adverse event, and
accordingly, the list may comprise percentages of patients
prescribed the medication who experienced the identified adverse
event. At step 430, the analyzer may retrieve from the adverse
event database a second list of medications prescribed to patients
for the second indication, the list comprising percentages of
patients prescribed each medication who experienced an adverse
event. In some embodiments, the list may be limited to adverse
event data for the identified adverse event, and accordingly, the
list may comprise percentages of patients prescribed the medication
who experienced the identified adverse event. At step 432, in some
embodiments, the analyzer may compare the first list and second
list to identify one or more medications with a different
percentage value in each list. At step 434, the analyzer may
retrieve one or more lists of molecular entities associated with a
corresponding each of the identified one or more medications. At
step 436, an output module of the computing device may present the
retrieved one or more lists of molecular entities to the user as
lists of molecular entities potentially affected by only one of the
first indication and the second indication.
[0129] Still referring to FIG. 4B and in more detail, at step 422,
an analyzer 320 may receive an identification of a first
indication. As discussed above, an indication may comprise a
disease, a symptom, an adverse effect, or any other such
circumstance which indicates the advisability or necessity of a
specific medical treatment or procedure. In some embodiments,
analyzer 320 may receive the identification of a first indication
from an input/output module, such as a web interface or application
interface. In some embodiments, a user may select the first
indication or input a name of the first indication into a text
entry field, and an input module may pass the identification of the
indication to the analyzer. In other embodiments, the user may
select the first indication from a list of indications. In many
embodiments, analyzer 320 may receive the identification of the
indication from a second computing device operated by or on behalf
of the user.
[0130] At step 424, the analyzer may receive an identification of a
second indication. The second indication may be similar to the
first indication, in some embodiments, while in other embodiments,
the second indication may comprise any indication. Indications may
be similar if they share symptoms; are subsets of a category of
indication (e.g. different types of cancer); if they are commonly
or functionally associated (e.g. nausea and vomiting); or via other
similar associations. In some embodiments, indications may be
similar if they are involve the same pathway, protein, or other
molecular entity. In some embodiments, analyzer 320 may receive the
identification of the second indication from an input/output
module, such as a web interface or application interface. In some
embodiments, a user may select the second indication or input a
name of the second indication into a text entry field, and an input
module may pass the identification of the indication to the
analyzer. In other embodiments, the user may select the second
indication from a list of indications. In many embodiments, the
analyzer may receive the identification of the second indication
from a second computing device operated by or on behalf of the
user.
[0131] At step 426, in some embodiments, the analyzer may receive
an identification of an adverse event. In some embodiments, the
adverse event may comprise an adverse event distinct from the first
indication and second indication. The adverse event may thus be
suspected of being caused by one or more drugs prescribed or
co-prescribed to patients with the first or second indication. For
example, in one embodiment, the two similar indications may
comprise depression and post-partum depression, and the adverse
event may comprise a rash. As depression is not typically
associated or functionally identified as causing a rash, clinicians
may suspect that the adverse event is not caused by the indication,
but by a medication. Thus, in many embodiments, the adverse event
may not be an adverse event corresponding to one of the indications
(e.g. an adverse event of fatigue for an indication of chronic
fatigue syndrome).
[0132] At step 428, the analyzer may retrieve a first list of
medications prescribed to patients with the first indication who
experienced the identified adverse event, and a second list of
medications prescribed to patients with the second indication who
experienced the identified adverse event. Retrieving the lists of
medications may comprise searching an adverse event database for
reports corresponding to the identified adverse event. Each report
may comprise patient demographic information, an identification of
the adverse event, an identification of an indication, an
identification of an outcome, and an identification of one or more
medications consumed by the patient. The adverse event database may
comprise a collated index of adverse events, normalized to be
searchable with standard terms and definitions (for example,
replacing abbreviations with full titles, etc.). In some
embodiments, the analyzer may retrieve a subset of adverse event
reports that include the identification of the adverse event. The
analyzer may then extract a second subset of adverse event reports
that include the identification of the first indication, and
extract a third subset of adverse event reports that include the
identification of the second indication. The analyzer then, in some
embodiments, may iteratively sort or count the extracted subsets of
adverse event reports to generate a table of medications identified
in the extracted subsets, sorted by count or percentage of listing
in the extracted subsets. In other embodiments, the tables may be
unsorted. For example, referring briefly to FIG. 4C, illustrated is
a flow chart of an embodiment of a method 428 for retrieving a list
of medications for an indication and adverse event. At step 450, as
discussed above, the analyzer may retrieve the first subset of
adverse event reports for the identified adverse event, and at step
452, the analyzer may extract a second subset of adverse event
reports from the first subset including the indication. Although
shown in this order, in many embodiments, these steps may be
reversed. For example, the analyzer may extract a subset of adverse
event reports for the indication, and may then extract a further
subset of adverse event reports corresponding to the identified
adverse event. Furthermore, in some embodiments, these steps may be
performed simultaneously as part of a Boolean search.
[0133] At step 454 of FIG. 4C, the analyzer may identify a first
medication in the extracted subset of adverse event reports for the
indication and identified adverse event. At step 456, the analyzer
may then search the extracted subset to identify the number and/or
percentage of times that the first medication is listed in the
adverse event reports. In some embodiments, the analyzer may search
the extracted subsets for records in which the first medication is
listed as the medication suspected of causing the identified
adverse reaction as opposed to being a co-prescribed or concomitant
medication, while in other embodiments, the analyzer may search the
extracted subsets for all appearances of the first medication. At
step 458, the analyzer may add the first medication and the count
or percentage to a list. In some embodiments, a percentage of the
reports in which the medication appears out of the total number of
adverse event reports for the indication and adverse event may be
more useful, while in other embodiments, a raw count may be
preferred. The list may be similarly sorted by either number. In
many embodiments, analyzer may iteratively repeat steps 454-458 for
each additional medication identified in the extracted subset of
adverse event reports. At step 460, in some embodiments utilizing
raw counts, the analyzer may determine a percentage for each
medication as discussed above. In some embodiments, the analyzer
may sort the list by the identified count or percentage to generate
an ordered list. Sorting may be done through any sort algorithm,
such as a bubble sort, quick sort, merge sort, or any other type of
sorting.
[0134] Returning to FIG. 4B, at step 430, the analyzer may retrieve
a second list of medications for the second indication and the
identified adverse event. Although shown for step 428 of FIG. 4B,
embodiments of the method shown in FIG. 4C may also be applied to
step 430 for retrieval of the second list of medications. In some
embodiments, steps 428 and 430 may be performed in any order, or
simultaneously, such as by a multi-threaded processor.
[0135] At step 432, the analyzer may compare the first list and
second list to identify a medication with a different percentage
value in each list. In some embodiments, if the medication appears
in 90% of adverse event reports for the first indication, but only
20% of adverse event reports for the second indication, the
difference in percentages may indicate an important distinction
between the two indications. Accordingly, in many embodiments, the
analyzer may identify a medication with a difference between the
count or percentage in the first list and the count or percentage
in the second list that is greater than a predetermined threshold
amount. Such a threshold may be a percentage, such as 5%, 10%, 20%
or any other value, or may be a number, such as 100 reports, 1000
reports, or any other value. As discussed above, in many
embodiments, ordering by percentages may be useful for certain
comparisons, such as where a first indication has a greater number
of adverse event reports than a second indication. In such
embodiments, percentages may be more easily compared than raw
counts. In other embodiments, the analyzer may determine
differences based on each medication's position in each list, the
list being ordered by percentage or count. This may be useful in
embodiments in which raw counts are used, for example. In similar
embodiments, the list may comprise an index number for each entry,
and the analyzer may compare index numbers of a medication in both
lists.
[0136] At step 434, in some embodiments, the analyzer may retrieve
a third list of molecular entities associated with the identified
medication from a medication information database. As discussed
above, in some embodiments, a medication information database may
comprise part of or be joined with an adverse event database. The
medication information database may identify a medication and known
targets, pathways, enzymes, transporters, or other molecular
entities associated with the medication.
[0137] At step 436, in some embodiments, the analyzer may present
the retrieved third list to the user as a list of molecular
entities potentially affected by only one of the first indication
and the second indication. As discussed above, if a first
indication causes activation of a particular protein and a second
indication does not, and a medication's interaction with the
activated protein causes the adverse effect, such adverse effect
differences may be detected in the adverse event reports,
indicating that the first indication and second indication interact
with the molecular entities affected by the medication in different
ways. This may be useful in identifying potential avenues for
research for the two indications.
[0138] In some embodiments, the analyzer may repeat steps 432-434
for additional medications appearing in both the first list and
second list. In one such embodiment, the analyzer may present a
plurality of lists of molecular entities for each identified
medication, while in other embodiments, the analyzer may merge the
lists of molecular entities. In one embodiment, the analyzer may
generate a combined list including all molecular entities in each
retrieved list, while in other embodiments, the analyzer may
generate an intersection list including only molecular entities in
all retrieved lists. In still other embodiments, the analyzer may
generate a combined list comprising a score for each molecular
entity. In one embodiment, each score may comprise a default score.
The analyzer may increase the default score for each molecular
entity appearing in a plurality of lists and/or decrease the
default score for each molecular entity appearing in one list. In
some embodiments, each molecular entity may be scored responsive to
the number of retrieved lists in which it appears. This may be used
to generate a priority of which molecular entities are most likely
associated with the adverse event rate differences. With a greater
number of medications inducing or suppressing an adverse effect at
a different rate in each indication, the analyzer may be able to
generate more accurate priorities of molecular entities associated
with the adverse event rate differences.
[0139] As discussed above, in some embodiments, a computing device
may comprise global molecular entity graph. Such a graph may
comprise a linked network of nodes representing molecular entities,
such as proteins or enzymes, and functional interactions between
the entities, such as a link between an enzyme and an organic
compound catalyzed by the enzyme. In some embodiments, the graph
may comprise a hypergraph with edges connecting to more than two
nodes, while in other embodiments, the graph may comprise a
two-dimensional graph with intermediate reaction nodes.
[0140] A global molecular entity graph may be used for identifying
molecular entities associated with a side effect or indication and
building an indication or side effect-specific model of molecular
interactions. Although the global molecular entity graph is not
indication or side effect specific, an analyzer may extract
subgraphs or subnetworks from the global molecular entity graph to
generate a model of entities related to a specified indication.
Building an indication or side effect specific molecular entity
model may allow for targeted pharmacological research regarding
entities previously unassociated with the indication or side
effect. In some embodiments, the analyzer may utilize an adverse
event database to identify medications associated with the
specified indication and/or adverse event. The analyzer may then
use a medication information database to identify molecular
entities, such as a proteins and enzymes, related to the identified
medications. In other embodiments, as discussed above, medication
information may be integrated into the adverse event database such
that each adverse event record further includes or is linked to
identifications of molecular entities associated with the
prescribed or consumed medications of the patient that experienced
the adverse event. Accordingly, in such embodiments, the analyzer
may utilize the database to identify molecular entities associated
with the specified indication and/or adverse event. In some
embodiments, the analyzer may identify molecular entities or
medications that are most highly associated with the selected
indication or side effect. For example, as discussed above, in some
embodiments, the analyzer may sort a retrieved list of medications
or molecular entities associated with adverse event reports for the
selected indication or side effect. In a further embodiment, the
analyzer may discard medications or molecular entities with a count
or percentage below a predetermined threshold. For example, in
building a side effect-specific model, it may be advantageous to
focus on molecular entities associated with the side effect in more
than 50% of the adverse event reports for the side effect, and
discard entities in fewer than 50% of the reports. The
predetermined threshold may be any value, and, in some embodiments,
may even include 0% or 100%, either allowing in all associated
entities, or restricting to entities that appear in every adverse
event record. Medications or entities may be sorted and ordered by
various statistical techniques, including proportional reporting
ratios (PRR), regularized PRR (normalized such that older
medications do not outweigh newer medications in the adverse event
reports merely due to amount of data collected, for example),
logistic regression, or other algorithms.
[0141] In many embodiments, the molecular entities identified at
this stage may include only entities known to be associated with
the identified medications. For example, the entities may include
known target proteins, but may not include unknown off-target
proteins or intermediate molecular entities involved in catalyzing
or metabolizing the medication. Furthermore, as multiple
medications may be associated with an indication or side effect,
the identified entities may comprise disjoint regions of the global
molecular entity graph. For example, referring briefly to FIG. 5A,
illustrated is a chart diagram of an embodiment of a global
molecular entity graph 500. Multiple molecular entities or nodes
may be linked to show functional interaction. A first subset of
entities 502 may be known to be associated with a first medication,
and a second subset of entities 504 may be known to be associated
with a second medication, the first medication and second
medication associated with a selected indication or side effect.
Including only the subsets 502 and 504 may comprise an incomplete
list of the entities responsible for or associated with
experiencing the selected indication or side effect.
[0142] Accordingly, the global molecular entity graph may be used
to expand or augment the identified set of entities by identifying
additional entities functionally related to known and identified
entities, such as subsets 502 and 504. In one embodiment, the set
of entities may be augmented by performing a shortest path analysis
between disjoint pairs of known entities, such as a first entity
identified as associated with a first medication (e.g. subset 502)
and a second entity identified as associated with a second
medication (e.g. subset 504). In some embodiments, edges between
nodes may be weighted based on relationships to other entities. For
example, edges to an intermediate node between two entities may be
more heavily weighted if the intermediate node is further connected
to a second intermediate node between both entities. In other
embodiments, edges between nodes may be weighted responsive to
identification of the node as related to an organ associated with
the side effect or indication, such as aspartate transaminase (AST)
being related to the liver with an indication of hepatitis.
Accordingly, weights may vary depending on the identified
indication or side effect. The analyzer may perform any type or
form of shortest path analysis, including Dijkstra's algorithm, a
Bellman-Ford algorithm, or any other type and form of routing
algorithm. Such analysis may, for example, indicate to include
entities 506 and not include entities 508 in the example embodiment
of FIG. 5A.
[0143] In other embodiments, the set of entities may be augmented
by scoring nodes in the global molecular entity graph with respect
to their inclusion in a subnetwork with desired properties. In one
embodiment, modifying scores may include increasing scores related
to an organ associated with the indication or side effect and
reducing scores of unrelated nodes. In another embodiment, scores
may be modified by increasing scores of nodes well connected to
other nodes within the subnetwork and decreasing scores of nodes
well connected to other nodes external to the subnetwork. This may
minimize connectivity to the remainder of the network, reducing the
likelihood of false positives and, if incorporated with the above
discussed embodiments, decreasing complexity of a shortest path
analysis.
[0144] In still other embodiments, pre-defined pathways within the
global molecular entity network (e.g. glycolysis, cAMP-dependent
pathway, etc.) may be scored with respect to their coverage of the
indication-relevant entities or entities known to be associated
with identified medications associated with the indication or side
effect. Merging high-scoring pathways may thus allow generating an
indication-specific subnetwork.
[0145] Referring now to FIG. 5B, illustrated is a flow diagram of
an embodiment of a method for extracting an indication-specific
model from a global molecular entity graph. In brief overview at
step 522, an analyzer or an input/output module in communication
with an analyzer may receive an identification of an indication or
side effect. At step 524, the analyzer may identify molecular
entities known to be associated with the indication or side effect.
At step 526, the analyzer may extract a subgraph of the identified
molecular entities from a global molecular entity graph. At step
528, the analyzer may augment the extracted subgraph to include
additional molecular entities and inter-connections. At step 530,
the analyzer may present the extracted subgraph to the user.
[0146] Still referring to FIG. 5B and in more detail, at step 522,
an analyzer executed by a computing device may receive an
identification of an indication or side effect. In some
embodiments, the analyzer may receive the indication from an
input/output module of the computing device. A user may select or
enter the indication or side effect into an input interface, such
as an application interface or web page interface. In many
embodiments, the user may use an application on a second computing
device to enter or select the indication, and the second computing
device may transmit the entered indication to the input/output
module of the computing device.
[0147] At step 524, in some embodiments, the analyzer may identify
one or more molecular entities known to be associated with the
selected or identified indication. Identifying a molecular entity
known to be associated with the selected or identified indication
may comprise, in some embodiments, retrieving adverse event data
associated with the selected or identified indication. As discussed
above, adverse event data associated with the indication may
comprise one or more adverse event records including identification
of consumed medications. In some embodiments, the medications in
adverse event records may be identified in or linked to
corresponding molecular entity information, such as via a
medication information database. Accordingly, by identifying an
indication, then medications associated with the indication, and
then molecular entities such as protein targets associated with the
medications, the analyzer may identify molecular entities
associated with the indication. In some embodiments, such as where
an adverse event database comprises medication information as
discussed above, adverse event records may comprise molecular
entity information, and thus, the analyzer may directly identify
medications associated with the indication.
[0148] As discussed above, in some embodiments, the analyzer may
generate a list of identified molecular entities. Such list may be
ordered through various statistical techniques, including PRR,
regularized PRR, logistic regression, or other means. In many
embodiments, the analyzer may include in the list only entities
appearing in adverse event records at a greater rate than a
predetermined percentage or number threshold or corresponding to
medications appearing in adverse event records at a greater rate
than the predetermined percentage or number threshold. This may
help reduce false positives and incidental, unrelated signals.
[0149] At step 526, the analyzer may extract a subgraph of the
identified molecular entities from a global molecular entity graph.
Extracting the subgraph may comprise identifying a network
comprising each of the identified molecular entities and augmenting
the network at step 528 with one or more additional entities and/or
connections, using any of the techniques discussed above. For
example, in some embodiments, extracting the subgraph may comprise
selecting pairs of the identified molecular entities and performing
a shortest path analysis to identify one or more intermediate
entities to be included in the subgraph. In other embodiments,
extracting the subgraph may comprise scoring additional nodes in
the network and adding the nodes to the subgraph based on node
scores being above a predetermined threshold. As discussed above,
nodes may be scored based on their relationship to the indication,
their relationship to an organ associated with the indication,
their relationship to a pathway associated with the indication,
their relationship to other nodes external to the subgraph or
internal to the subgraph (for example, decreasing the score of a
node with large numbers of connections to nodes not included in the
subgraph or increasing the score of a node with large numbers of
connections to nodes included in the subgraph), or other similar
relationships. In some embodiments, extracting the subgraph may
comprise scoring pre-defined pathways in the global molecular
entity graph with respect to their coverage of the identified
molecular entities and merging high scoring pre-defined pathways to
generate the subgraph network. Accordingly, in many embodiments,
steps 526 and 528 may be considered as combined steps of extracting
a subgraph based on the identified molecular entities and
augmenting the subgraph with additional nodes using the techniques
discussed herein.
[0150] At step 530, in some embodiments, the analyzer or an output
module connected to the analyzer may present the extracted and
augmented subgraph to a user. In some embodiments, the subgraph may
be presented as a visual graph. In many such embodiments, the
visual graph may be generated by a display module, as discussed
above. For example, the display module may generate a visual graph
of the molecular entities and interconnections as an image, and may
relocate entities as necessary to avoid intersecting connections.
In some embodiments, the display module may generate an interactive
image allowing entities to be selected for additional information,
moved or highlighted, or otherwise manipulated. In some
embodiments, the subgraph may be presented as an index or array of
molecular entities and connected entities. In a further such
embodiment, entities in the subgraph may be ordered based on number
of connections to other entities in the subgraph, identifying
entities that may be most important to the selected indication.
[0151] In some instances, activating a pathway or protein may
result in different side effects or adverse events than
deactivating the pathway or protein. Using the multivariate
analysis techniques discussed herein, these differences may be
readily examined by extracting, from a subset of adverse event data
associated with a pathway or protein, a further subset of adverse
event data based on whether a drug was an agonist or activator of
the protein or pathway, or whether the drug was an antagonist or
inhibitor of the protein or pathway. For example, referring briefly
to FIG. 5C, illustrated is an example diagram of an embodiment of a
subset of a global entity graph associated with a pathway 550. The
subset may be extracted from a global molecular entity graph using
any of the techniques discussed above. In some embodiments, the
extracted graph may comprise one or more molecular entities 552.
Some of the molecular entities may comprise entities 554a-554c that
are known to be activated or inactivated by agonist or antagonist
drugs. For example, a medication information database may indicate
that a first molecular entity 554a is activated by a first
medication, or that a second molecular entity 554b is inactivated
by a second medication. In some embodiments, a molecular entity may
be activated by a first medication and inactivated by a second
medication. Thus, in many embodiments, a pathway or protein may be
activated by one or more medications and deactivated by one or more
medications. By comparing subsets of adverse event data associated
with the pathway or protein based on whether the patient
experiencing the adverse event consumed an agonist or antagonist, a
side effect profile specific to activating or inactivating the
pathway or protein may be generated, and compared to general
adverse event data for the pathway or for a different activating
state to generate distinct adverse event comparison profiles.
[0152] Referring now to FIG. 5C, illustrated is a flow chart of an
embodiment of a method for extracting and comparing subsets of
adverse event data based on activation state of a molecular entity.
In brief overview, at step 570, a multivariate analyzer may
receive, from a user, an identification of a molecular entity. In
some embodiments, the entity may comprise a pathway, while in other
embodiments, the entity may comprise a protein, or any other
entity. At step 572, the analyzer may retrieve, from a medication
information database, an identification of one or more medications
affecting the pathway or entity. At step 574, the analyzer may
identify a subset of the one or more medications that are agonists
or activators of the entity or one or more entities of the pathway,
or a subset of antagonists or inhibitors of the entity or one or
more entities of the pathway. At steps 576, the analyzer may
retrieve, from an adverse event database, adverse event data
records including the identified subset of agonists or antagonists.
In some embodiments, steps 574-576 may be repeated. In other
embodiments, adverse event data records may be retrieved for the
medications identified at step 572, to compare an overall side
effect profile with an activation state profile. At step 578, the
extracted records for different subsets or for the entire set of
identification medications may be compared to identify one or more
differences in the adverse event profiles for the activation
states.
[0153] Still referring to FIG. 5D and in more detail, in some
embodiments, at step 570, an analyzer may receive an identification
of a molecular entity from a user, such as a pathway or protein. In
some embodiments, the analyzer may receive the identification via a
web interface or application interface, from a remote computing
device operating on behalf of the user, or from an input device
connected to the computing device executing the analyzer. In many
embodiments, the analyzer may receive an identification of a
pathway, and may then retrieve from a global molecular entity graph
or a molecular entity information database, an identification or
subset of entities associated with the pathway, using any of the
techniques discussed herein.
[0154] At step 572, the analyzer may retrieve, from a medication
information database, an identification of medications associated
with the entity. For example, in one embodiment in which the entity
is a protein, the analyzer may retrieve an identification of
medications known to affect the protein. In another embodiment in
which the entity is a pathway, the analyzer may identify, from the
global molecular entity graph or an entity database, a set of
entities, including proteins, associated with the pathway. The
analyzer may then retrieve, from the medication information
database, an identification of medications known to affect the set
of entities associated with the pathway.
[0155] At step 574, in some embodiments, responsive to a request
from the user, the analyzer may identify a subset of the
medications responsive to their activation or inactivation of one
or more of the entities of the pathway or an identified protein.
For example, in one embodiment, a user may request to identify
adverse event data based on activation of the pathway, and the
analyzer may identify a subset of the medications that are agonists
or activators of entities of the pathway. In another embodiment,
the user may request to identify adverse event data based on
inhibition of the pathway, and the analyzer may identify a subset
of the medications that are antagonists or inhibitors of entities
of the pathway. In many embodiments, whether a medication is an
agonist or antagonist of an entity may be identified in a
medication information database. In some embodiments in which a
medication is an agonist of one entity in the pathway and an
antagonist of another entity of the pathway, such medications may
be excluded from the identified subset. In other embodiments, such
medications may be included in the identified subset.
[0156] At step 576, the analyzer may retrieve, from an adverse
event database, adverse event data associated with the identified
subset of medications. In some embodiments, retrieving the adverse
event data may comprise retrieving adverse event records for a
medication in the identified subset of medications, while in other
embodiment, retrieving the adverse event data may comprise querying
a database for records associated with the medication. In some
embodiments, the analyzer may retrieve adverse event records of
patients only taking medications in the identified subset of
medications. In other embodiments, the analyzer may retrieve
adverse event records of patients taking medications in the
identified subset of medications and other medications unrelated to
the pathway, but excluding medications with the other activation
state of the pathway. For example, for a request for adverse event
data associated with activating a pathway, the analyzer may
retrieve adverse event records of patients taking any medication
identified as an agonist for a protein in the pathway, but
excluding any adverse event records of patients taking any
medication identified as an antagonist for a protein in the
pathway. This may be done to exclude adverse event data associated
with patients who are consumed both activating and inhibiting
medications.
[0157] In some embodiments, it may be more helpful to identify
adverse event records associated with activating or inhibiting a
plurality of molecular entities in a pathway. For example,
inhibiting one protein in a pathway may not have the effect of
inhibiting the entire pathway. Accordingly, in some embodiments,
the analyzer may identify a plurality of molecular entities in a
pathway, and may identify which medication in the identified subset
of medications activates or inactivates which of the plurality of
molecular entities. In such embodiments, the analyzer may retrieve
adverse event records for patients consuming one or more
medications, such that all of the identified entities was activated
or inactivated by the medications. For example, in one such
embodiment in which a first protein is activated by a first
medication, and a second protein is activated by a second
medication, the analyzer may retrieve only adverse event records
associated with patients consuming both medications. Similarly, if
a third medication activates both proteins, the analyzer may
retrieve adverse event records associated with patients consuming
the third medication. Thus, the analyzer may build a side effect
profile for patients who have, through one or more medications,
activated or inactivated all of the identified entities in the
pathway. In some embodiments, all of the entities may be
identified, while in other embodiments, certain entities of
interest may be identified. Additionally, though discussed in terms
of pure activation or inactivation states, the above techniques may
be applied to mixed activation or inactivation states of a
plurality of entities. Thus, in one example embodiment, the
analyzer may retrieve adverse event of patients taking a medication
that activated a first protein and inhibited a second protein, or a
first medication that activated the first protein and a second
medication that inhibited a second protein, allowing complex
analyses.
[0158] In many embodiments, steps 574-576 may be repeated for
different activation states, such as for activating a pathway vs.
inhibiting the pathway. In some embodiments, adverse event data may
be retrieved for all medications associated with the pathway,
regardless of activation state. This may be done to provide a
control group or allow comparisons to a particular activation
state.
[0159] In some embodiments, at step 578, the analyzer or a display
module may display side effect profiles or adverse event profiles
associated with the one or more sets of adverse event data
retrieved at step 576. Such profiles may comprise identifications
of adverse events experienced by patients in the extracted subset
of records, including identifications of adverse events over time,
proportional reporting rates, an ordered list of medications, an
ordered list of indications, an ordered list of outcomes, or any
other data. In some embodiments, the analyzer may generate a
difference profile or identify one or more differences between two
profiles. For example, the analyzer may identify indications in
different positions or percentages between two profiles, identify
differences in the rates of adverse events, or perform other
comparisons. Such difference profiles or differences may further be
displayed to the user, allowing investigation into adverse event
differences.
[0160] Adverse event data may also be used to predicatively
identify unknown targets for medications. Because adverse events
may be due to physiological reactions from interaction of molecular
entities with pharmaceutical compounds, a "backwards" analysis of
observed adverse event data may enable identification of molecular
entities previously unknown to interact with the pharmaceutical
compound. Referring now to FIG. 6A, illustrated is a diagram of a
method of utilizing side effect profile dissimilarities to identify
likely unknown targets of a medication. A first medication may have
a first side effect profile 602 comprising a statistical index of
one or more side effects experienced by patients or clinical trial
participants consuming the medication, in some embodiments, sorted
by frequency or percentage of occurrence, as discussed above. A
second, similar medication, may have a second side effect profile
604 that may share some, but not all, characteristics with the
first side effect profile 602. In some embodiments, the second
similar medication may comprise a second medication in the same
drug class as the first medication, while in other embodiments, the
second similar medication may comprise a second medication with an
identified known target shared with the first medication, or known
to be affecting the same molecular entity as the first medication.
In some embodiments, the second side effect profile 604 may include
one or more different side effects from the first side effect
profile 602, or may include different frequencies or percentages of
occurrence for one or more side effects from those of the first
side effect profile 602. A multivariate analyzer may generate a
difference profile 606 that identifies differences between the
first side effect profile 602 and the second side effect profile
606. For example, a first medication such as lapatinib, may have a
first side effect profile 602 that includes rash as a side effect
at a very high rate, and may be known to bind to Human Epidermal
Growth Factor Receptor 2 (HER2). A second medication may be
selected that also binds to HER2, such as Herceptin, which may have
a second side effect profile 604 that does not include rash as a
side effect or includes rash only at a very low frequency.
Accordingly, an analyzer may generate a difference profile or
subset of the first medication side effect profile 606 that
includes rash at a high frequency.
[0161] The analyzer may compare the difference profile 606 to other
medication side effect profiles to identify another medication that
includes the identified differences in its side effect profile 608.
In some embodiments, the analyzer may limit the comparison to other
medications in the same drug class or type, such as kinase
inhibitors. For example, given a difference profile 606 including
rash at a high frequency, the analyzer may identify that rash is
also commonly associated with medications such as gefitinib and
erlotinib. Known targets of the identified other medication may
then be indicated as likely targets of the first medication. For
example, Epidermal Growth Factor Receptor (EGFR) is a known target
of gefitinib and erlotinib (as well as being a known target of
lapatinib, but not Herceptin). If it was not known that lapatinib
bound to EGFR, comparison of its difference side effect profile to
the side effect profiles of gefitinib or erlotinib would indicate
that EGFR is a likely target of lapatinib. Thus, through side
effect profile comparisons and difference profiles,
previously-unknown affected molecular entities for medications may
be quickly identified for confirmation through targeted
research.
[0162] Referring now to FIG. 6B, illustrated is a flow chart of an
embodiment of a method for identifying unknown likely targets of a
first medication via comparison of adverse event data. In brief
overview, at step 622, an analyzer may receive an identification of
a first medication. At step 624, the analyzer may identify a
second, similar medication. At step 626, the analyzer may retrieve
side effect profiles for the first medication and the second
medication. At step 628, the analyzer may generate a difference
profile for the first medication. At step 630, the analyzer may
identify a third medication with a side effect profile similar to
the difference profile. At step 632, the analyzer may retrieve a
list of molecular entities or targets associated with the third
medication. In some embodiments, steps 630 and 632 may be repeated
for a plurality of medications. At step 634, the analyzer may
present the retrieved list as potential targets of the first
medication.
[0163] Still referring to FIG. 6B and in more detail, at step 622,
an analyzer executed by a computing device may receive an
identification of a first medication. In some embodiments, the
analyzer may receive the identification of the first medication
from an input/output module of the computing device. A user may
select or enter the medication into an input interface, such as an
application interface or web page interface. In many embodiments,
the user may use an application on a second computing device to
enter or select the medication, and the second computing device may
transmit the entered medication to the input/output module of the
computing device.
[0164] At step 624, the analyzer may identify a similar second
medication. In some embodiments, the second similar medication may
comprise a second medication in the same drug class as the first
medication, while in other embodiments, the second similar
medication may comprise a second medication with an identified
known target shared with the first medication, or known to be
affecting the same molecular entity as the first medication. In
still other embodiments, the second similar medication may comprise
a medication structurally similar to the first medication.
[0165] At step 626, the analyzer may retrieve a first side effect
profile associated with the first medication and a second side
effect profile associated with the second medication. As discussed
above, a side effect profile may comprise a statistical index of
one or more side effects experienced by patients or clinical trial
participants consuming the medication. The analyzer may retrieve
each side effect profile by searching an adverse event database for
adverse event records including the medication. In some
embodiments, the analyzer may sort each side effect profile by
frequency or percentage of occurrence of each side effect, as
discussed above.
[0166] At step 628, the analyzer may generate a difference profile
that identifies differences between the first side effect profile
and the second side effect profile. In some embodiments, generating
a difference profile may comprise subtracting a frequency of
occurrence of a side effect in the second side effect profile from
a frequency of occurrence of the side effect in the first side
effect profile. In other embodiments, generating a difference
profile may comprise discarding each side effect in the first side
effect profile for which the second side effect profile includes
the side effect at a frequency of occurrence within a predetermined
threshold. For example, if a first side effect profile includes a
first side effect with an 80% occurrence rate, and the second side
effect profile includes the first side effect with a 75% occurrence
rate, and the predetermined threshold is 10%, then the first side
effect may be discarded from the resulting difference profile.
[0167] At step 630, the analyzer may identify a third medication
with a third side effect profile similar to or comprising the
difference profile. In one embodiment, a side effect profile is
similar to the difference profile if the side effect profile
includes one or more of the side effects in the difference profile
at a frequency of occurrence within a predetermined threshold of
the value in the difference profile. For example, if the difference
profile includes a side effect with an 80% occurrence rate, and the
side effect profile includes the side effect with a 65% occurrence
rate, and the predetermined threshold is 20%, the side effect
profile may be considered similar to the difference profile. In
such embodiments, a predetermined threshold for similarity between
the difference profile and the side effect profile may be the same
as, or different from the predetermined threshold discussed above
for generating the difference profile. In other embodiments, the
analyzer may subtract a frequency of occurrence of a side effect in
the difference profile from a frequency of occurrence of the side
effect in the third side effect profile, and if the result is zero
or within a predetermined value, the profiles may be identified as
similar. In many embodiments, either of the difference profile or
the third side effect profile may include additional side effects
not included in the corresponding other profile. Nonetheless, a
profile may be identified as similar based on similar values for
identified side effects. In some embodiments, similarities must
exist between a plurality of side effect occurrence frequencies
before a third side effect profile may be identified as
similar.
[0168] In one embodiment, the analyzer may identify the third
medication by searching an adverse event database for all records
including a first side effect in the difference profile. For each
medication in the identified records, the analyzer may then search
the adverse event database for all adverse events associated with
the medication. The analyzer may then identify a frequency of
occurrence of the first side effect by identifying the percentage
of adverse event records for the medication which include the first
side effect. This process may be repeated iteratively for
additional medications and/or additional side effects to build a
side effect profile for the medication. Additionally, in many
embodiments, the analyzer may pre-generate side effect profiles for
medications, allowing identification at step 630 to be performed
quickly using the pre-generated profiles. In some embodiments, the
analyzer may limit the comparison and identification to other
medications in the same drug class or type.
[0169] At step 632, the analyzer may retrieve a list of targets
associated with the identified third medication. In some
embodiments, as discussed above, the analyzer may retrieve the list
of targets from a medication information database. In many
embodiments, steps 630-632 may be repeated iteratively to identify
additional medications with side effect profiles similar to the
difference profile.
[0170] At step 634, the analyzer may present the retrieved list of
targets as potential unknown targets of the first medication. In
some embodiments, the analyzer may remove from the retrieved list
any known targets of the first medication, while in other
embodiments, the analyzer may add any known targets of the first
medication not included in the retrieved list. In some embodiments
in which steps 630-632 are repeated for a plurality of medications,
the analyzer may generate a union of the retrieved lists of
targets, while in other embodiments, the analyzer may take an
intersection of the retrieved lists of targets. This may be done to
increase the number of potential targets or decrease the number of
potential targets, respectively. For example, utilizing an
intersection of lists of targets of medications identified as
having side effect profiles comprising or at least partially
similar to the difference profile may result in removing targets
that are associated with less than all of the medications, and thus
may not contribute to the occurrence of the side effect.
[0171] Molecular entity interactions, even for a single drug, may
be complex. With multiple drugs consumed by a patient, and
information about each medication in a text-based form, it may
difficult to identify interactions or treatment redundancies. As a
result, physicians tend to use only known drug-drug interactions in
considering prescriptions. Furthermore, in many instances, patients
may be prescribed drugs with redundant interactions, resulting in
potential unpredictable side effects. For example, a first drug may
need to be catalyzed by a first enzyme into a bioavailable
compound, and the drug dosage may be calculated based on normal
levels of the enzyme. If a patient is prescribed a second drug that
is also catalyzed by the first enzyme, the enzyme may not be
available in sufficient amounts to catalyze both drugs. In such
cases, the first drug may not be present in sufficient amounts of
its bioavailable form to treat the indication, or may be present in
its non-catalyzed form at potentially toxic levels. Even if
non-toxic, in some instances, the combination of drugs may result
in one being excreted unprocessed by the patient, resulting in
potentially expensive waste. Accordingly, it may be useful to
physicians and patients self-managing care, as well as insurance
companies or health care providers, to have an intuitive tool for
examining molecular dependencies of a patient's prescription load,
including all drugs, and the targets, carriers, metabolizing
enzymes, transporters, pathways, and other molecular entities
involved with each medication.
[0172] Referring now to FIG. 7A, illustrated is a screenshot of an
example of an embodiment of a molecular entity dependency graph
that provides intuitive identification of redundancies and
molecular interactions between medications in a patient's
prescription load. In some embodiments, a display module,
embodiments of which are discussed above, may generate the
dependency graph responsive to identification of a patient's
prescription load. The display module and/or an analyzer may
retrieve, from a medication information database, an identification
of molecular entities associated with each medication prescribed to
the patient and their associations and inter-associations for
display in the dependency graph. In some embodiments, the
dependency graph may comprise a radial graph of a plurality of
molecular entities as radial entries. The molecular entities may be
grouped into sub-groups of medications 702 prescribed to a patient;
targets 704 of the medications 702; enzymes 706 catalyzing the
medications 702; membrane transporters 708 of the medications 702;
carriers 710 such as a carrier protein utilized by the medications
702; and/or pathways 712 associated with the medications 702.
Molecular entities in the radial graph may be visually linked by
entity associations 714. In some embodiments, the radial entries
may include mapped mutational information for the patient, such as
identified genetic variants for the patient. Such variants may be
linked with other molecular entities in the graph, for example,
corresponding protein targets 704 whose activation is modified by
the variant. Although shown linking entities 704-712 to medications
702, in some embodiments, pathways 712 may be visually linked to
other molecular entities such as target proteins 704 associated
with the pathway. As shown, in many embodiments, entity associates
714 may comprise splines, and may be generated to be grouped with
other associations between a first subcategory of entities and a
second subcategory of entities. This may help to visually separate
out entity associations, as opposed to depicting entity
associations with straight lines from one radial entry to another.
For example, a straight line from a first medication 702 to a first
carrier 710 may intersect with a straight line from a second
medication 702 to a target 704, potentially visually confusing the
two lines. Additionally, through the use of splines as shown in
FIG. 7A, a plurality of entity associations 714 from one subgroup
of entities to another subgroup of entities may be substantially
parallel until splitting out at each end, reducing visual
confusion.
[0173] In some embodiments, the dependency graph may be
interactive. For example, a display module may provide the
dependency graph to an input/output module, such as a web server or
server-side application, which may allow user interaction with the
graph. In some embodiments, the user may select a first molecular
entity, such as by clicking on the first molecular entity. In one
such embodiment, the display module and/or input/output module may
hide entity associations 714 not connected to the selected
molecular entity. Referring now to FIG. 7B, illustrated is a
screenshot of an example of an embodiment of a dependency graph
allowing user interaction. As shown, in such embodiments, a user
may select an entity 716, and a subgroup of entity associations 714
associated with only that entity 716 may be displayed. In some
embodiments, radial entries connected to the subgroup may be
highlighted or in darker text, as shown, while other radial entries
may be faded or presented in lighter text, to visually distinguish
associated entities and non-associated entities.
[0174] Referring briefly to FIG. 7C, illustrated is another
screenshot of an example of an embodiment of a dependency graph
allowing user interaction. As shown, in some embodiments, the
display module and/or input/output module may be configured to
allow a user to select a plurality of entities 716a-716b. The
display module may display corresponding entity associations
714a-714b for each of the plurality of selected entities, allowing
direct comparison of two molecular entities, such as two
medications 702. In some embodiments, the display module may show
entity associations 714a for a first selected entity 716a in a
first color or shade, and entity associations 714b for a second
selected entity 716b in a second color or shade. This may be
particularly helpful when each selected entity is associated with
the some of the same other entities. For example, as shown in FIG.
7C, the two selected entities 716a-716b have associations with many
of the same molecular entities. In a further embodiment,
associations connected to a first selected entity may be displayed
in a first color, associations connected to a second selected
entity may be displayed in a second color, and display module may
merge the colors of overlapping associations to display a third
color representing shared associations. Returning briefly to FIG.
7A, as shown, in some embodiments, the display module may be
configured to optionally display selected entities and
corresponding associations in a highlighted or darker color, and
non-selected entities and corresponding associations in a
non-highlighted or lighter color. In one such embodiment, a user
need not click to select an entity, but rather the display module
may highlight entities and corresponding associations 714 as the
user moves a cursor over each radial entry.
[0175] In some embodiments, the dependency graph may allow a user
to easily identify redundant medications. For example, a patient
may be prescribed a first pain reliever and a second pain reliever,
which may act in a similar way. The two medications may both be
associated with many of the same molecular entities. If the two
medications target different proteins, but utilize the same
enzymes, transporters, and pathways, a simple target comparison may
not identify a potential interaction (as well as potentially
missing off-target interactions with proteins) that may cause an
adverse effect or reduced efficacy of one or both medications. As
the dependency graph intuitively highlights such interactions, a
patient self-managing care or an insurance provider who lacks an
advanced biology education may still be able to identify potential
concerns or reduced efficacies for further discussion with a
physician. In some embodiments, this may also allow identification
of drugs with similar or identical interactions, raising questions
of whether both drugs are needed for treatment. Reducing or
eliminating one may reduce patient or insurance provider cost,
increase efficacy of the remaining drug or drugs, and reduce
unpredictable effects due to drug-drug interactions.
[0176] In some embodiments, adverse event data related to dangerous
or efficacious combination therapies may be used with
patient-specific genomic information to optimize or de-risk therapy
for the patient. For example, in one embodiment, adverse event data
may indicate that a combination therapy targeting a first protein
(protein A) with a first medication (drug A) and targeting a second
protein (protein B) with a second medication (drug B) may have a
high rate of adverse side effects and/or negative outcomes. In
addition to recognizing that drug A and drug B should not be
co-prescribed to a patient, by identifying patient variants
associated with the molecular entities protein A and protein B, it
may even be determined that either of drug A or drug B should not
be prescribed to the patient alone. For example, if the patient has
a genetic mutation that inactivates protein B and drug B is an
antagonist (such that normal operation of drug B blocks binding of
protein B, for example), then physiologically, the patient's system
may be equivalent to a normal patient consuming drug B.
Accordingly, prescribing drug A alone to the patient may
unintentionally result in adverse events normally seen through the
combination of drug A and drug B.
[0177] Similar relationships may result based on whether the
mutation is inactivating or activating of the protein, and whether
the drug is an agonist or antagonist. For example, in an embodiment
in which drug A is an agonist, drug B is an agonist, and the
combination of drug A and drug B results in an adverse event:
[0178] a. If the patient has an activating mutation for protein A,
then drug B should be contraindicated. [0179] b. If the patient has
an inactivating mutation for protein A, then drug B may be
indicated. [0180] c. If the patient has no mutation (i.e. a
wildtype) for protein A, then drug B may be indicated. [0181] d. If
the patient has an activating mutation for protein B, then drug A
should be contraindicated. [0182] e. If the patient has an
inactivating mutation for protein B, then drug A may be indicated.
[0183] f. If the patient has no mutation (i.e. a wildtype) for
protein B, then drug A may be indicated.
[0184] Similarly, if drug A is an antagonist, drug B is an
antagonist, and the combination of drug A and drug B results in an
adverse event: [0185] a. If the patient has an inactivating
mutation for protein A, then drug B should be contraindicated.
[0186] b. If the patient has an activating mutation for protein A,
then drug B may be indicated. [0187] c. If the patient has no
mutation (i.e. a wildtype) for protein A, then drug B may be
indicated. [0188] d. If the patient has an inactivating mutation
for protein B, then drug A should be contraindicated. [0189] e. If
the patient has an activating mutation for protein B, then drug A
may be indicated. [0190] f. If the patient has no mutation (i.e. a
wildtype) for protein B, then drug A may be indicated.
[0191] Likewise, if drug A is an agonist, drug B is an antagonist,
and the combination of drug A and drug B results in an adverse
event: [0192] a. If the patient has an activating mutation for
protein A, then drug B should be contraindicated. [0193] b. If the
patient has an inactivating mutation for protein A, then drug B may
be indicated. [0194] c. If the patient has no mutation (i.e. a
wildtype) for protein A, then drug B may be indicated. [0195] d. If
the patient has an inactivating mutation for protein B, then drug A
should be contraindicated. [0196] e. If the patient has an
activating mutation for protein B, then drug A may be indicated.
[0197] f. If the patient has no mutation (i.e. a wildtype) for
protein B, then drug A may be indicated.
[0198] Similarly, if drug A is an antagonist, drug B is an agonist,
and the combination of drug A and drug B results in an adverse
event: [0199] a. If the patient has an inactivating mutation for
protein A, then drug B should be contraindicated. [0200] b. If the
patient has an activating mutation for protein A, then drug B may
be indicated. [0201] c. If the patient has no mutation (i.e. a
wildtype) for protein A, then drug B may be indicated. [0202] d. If
the patient has an activating mutation for protein B, then drug A
should be contraindicated. [0203] e. If the patient has an
inactivating mutation for protein B, then drug A may be indicated.
[0204] f. If the patient has no mutation (i.e. a wildtype) for
protein B, then drug A may be indicated.
[0205] Although discussed in terms of a pair of interacting drugs,
in many embodiments, the analysis may be extended to any number of
interacting medications. For example, if it is observed that four
drugs prescribed in combination results in a high rate of adverse
events, patient genetic variant information relating to the
molecular entities targeted by each drug may be analyzed to
determine if a single drug, pair of drugs, or trio of drugs should
be contraindicated, responsive to corresponding variants for three
targets, two targets, or one target respectively. In other
embodiments, a drug may have a plurality of target proteins, and
the system may contraindicate other drugs responsive to the patient
having corresponding variants for each protein. Thus, for example,
if drug A is an antagonist of proteins A and C, in some
embodiments, drug B may be contraindicated only if the patient has
inactivating mutations for both of proteins A and C.
[0206] Referring now to FIG. 8, illustrated is a flow chart of an
embodiment of a method for personalized de-risking of medications
based on genomic information of a patient and adverse event data of
combination therapies. In brief overview, at step 802, an analyzer
executed by a computing device may receive an identification of a
genomic variant of a patient altering activity of a first protein.
At step 804, the analyzer may identify a first medication targeting
the first protein. At step 806, the analyzer may receive an
identification of a second medication targeting the second protein
considered as a potential medication to be prescribed. At step 808,
the analyzer may identify a likelihood of an adverse event
occurring from co-medication of the first medication and second
medication. At step 810, the analyzer may determine that an adverse
event is likely to occur for the patient. At step 812, the analyzer
may contraindicate the second medication.
[0207] Still referring to FIG. 8 and in more detail, in one
embodiment, an analyzer may receive an identification of a genomic
variant of a patient altering activity of a first protein. In one
embodiment, the analyzer may receive a list of variants of the
patient. In some embodiments, in which the analyzer receives a
plurality of variants, the analyzer may select a variant and repeat
the method of FIG. 8 iteratively. In some embodiments, the list of
variants may explicitly identify corresponding proteins, while in
other embodiments, the analyzer may retrieve identifications of one
or proteins corresponding to each variant from a genetic
information database. In some embodiments, the analyzer may receive
the identification of genomic variants from an input/output module,
as discussed above. In some embodiments, a user of a second
computing device may transfer or upload a list of variants to the
analyzer, such as via a web interface or application.
[0208] At step 804, the analyzer may identify a first medication
targeting the first protein. In one embodiment, the analyzer may
search a medication information database for medications identified
as targeting the first protein. In another embodiment, the analyzer
may utilize an adverse event database that includes in adverse
event records identification of target proteins targeted by
medications consumed by the person experiencing the adverse event.
The analyzer may query the database to retrieve a list of
medications associated with the first protein.
[0209] At step 806, the analyzer may receive an identification of a
second medication for consideration for prescription to the
patient. The second medication may target a second protein. In some
embodiments, a user may select a second medication from a list of
medications, while in other embodiments, the user may enter a name
or part of a name of a medication through a web interface or
application interface, as discussed above.
[0210] At step 808, the analyzer may determine whether an adverse
event is likely to occur if both the first medication and second
medication are prescribed to a patient. In some embodiments, the
analyzer may query an adverse event database to retrieve an
identification of a number of adverse event records including both
medications as consumed by the person experiencing the adverse
event. The adverse event database may, in some embodiments,
identify a number of times each drug was prescribed or number of
times the combinations of drugs were prescribed, such that the
analyzer may determine a ratio of adverse event occurrences to
total number of prescriptions. In other embodiments, such as where
such non-adverse event data is unavailable, the analyzer may query
the adverse event database to determine a ratio of serious outcomes
to total number of adverse events for the combination of
medications. For example, if a serious outcome, such as death or
disability occurs in the majority of adverse event reports for the
two medications, the combination may be considered to have very
high risk. In comparison, if a serious outcome occurs in only a
slim minority or none of the adverse event reports, with
non-serious outcomes dominating the records, then the combination
may be considered to have a low risk. Thus, in such embodiments,
the analyzer may determine whether an adverse event including a
serious outcome is likely to occur if both the first medication and
second medication are prescribed to a patient.
[0211] At step 810, the analyzer may determine that an adverse
event is likely to occur for the patient if the patient is
prescribed the second medication, responsive to determining that an
adverse event is likely to occur if the patient comedicated with
the first medication and the second medication and that the patient
has a genetic mutation affecting a protein corresponding to
activity of the first medication with the protein. As discussed
above, this determination may be responsive to whether the mutation
is activating or non-activating, and whether the medication is an
agonist or antagonist, respectively.
[0212] At step 812, responsive to determining that an adverse event
is likely to occur for the patient if the patent is prescribed the
second medication, the analyzer may contraindicate the second
medication. In some embodiments, contraindicating the medication
may comprise generating a list of contraindicated medications for
display to the user.
[0213] As discussed above, in many embodiments, steps 806-812 may
be iteratively repeated for additional medications, to de-risk a
patient's prescription load. Accordingly, at step 808, the analyzer
may search for adverse events with a pair of medications, trio of
medications, or more medications, responsive to the number of
medications identified by the user. Additionally, in some
embodiments, steps 806-812 may be iteratively repeated for
alternate, similar medications to the identified second medication.
For example, in one such embodiment, having determined that the
patient will likely experience an adverse event upon consuming the
identified second medication, the analyzer may repeat steps 806-812
for a third medication in the same drug class or type as the second
medication. For example, if the analyzer identifies that, due to a
genetic mutation in a patient and based on adverse event data, the
patient will likely experience an adverse event upon consuming
gefitinib, the analyzer may repeat the analysis for erlotinib,
another kinase inhibitor. If the analyzer determines that the third
medication may not induce an adverse effect in the patient, the
analyzer may identify the third medication as a potential alternate
prescription. This may allow the system to automatically identify
safer alternative medications for consideration.
[0214] Furthermore, in a similar embodiment, patient genomic
information may be used to determine if, for example, a mutation in
a protein will decrease the binding affinity of a specific drug,
leading to the drug building up to toxic levels and causing adverse
events if consumed by the patient. Such proteins may comprise any
proteins that interact with and/or are critical to the mode of
action, metabolism, or passage of the drug through the patient
system, or otherwise directly interact with the drug at the
pharmacokinetic or pharmacodynamics levels. Accordingly, in such
embodiments, the model of the drug's passage and mode of action
within the patient system may be analyzed against patient variant
information. This may allow identification of mutations in genes
that do not directly interact with the drug, but whose functions
regulate the activity of a gene or protein that does. Similarly, in
some embodiments, the above methods and systems may be used to
identify mutations in genes that affect the expression or binding
affinities for off-target proteins that may lead to adverse events.
For example, over-expressed off-target proteins may act as
"molecular sinks" for a drug, decreasing the therapeutic efficacy
of the medication. Identifying such interactions with the
above-discussed systems may allow contraindication of apparently
unrelated medications, reducing the incidence of previously
unpredictable adverse events.
[0215] Furthermore, by collecting and analyzing patent-specific
genomic information, adverse event profiles may be generated based
on a genetic mutation. For example, variant identifications of
patients that suffered a specific adverse event may be compared to
identify genetic commonalities, which may be used to potentially
de-risk new patients.
[0216] In another embodiment, homologous family members of proteins
may be identified as likely off-target candidates. For example,
using knowledge about the diseases caused by mutations in these
candidates, the analysis system may predict potential adverse
events induced by consumption of drugs targeting the homologous
family members by the patient.
[0217] In some embodiments, a multivariate analysis system may be
able to reduce false signals in planned clinical trials by
identifying medications to be contraindicated for a cohort. For
example, in many instances, a disease and a side effect may differ
only due to the side effect being drug-induced. Accordingly, the
side effect may be thought of as a drug-induced disease. For
manufacturers and researchers developing new pharmaceuticals, it
may be important during trials to avoid including patients taking
other drugs that may induce the same side effect as the disease in
question. Furthermore, it may be desirable to screen all patient
co-medications for drug interactions at many levels, including on a
molecular basis.
[0218] In some embodiments, it may be desirable to exclude drugs
from a proposed clinical trial with side effect profiles that
include side effects corresponding to a disease that is the subject
of the clinical trial. For example, in one embodiment, if a
proposed clinical trial is examining the effect of drug A in
indication A, but adverse event data indicates that a side effect
corresponding with indication A is also inducible by drug B, then
the analysis system may contraindicate drug B from the clinical
trial. The inclusion of such contraindicated drugs may result in
false negatives, as they have a chance of counteracting any
therapeutic effects of drug A on the disease. In another
embodiment, if a clinical trial is examining the combined effects
of two approved drugs for investigation into potential combination
therapies, the analysis system may be used to examine the safety
profile of the combination and include potential safety issues in
the trial protocol.
[0219] In some embodiments, as discussed above, analysis may be
performed on a molecular basis. For example, in one such embodiment
with a first drug targeting a first protein to be used for a
clinical trial, a multivariate analysis system may retrieve a side
effect profile for the protein, based on adverse event data for all
medications targeting the protein. In other embodiments, molecular
entities functionally related to the protein may be identified, and
side effect profiles for medications targeting those molecular
entities may be retrieved. In many embodiments in which molecular
entity information is integrated into adverse event records as
discussed above, side effect profiles may be generated for the
molecular entities directly, and then medications associated with
high risk entities may be identified for contraindication.
[0220] Referring now to FIG. 9, illustrated is a flow chart of an
embodiment of a method for identifying a medication for
contraindication from a clinical trial of another medication. In
brief overview, at step 902, an analyzer executed by a computing
device may receive an identification of an indication for a
clinical trial. At step 904, the analyzer may retrieve adverse
event data for a side effect corresponding to the indication. At
step 906, the analyzer may generate an ordered list of one or more
medications consumed by patients that experienced the side effect.
At step 908, the analyzer may select one or more medications from
the list, and at step 910 may display the one or more medications
as contraindicated from the clinical trial.
[0221] Still referring to FIG. 9 and in more detail, at step 902,
an analyzer executed by a computing device may receive an
identification from a user of an indication for a clinical trial.
In some embodiments, the user may select or enter the indication
via a web interface or application interface. The user may utilize
the same computing device or a second computing device connected to
the first computing device via a network.
[0222] In some embodiments, at step 904, the analyzer may retrieve
adverse event data for a side effect corresponding to the
indication from an adverse event database. In some embodiments, the
analyzer may query the database for records including the side
effect corresponding to the indication. Such records may comprise
identifications of the side effect and outcome experienced by the
patient, medications consumed by the patient, patient demographic
information, and any other relevant information. In some
embodiments, the records may comprise identifications of molecular
entities corresponding to the medications, while in other
embodiments, such identifications may be in a second medication
information database.
[0223] At step 906, the analyzer may generate a list of medications
identified in each retrieved record. In some embodiments, the
analyzer may count the number of times each medication appears in
the retrieved records in order to order the list via frequency of
appearance. In some embodiments, each medication may be scored in
the list or have an associated frequency value and/or statistical
percentage or rate of appearance. In some embodiments, the analyzer
may determine one or more statistical measures for the medication,
such as reporting odds ratio (ROR), incidence rate ratio, or
proportional reporting ratio (PRR), or may apply one or more
statistical algorithms, such as a multi-item gamma poisson shrinker
(MGPS) algorithm.
[0224] At step 908, the analyzer may identify one or more
medications from the list to be contraindicated. In some
embodiments, the analyzer may select all medications in the list to
be contraindicated, while in other embodiments, the analyzer may
select a subset of medications in the list. For example, in one
embodiment, the analyzer may select all medications in the list
associated with a particular organ that is the subject of the
clinical trial. In another embodiment, the analyzer may select all
medications in the list of a particular drug class or type. In
still another embodiment, the analyzer may select medications
having a statistical value or ratio above a predetermined
threshold. For example, the analyzer may select all medications
having a PRR or MGPS value over 2 and discard other medications
from the list.
[0225] At step 910, the analyzer may display the identified one or
more medications as medications to be contraindicated from the
trial. In some embodiments, the analyzer may display one or more
statistically likely side effects that may be induced by each
contraindicated medication.
[0226] In some embodiments, the analyzer may further identify
combinations of medications to be contraindicated for the trial.
For example, in some instances, a side effect corresponding to the
indication may appear when two medications are consumed by a
patient, but not when either is consumed alone. From the adverse
event data, the analyzer may identify that each medication is
included individually in adverse event records for the side effect.
The analyzer may then compare pairs or sets of identified
medications for frequency of co-appearance within each retrieved
record. Medications that appear together at a high frequency within
the adverse event records may be identified as a contraindicated
combination.
[0227] In some embodiments, a multivariate analysis of adverse
event data may be further used to identify novel combination
therapies for research by generating cohorts of patients conforming
to specific clinical and treatment variables. Cohorts can be
compared in terms of patient outcomes, with variables examined for
potential clinical effects. For example, adverse event data for a
first cohort of patients with cancer who have taken an
anti-neoplastic agent may be retrieved and compared to adverse
event data for a second cohort of patients with cancer who have
taken an anti-neoplastic agent plus another class of drug. The sets
of adverse event data for each cohort may be compared to identify
if the other class of drug has any effect on the death rate of
cancer patients across cancer indications. Drugs which appear to
decrease the death rate or are associated with a lower death rate
in adverse event reports may then be potential candidates for
combination therapy. Furthermore, such analysis may be done for any
molecular entity.
[0228] For example, and referring briefly to FIG. 10A, illustrated
is a Venn diagram of an example of an embodiment of defining
cohorts within adverse event data and extracting difference
profiles for a cohort. Adverse event data for an indication 1002
may be retrieved from an adverse event database through a query by
an analyzer. The query may further comprise additional variables to
define cohorts 1004A-1004C or patients defined by the variable, and
adverse event data for each cohort may be retrieved. In many
embodiments, patients may be in multiple cohorts. For example, a
first cohort may be defined as patients who consumed a first drug,
and a second cohort may be defined as patients who consumed a
second drug. Accordingly, patients consuming both drugs may be
placed in both cohorts. Variables for defining cohorts may be of
different types. For example, a first cohort may be defined as
patients who are over a specified age, and a second cohort may be
defined as patients who consumed a medication that was catalyzed by
a specified enzyme. The analyzer may extract a distinct adverse
event profile for a cohort 1006. In some embodiments, the analyzer
may compare adverse event profiles between cohorts to generate a
difference profile, while in other embodiments, the analyzer may
generate a query that excludes members of other cohorts from the
cohort for which the distinct profile is created. In still other
embodiments, the analyzer may retrieve identifications of adverse
event records for each cohort, and then eliminate any records
shared by each cohort. The analyzer may then determine rates of
various outcomes for the records identified in the difference
profile, and may compare this to rates of various outcomes for
other cohorts, or the indication as a whole. Differences in the
rates may thus indicate potential combination therapies.
[0229] Referring now to FIG. 10B, illustrated is a flow chart of an
embodiment of a method for identifying potential combination
therapies for research via adverse event data. In brief overview,
at step 1022, an analyzer may receive an identification of an
indication. At step 1024, the analyzer may retrieve adverse event
data for the identified indication. At step 1026, the analyzer may
receive an identification of a patient cohort. In many embodiments,
the patient cohort may be defined by a molecular entity, while in
other embodiments, the patient cohort may be defined by demographic
information or a genotype. At step 1028, the analyzer may extract a
subset of adverse event data for the patient cohort. In some
embodiments, steps 1026-1028 may be repeated for additional
cohorts. At step 1030, the analyzer may compare the extracted
subsets to generate a collated list of differences between the
patient cohorts. At step 1032, the analyzer or an output module
connected to the analyzer may display the collated list of
differences. Although shown in one order in FIG. 10B, as discussed
above, in some embodiments in which the analyzer uses multivariate
queries with Boolean operations to retrieve adverse event data from
the adverse event database, many of the steps may be collapsed into
a single step.
[0230] Still referring to FIG. 10B and in more detail, in one
embodiment at step 1022, an analyzer may receive an identification
of an indication from a user. In some embodiments, the analyzer may
receive the identification via a web interface or application
interface communicating via an input/output module. As discussed
above, the user may operate an application on the same computing
device as the analyzer, or on a different computing device
communicating with the first computing device via a network.
[0231] At step 1024, in some embodiments, the analyzer may retrieve
adverse event data for the identified indication from an adverse
event database. As discussed above, adverse event data may comprise
records of adverse events experienced by patients, and may identify
an indication for which the patient was being treated or may
identify a side effect experienced by the patient corresponding to
the indication.
[0232] At step 1026, the analyzer may receive an identification of
a first patient cohort. The patient cohort may be defined by a
molecular entity, such as patients consuming a first medication,
patients consuming a medication targeting a first protein, patients
consuming a medication targeting a first pathway, patients
consuming a medication related to a first drug class, etc. In other
embodiments, the patient cohort may be defined by demographic
information, such as age or gender, or may be defined by patients
having specified genetic mutations or wildtypes. In many
embodiments, multiple variables may be used to define a patient
cohort, such as men over 50 being treated for high cholesterol.
[0233] At step 1028, the analyzer may extract a subset of adverse
data experienced by the identified first patient cohort. In some
embodiments, the analyzer may extract data relating to side effects
experienced by the first patient cohort being treated for the
identified indication, while in other embodiments, the analyzer may
extract data relating to patient outcomes of the first patient
cohort. Such data may comprise raw numbers of adverse events for
each side effect and/or outcome, or proportional reporting ratios
or other statistical identifiers for each side effect and/or
outcome. The analyzer may repeat steps 1026-1028 for a plurality of
cohorts with at least one modified variable, such as an included or
excluded molecular entity, changed demographic information,
etc.
[0234] At step 1030, the analyzer may compare the extracted subsets
for different patient cohorts to identify statistical differences
between side effects and/or outcomes between cohorts. In one
embodiment, comparing the extracted subsets may comprise generating
difference values for each statistical value of a side effect
and/or outcome. For example, if 30% of a first cohort is listed as
having died as a result of the indication and/or side effect, and
10% of a second cohort is listed as having died as a result of the
indication and/or side effect, a difference value of -20% may be
identified for the second cohort. In many embodiments, difference
values beyond a predetermined threshold may indicate a potentially
significant result of the modified variable between the cohorts. In
some embodiments, comparing the extracted subsets of adverse event
data may comprise generating an index of side effects and/or
outcomes experienced by the patients and sorting the index by
percentage or raw number. The analyzer may then compare the
positions of individual side effects and/or outcomes within the
generated index for each cohort. In many embodiments, the analyzer
may generate a collated list of one or more statistical differences
between the side effect profiles for each cohort. As discussed
above, in many embodiments, the list may be limited to statistical
differences above a predetermined threshold, such as difference
percentages over a predetermined rate, or altered index positions
greater than a predetermined number.
[0235] At step 1032, the analyzer or a display module or output
module connected to the analyzer may display the generated list of
statistical differences to the user. The list may be used to
identify statistically significant differences in adverse events
experienced by each cohort, and potentially attributable to the
modified variable or variables between the cohorts. This may point
to potential combination therapies for reducing risk or increasing
efficacy of therapy.
[0236] By integrating an adverse event database with molecular
entity information, such as the global molecular entity graph
discussed above, a multivariate analysis system may be able to
predict a likely side effect profile for even new, untested
medications. Specifically, a predicted side effect profile may be
generated based on intersections of side effect profiles of other
medications that affect the same or related molecular entities,
such as the nearby target proteins, involve the same pathways, or
are otherwise similarly related. To generate a predicted side
effect profile for a new drug targeting a novel or previously
un-targeted protein target, an analyzer may query an adverse event
database for records pertaining to patients who have taken drugs or
combinations of drugs that target or affect molecular entities in
the vicinity of the novel target within a global molecular entity
graph. By examining the side effect profiles associated with the
connected targets, one can look for commonalities that might also
be expected with the novel target. For example, referring briefly
to FIG. 11A, illustrated is a graph of an example of a region of an
example embodiment of a global molecular entity graph or molecular
entity network comprising a plurality of molecular entities 1106
connected via functional links. To generate a predicted side effect
profile for a new drug targeting novel target protein 1102, an
analyzer may query an adverse event database for adverse event
records of patients who consumed a first approved drug targeting a
first protein A 1104A; adverse event records of patients who
consumed a second approved drug targeting a second protein B 1104B;
and records of patients who consumed both drugs. Intersections
and/or difference profiles may be generated based on these
retrieved adverse event records to a generate side effect profile
of adverse event records that likely involved the novel target
1102, even if it was not realized at the time. For example, a
patient who consumed both the first drug and second drug targeting
proteins A and B likely affected their processing of the novel
target protein 1102, for example by reducing availability of an
enzyme needed to catalyze the protein 1102, resulting in higher
systemic levels of the protein than normal. In some embodiments,
this may have a similar effect as a novel drug that acts as an
agonist of the protein, for example. Accordingly, side effects
experienced by such a patient may be similar to side effects that
may be experienced by a patient consuming the novel drug.
[0237] Referring now to FIG. 11B, illustrated is a flow chart of an
embodiment of a method for generating a predicted side effect
profile for a medication targeting a novel target. In brief
overview, at step 1122, an analyzer or input module may receive an
identification of a novel drug target. At step 1124, the analyzer
may identify a second target functionally connected to the novel
drug target in a global molecular entity graph. At step 1126, the
analyzer may identify a medication targeting the second target. At
step 1128, the analyzer may retrieve a side effect profile for the
identified medication targeting the second target. In some
embodiments, the analyzer may output the retrieved side effect
profile at step 1132 for display to the user as a predicted side
effect profile of the novel drug target. In many embodiments, the
analyzer may repeat steps 1126-1128 to retrieve side effect
profiles for one or more additional medications targeting the
second target, while in other embodiments, the analyzer may repeat
steps 1124-1128 to identify one or more additional targets and
additional medications. At step 1130, the analyzer may generate an
intersection side effect profile of the retrieved side effect
profiles, and at step 1132, may output the retrieved side effect
profile for display to the user as a predicted side effect profile
of the novel drug target.
[0238] Still referring to FIG. 11B and in more detail, at step
1122, an analyzer executed by a computing device may receive an
identification of a novel drug target from a user. The novel drug
target may comprise a molecular entity, such as a protein, enzyme,
transporter, or other entity that may be known, but not previously
targeted by a medication. Functional relationships or connections
to other molecular entities from the novel drug target may also be
known, such as the inclusion of the novel drug target in a global
molecular entity graph. In some embodiments, the analyzer may
receive the identification of the novel drug target via an
application executed by the computing device used by the user,
while in other embodiments, the analyzer may receive the
identification via a web interface or application interface via a
network from a second computing device.
[0239] At step 1124, the analyzer may identify a second target
functionally connected to the novel drug target in a global
molecular entity graph. In one embodiment, the analyzer may select
a nearby drug target using a shortest path algorithm. In another
embodiment, the analyzer may select a nearby drug target with the
most interconnections to nodes also connected to the novel drug
target. For example, if the novel drug target is connected to five
additional nodes, two of which are also connected to a first target
and three of which are connected to a second target, the analyzer
may select the second target based on the additional shared node.
In some embodiments, a combination of these approaches may be used.
For example, the analyzer may select a nearby target that has the
most independent paths to the novel target of less than a
predetermined length. In some embodiments, the analyzer may even
select such a target over a second target that has fewer, but
shorter paths. For example, if a first nearby target has five paths
to the novel target, each path traversing one intermediate node
(i.e. length two), the analyzer may select this target over a
second nearby target that has only one path that directly connects
to the novel target (i.e. length one). In some embodiments, nearby
targets may be selected based on their relationship to the same
organ involved with the first target. In other embodiments, nearby
targets may be scored based on their inclusion in a common pathway
or pathways with the novel target, and the analyzer may select the
highest scoring target. In still other embodiments, nearby targets
may be scored based on their number of connections to nodes in a
shared pathway with the novel target. In a further embodiment, a
target's score may be reduced based on its number of connections to
nodes in pathways not shared with the novel target. In still other
embodiments, combinations of a plurality of these techniques may be
used to generate a score for each nearby target, and the analyzer
may select a high scoring target. In repeated iterations, the
analyzer may select additional targets scoring above a
predetermined threshold.
[0240] At step 1126, the analyzer may identify a medication
targeting the second target. In one embodiment, the analyzer may
query a medication information database for one or more medications
identified as targeting the second target. In some embodiments, the
analyzer may identify medications that are known to have off-target
effects on the second target. In some embodiments, the analyzer may
identify a plurality of medications targeting the second target and
may repeat steps 1126-1130 iteratively for each of the plurality of
medications.
[0241] At step 1128, in some embodiments, the analyzer may retrieve
from an adverse event database or generate from records retrieved
from the adverse event database a side effect profile for the
identified medication. As discussed above, the side effect profile
may comprise an identification of all side effects or adverse
events listed in the adverse event database as experienced by
patients consuming the medication, along with a score, raw number,
percentage or proportional reporting ratio, or other metric to
identify a statistical rate for each side effect. In some
embodiments, the analyzer may return the side effect profile as a
predicted side effect profile for the novel target at step 1132 for
display to the user. This may be done, for example, if the second
target is only targeted by one medication. Typically, however, the
analyzer may repeat steps 1126-1128 for additional medications
identified as targeting the second target, and/or steps 1124-1128
for additional targets nearby the novel target in the global
molecular entity graph.
[0242] At step 1130, in some embodiments, the analyzer may compare
a plurality of retrieved side effect profiles to generate an
intersection profile. In one embodiment, an intersection profile
may comprise one or more side effects or adverse events present in
each retrieved side effect profile. In another embodiment, an
intersection profile may comprise one or more side effects or
adverse events present in each retrieved side effect profile with a
similar reporting percentage or PRR, such as within a predetermined
range. This may be useful to discard false positives where a side
effect profile includes large numbers of side effects only
associated with a few records. In some embodiments, an intersection
profile may be further differentiated by outcome. For example, the
intersection profile may comprise one or more side effects or
adverse events present in each retrieved side effect profile with a
similar reporting percentage and similar rate of serious or
non-serious outcomes. This may be an important distinction, for
example, if two side effect profiles experience a side effect at
the same rate, but one has a much higher rate of serious
outcomes.
[0243] At step 1132, the intersection profile may be presented to
the user as a predicted side effect profile for the drug targeting
the novel target. In one embodiment, a display module or output
module may generate a table, list or index of the intersection
profile for display to the user. In some embodiments, the
intersection profile may be transmitted to a second computing
device for display to the user. Such predicted side effect profiles
may be used to establish safety measures for a trial protocol for
the drug. Furthermore, in some embodiments, while an intersection
profile may be more narrowly tailored to the target protein, the
analyzer may instead generate a union or combination profile at
step 1130. This may be done to ensure that all potential side
effects are included in the predicted side effect profile. In such
embodiments, the combination profile may comprise a combination of
the retrieved side effect profiles. In some embodiments, duplicate
entries in the side effect profiles, such as one side effect that
appears in each profile at a similar rate, may be removed. In other
embodiments, duplicate entries may be more highly scored, such as
with a confidence value. Thus, a side effect that appears in only
one profile may be included in the combination profile but scored
lower than a side effect that appears in a plurality of profiles at
similar rates. The latter may be more likely to occur with the new
drug. Scores or confidence values may be displayed to the user
along with profile to aid in predicting likely side effects.
[0244] In some embodiments, by integrating patient or trial
participant-specific genetic information with adverse event data, a
multivariate analysis system may be able to identify genetic
variants associated with adverse events in a clinical trial. This
may enable deeper levels of interpretation of safety signals than
are available through purely observational means, allowing in-depth
insights into the molecular protagonists and pathways involved in
eliciting drug side effects. On the one hand, a multivariate
analysis may detect drugs that induce specific clinical side
effects. Exploration of the underlying molecular mechanisms of
offending drugs allows researchers and clinicians to hone in on the
activity of targets and off-targets whose drug-induced perturbation
leads to specific adverse phenotypes. On the other hand, the
multivariate analysis may capture and contextualize relevant
published information, providing another level of gene
prioritization in association with specific side effects. Combining
these techniques and integrating other clinico-molecular
information may provide the ability to efficiently analyze patient
specific genomic information in search of genetic factors that
influence a drugs risk profile.
[0245] For example, and referring briefly to the block diagram
illustrated in FIG. 12A, in one embodiment involving a clinical
trial where a serious and unexpected adverse reaction is
encountered, a researcher may generate complete genome sequence
information for the affected patient or patients, and then attempt
to identify a causal genetic predisposition or predispositions to
the observed effect. Such sequence information may comprise
identifications of the patient's specific genetic mutations and
variants. In many embodiments, the sequence information may be
obtained from an external provider of genomic information. The
sequence may be analyzed to detect variants from wildtypes, and
each variant may be mapped to one or more corresponding molecular
entities based on their relationship to the entities, such as
whether they are activating or inactivating of a protein, etc. By
combining information and knowledge about the molecular mechanisms
associated with side effects with complete genomic sequencing,
researchers can quickly identify genetic factors that may increase
a patient's risk of drug-induced side effects. The multivariate
analyzer may determine, from adverse event data associated with
molecular entity information, which molecular entities may be
responsible for an adverse event, and correspondingly, whether the
event may be likely to occur in the general trial population or
whether it is associated with a specific variant or variants of the
affected patient.
[0246] Referring now to FIG. 12B, illustrated is a flow chart of an
embodiment of a method of identifying genetic variants associated
with adverse events. In brief overview, at step 1202, an analyzer
executed by a computing device may receive an identification of an
adverse event experienced by a patient or participant in a clinical
trial of a first medication. At step 1204, the analyzer may query
an adverse event database for records associated with the adverse
event to generate an ordered list of one or more protein targets
most associated with the event. At step 1206, the analyzer may
receive an identification of one or more genetic variants of the
participant or patient. At step 1208, the analyzer may modify the
order of the list of one or more protein targets responsive to
targets in the list corresponding to the identified one or more
genetic variants. At step 1210, the analyzer or an output module
connected to the analyzer may output the modified list to a user as
a prioritized list of variants potentially responsible for the
adverse event.
[0247] Still referring to FIG. 12B and in more detail, at step
1202, a multivariate analyzer executed by a computing device may
receive, from a user, an identification of an adverse event
experienced by a participant of a clinical trial of a first
medication. In some embodiments, the analyzer may receive the
identification of the adverse event via an input module, such as a
web interface or application interface. In many embodiments, the
analyzer may receive the identification from a second computing
device via a network.
[0248] At step 1204, the analyzer may query an adverse event
database for one or more adverse event records associated with the
adverse event. As discussed above, in some embodiments, each record
may comprise or be linked to identifications of one or more protein
targets targeted by drugs consumed by the person who experienced
the adverse event for which the record was generated. In other
embodiments, each record may comprise identifications of one or
more medications consumed by the person who experienced the adverse
event, and the analyzer may retrieve one or more corresponding
protein targets for the one or more medications from a medication
information database. The analyzer may generate an ordered list of
the proteins based on the frequency with which the protein (or a
medication targeting the protein) appears in the adverse event
records. In some embodiments, the analyzer may include a PRR or
percentage rate with which each protein appears in or is associated
with the adverse event records. In one embodiment, the analyzer may
generate a score for each protein based on the order of the protein
within the list or the identified rate.
[0249] At step 1206, the analyzer may receive an identification of
one or more genetic variants of the participant who experienced the
adverse event in the clinical trial. In some embodiments, the user
of the computing device may provide a list of variants to the
analyzer, while in other embodiments, the user of the computing
device may provide a full or partial genetic sequence of the
participant, and the analyzer may identify one or more variants
within the genetic sequence through comparison with a database of
genetic wildtypes.
[0250] At step 1208, the analyzer may modify the order of the list
of proteins for protein targets corresponding to identified genetic
variants of the participant. In some embodiments, the analyzer may
increase a score associated with each protein in the ordered list
responsive to the participant having a variant associated with the
protein, or decrease scores associated with each protein in the
ordered list responsive to the participant not having a variant or
having a wildtype associated with the protein. In a further
embodiment, the analyzer may increase a score of a protein targeted
by the first medication if the participant has a genetic variant
corresponding to the protein. In some embodiments, the analyzer may
increase the scores of proteins in the list associated with an
organ related to the adverse event, such as increasing the score of
proteins associated with the kidneys if the participant experienced
renal failure. Accordingly, the analyzer may modify the order of
the list of proteins and/or score of each protein to generate a
prioritized list of potential targets inducing the adverse event in
the participant. At step 1210, the analyzer or an output module may
present the modified list to the user as a prioritized list of
proteins potentially responsible for the experienced adverse event.
In a further embodiment, the analyzer or output module may present
the modified list with corresponding genetic variants of the
patient. Accordingly, the list may identify the genetic variants
and proteins most likely to be associated with inducing of the
adverse event.
[0251] It may be helpful to briefly discuss examples of embodiments
of an interface for performing multivariate analysis of adverse
event data. One skilled in the art may readily appreciate that many
other interfaces may be utilized, and as such, the examples should
be considered non-limiting.
[0252] Referring first to FIGS. 13A-13Y, illustrated are
screenshots of example embodiments of an interface for performing
multivariate analysis of adverse event data. In some embodiments,
the interface may be accessed through a web browser, while in other
embodiments, the interface may be provided as part of an
application. As shown in FIG. 13A, the interface may comprise a
home page or screen with one or more search boxes or links. As
shown in FIG. 13B, in response to a user entering a full or partial
search term, the interface may display a list of results,
comprising entity names matching the search, type of entity, number
of adverse events in an adverse event database associated with the
entity, most frequent drugs co-medicated with the entity, most
frequent indications for which the entity is prescribed, and most
frequent reactions associated with the entity in the adverse event
database. Similarly, as shown in FIG. 13C, searches may be done for
other entities or entity types.
[0253] Once an entity is selected from the search results, the
interface may display a dashboard of statistical data as shown in
the embodiment of FIG. 13D. Statistical data may include graphs of:
numbers of adverse events associated with the entity by year;
number of adverse events by indications; number of adverse events
by reactions; number of adverse events by outcomes; and number of
adverse events by drugs. In many embodiments, only the highest
numbered indications, reactions, or drugs may be displayed on the
dashboard, due to space limitations.
[0254] Navigation links in FIG. 13D provide access to further
detailed information. For example, as shown in FIG. 13E, the
interface may provide a list of drugs associated with the entity in
adverse event data, along with statistical data regarding their
frequency in the reports. Similarly, as shown in FIG. 13F, the
interface may provide a list of Anatomical Therapeutic Chemical
(ATC) classes, grouped by level, associated with the entity in
adverse event data, along with statistical data regarding their
frequency in the reports. In some embodiments, similar lists may be
displayed by the interface, including indications (as shown in FIG.
13G); reactions (as shown in FIG. 13H); molecular targets (as shown
in FIG. 13I); and molecular mechanisms (as shown in FIG. 13J).
[0255] In many embodiments, as shown in FIG. 13K, the interface may
provide access to individual adverse event reports for the entity.
In some embodiments, the interface may also provide identifications
of numbers of adverse events for the entity associated with
individual drugs (FIG. 13L); ATC classes (FIG. 13M); indications
(FIG. 13N); reactions (FIG. 13O); molecular targets or molecular
mechanisms (not shown for brevity). The interface may further
provide access to literature associated with the entity in a
medical literature server or accessible over a network, as shown in
FIG. 13P. In some embodiments, as shown in FIG. 13Q, the interface
may provide detailed information about the entity. Similarly, the
interface may provide information about molecular mechanisms
associated with the entity, as shown in FIG. 13R.
[0256] As discussed above in connection with FIG. 13K, the
interface may provide access to individual adverse event reports
for the entity, as shown in FIG. 13S. The adverse event reports may
comprise demographic information for the patient who experienced
the adverse event, and information regarding outcomes, consumed
medications, reactions, and indications. As discussed above, in
many embodiments, the interface may provide a radial dependency
graph, specific to the adverse event report, as shown in FIG.
13T.
[0257] In some embodiments, the interface may provide information
regarding pathways, such as a graph or portion of a global
molecular entity graph showing functional relationships among
entities associated with a pathway, as shown in FIG. 13U. As
discussed above, in many embodiments, the interface may also
provide such graphs as a result of analysis of a global molecular
entity graph.
[0258] In many embodiments, the interface may provide functions for
comparing two entities directly. For example, as shown in FIG. 13V,
the interface may provide for side-by-side searching of entities,
including different entity types, as well as side-by-side
comparison of adverse event data, as shown in FIG. 13W.
[0259] In some embodiments, as discussed above, the interface may
provide functions to generate cohorts for extraction of
cohort-specific adverse event data. Boolean queries may be crafted
defining the cohort and managed through a cohort interface, as
shown in FIG. 13X. Upon processing and extraction, adverse event
data specific to the cohort may be displayed and investigated, as
shown in FIG. 13Y. In some embodiments, the interface may comprise
a utility for building cohort definitions, as well as providing a
preview of what records may be included in the defined cohort.
[0260] Referring briefly to FIGS. 14A-C, as discussed above, in
some embodiments, a multivariate analyzer may compare side effect
profiles to generate intersection or union profiles for
investigation of combination therapies, prediction of side effects
for novel targets, or other purposes. Referring first to FIG. 14A,
illustrated is an example embodiment of a list of a side effect
profile for a first medication. The list may be sorted based on
frequency of reaction, for example, or based on frequency of a
particular outcome, such as death. Similarly, in FIG. 14B,
illustrated is an example embodiment of a list of a side effect
profile for a second medication. As shown, lists may be of
different length, for example, due to less data being available or
due to a reduced variety of side effects for one medication. As
shown in FIG. 14C, in some embodiments, side effect profiles may be
directly compared and cross referenced, allowing determinations of
differences in reactions between medications and generation of
intersection or union profiles.
[0261] In summary, by permitting the direct assessment of
relationships between the human proteome and drug-induced
phenotypes, the systems and methods discussed herein provide
efficient and intuitive approaches to the analysis and molecular
dissection of adverse event data information. Patient specific
clinico-molecular data may be integrated with the systems,
providing advanced treatment decision support.
[0262] It should be understood that the systems described above may
provide multiple ones of any or each of those components and these
components may be provided on either a standalone machine or, in
some embodiments, on multiple machines in a distributed system. The
systems and methods described above may be implemented as a method,
apparatus or article of manufacture using programming and/or
engineering techniques to produce software, firmware, hardware, or
any combination thereof. In addition, the systems and methods
described above may be provided as one or more computer-readable
programs embodied on or in one or more articles of manufacture. The
term "article of manufacture" as used herein is intended to
encompass code or logic accessible from and embedded in one or more
computer-readable devices, firmware, programmable logic, memory
devices (e.g., EEPROMs, ROMs, PROMs, RAMs, SRAMs, etc.), hardware
(e.g., integrated circuit chip, Field Programmable Gate Array
(FPGA), Application Specific Integrated Circuit (ASIC), etc.),
electronic devices, a computer readable non-volatile storage unit
(e.g., CD-ROM, floppy disk, hard disk drive, etc.). The article of
manufacture may be accessible from a file server providing access
to the computer-readable programs via a network transmission line,
wireless transmission media, signals propagating through space,
radio waves, infrared signals, etc. The article of manufacture may
be a flash memory card or a magnetic tape. The article of
manufacture includes hardware logic as well as software or
programmable code embedded in a computer readable medium that is
executed by a processor. In general, the computer-readable programs
may be implemented in any programming language, such as LISP, PERL,
C, C++, C#, PROLOG, or in any byte code language such as JAVA. The
software programs may be stored on or in one or more articles of
manufacture as object code.
[0263] Having described certain embodiments of methods and systems
for providing systems and methods for molecular analysis of adverse
event data, it will now become apparent to one of skill in the art
that other embodiments incorporating the concepts of the invention
may be used.
* * * * *