Approximation algorithms for NP-hard problems
Approximation algorithms for NP-hard problems
Diffusion Kernels on Graphs and Other Discrete Input Spaces
ICML '02 Proceedings of the Nineteenth International Conference on Machine Learning
An efficient rigorous approach for identifying statistically significant frequent itemsets
Proceedings of the twenty-eighth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Evaluating Between-Pathway Models with Expression Data
RECOMB 2'09 Proceedings of the 13th Annual International Conference on Research in Computational Molecular Biology
Detecting disease-specific dysregulated pathways via analysis of clinical expression profiles
RECOMB'08 Proceedings of the 12th annual international conference on Research in computational molecular biology
Connected set cover problem and its applications
AAIM'06 Proceedings of the Second international conference on Algorithmic Aspects in Information and Management
Proceedings of the ACM Conference on Bioinformatics, Computational Biology and Biomedicine
| Hi-index | 0.00 |
Recent genome sequencing studies have shown that the somatic mutations that drive cancer development are distributed across a large number of genes This mutational heterogeneity complicates efforts to distinguish functional mutations from sporadic, passenger mutations Since cancer mutations are hypothesized to target a relatively small number of cellular signaling and regulatory pathways, a common approach is to assess whether known pathways are enriched for mutated genes However, restricting attention to known pathways will not reveal novel cancer genes or pathways An alterative strategy is to examine mutated genes in the context of genome-scale interaction networks that include both well characterized pathways and additional gene interactions measured through various approaches We introduce a computational framework for de novo identification of subnetworks in a large gene interaction network that are mutated in a significant number of patients This framework includes two major features First, we introduce a diffusion process on the interaction network to define a local neighborhood of “influence” for each mutated gene in the network Second, we derive a two-stage multiple hypothesis test to bound the false discovery rate (FDR) associated with the identified subnetworks We test these algorithms on a large human protein-protein interaction network using mutation data from two recent studies: glioblastoma samples from The Cancer Genome Atlas and lung adenocarcinoma samples from the Tumor Sequencing Project We successfully recover pathways that are known to be important in these cancers, such as the p53 pathway We also identify additional pathways, such as the Notch signaling pathway, that have been implicated in other cancers but not previously reported as mutated in these samples Our approach is the first, to our knowledge, to demonstrate a computationally efficient strategy for de novo identification of statistically significant mutated subnetworks We anticipate that our approach will find increasing use as cancer genome studies increase in size and scope.