Network-Based Integration of Disparate Omic Data To Identify "Silent Players" in Cancer

• DOI: 10.1371/journal.pcbi.1004595
• Publication Date: 2015-12-18T18:42:35Z
• Authors: Matthew Ruffalo; Mehmet Koyutürk; Roded Sharan
• Keywords: Genetic Markers; QH301-705.5; Gene Expression Profiling; 0206 medical engineering; Chromosome Mapping; Genomics; 02 engineering and technology; Neoplasm Proteins; 3. Good health; Neoplasms; Databases, Genetic; Data Mining; Humans; Biology (General); Algorithms; Genetic Association Studies; Silent Mutation; Research Article; Genes, Neoplasm; Signal Transduction

Development of high-throughput monitoring technologies enables interrogation cancer samples at various levels cellular activity. Capitalizing on these developments, public efforts such as The Cancer Genome Atlas (TCGA) generate disparate omic data for large patient cohorts. As demonstrated by recent studies, heterogeneous sources provide the opportunity to gain insights into molecular changes that drive pathogenesis and progression. However, are limited vast search space a result low statistical power make new discoveries. In this paper, we propose methods integrating using interaction networks, with view gaining mechanistic relationship between different Namely, hypothesize genes play role in development progression may be implicated neither frequent mutation nor differential expression, network-based integration expression can reveal "silent players". For purpose, utilize network-propagation algorithms simulate information flow cell sample-specific resolution. We then use propagated signals identify not necessarily mutated or differentially expressed genes, but have an essential tumor outcome. test proposed method breast glioblastoma multiforme obtained from TCGA. Our results show important proteins readily revealed data, providing beyond what gleaned analyzing types isolation.