Combinatorial Detection of Conserved Alteration Patterns for Identifying Cancer Subnetworks

摘要

Background Advances in large-scale tumor sequencing have led to an understanding that there are combinations of genomic and transcriptomic alterations specific to tumor types, shared across many patients. Unfortunately, computational identification of functionally meaningful and recurrent alteration patterns within gene/protein interaction networks has proven to be challenging. Findings We introduce a novel combinatorial method, cd-CAP ( c ombinatorial d etection of c onserved a lteration p atterns), for simultaneous detection of connected subnetworks of an interaction network where genes exhibit conserved alteration patterns across tumor samples. Our method differentiates distinct alteration types associated with each gene (rather than relying on binary information of a gene being altered or not) and simultaneously detects multiple alteration profile conserved subnetworks. Conclusions In a number of The Cancer Genome Atlas datasets, cd-CAP identified large biologically significant subnetworks with conserved alteration patterns, shared across many tumor samples.