Drug discovery methods
First Claim
Patent Images
1. A method for evaluating user-supplied genomics data, comprising:
- receiving a set of genes, selected from user-supplied genomics data, that are used to generate a library of profiles;
computing a triangles score for each gene in the selected set by computing a count of three-neighbor loops including the gene, the loops defined by edge connections stored in a structured database, wherein the structured database is structured according to predetermined, causal relationships among genes and/or gene products;
generating a plurality of seed sets based on the computed triangles scores, each seed set having a unique subset of the set of genes, wherein each profile, from the profile library, is a seed set, from the plurality of seed sets, that is converted into a subnetwork having additional genes selected from the structured database according to one or more criteria;
identifying one or more profiles, from the profile library, including respective subsets of data that overlap at least a portion of the user-supplied genomics data;
determining, for each such overlapped profile, whether the overlap with the user-supplied genomics data is statistically significant; and
presenting a list of the one or more overlapped profiles determined to have a statistically significant overlap,wherein each of the receiving, computing, generating, identifying, determining, and presenting is performed by a processing system.
2 Assignments
0 Petitions
Accused Products
Abstract
Methods for identifying disease-related pathways that can be used to identify drug discovery targets, to identify new uses for known drugs, to identify markers for drug response, and related purposes.
69 Citations
19 Claims
-
1. A method for evaluating user-supplied genomics data, comprising:
-
receiving a set of genes, selected from user-supplied genomics data, that are used to generate a library of profiles; computing a triangles score for each gene in the selected set by computing a count of three-neighbor loops including the gene, the loops defined by edge connections stored in a structured database, wherein the structured database is structured according to predetermined, causal relationships among genes and/or gene products; generating a plurality of seed sets based on the computed triangles scores, each seed set having a unique subset of the set of genes, wherein each profile, from the profile library, is a seed set, from the plurality of seed sets, that is converted into a subnetwork having additional genes selected from the structured database according to one or more criteria; identifying one or more profiles, from the profile library, including respective subsets of data that overlap at least a portion of the user-supplied genomics data; determining, for each such overlapped profile, whether the overlap with the user-supplied genomics data is statistically significant; and presenting a list of the one or more overlapped profiles determined to have a statistically significant overlap, wherein each of the receiving, computing, generating, identifying, determining, and presenting is performed by a processing system. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
-
Specification