PT - JOURNAL ARTICLE AU - James T. Webber AU - Max V. Ranall AU - Swati Kaushik AU - Sourav Bandyopadhyay TI - Connecting tumor genomics with therapeutics through multi-dimensional network modules AID - 10.1101/083410 DP - 2016 Jan 01 TA - bioRxiv PG - 083410 4099 - http://biorxiv.org/content/early/2016/10/25/083410.short 4100 - http://biorxiv.org/content/early/2016/10/25/083410.full AB - Recent efforts have catalogued genomic, transcriptomic, epigenetic and proteomic changes in tumors, but connecting these data with effective therapeutics remains a challenge. In contrast, cancer cell lines can model therapeutic responses but only partially reflect tumor biology. Bridging this gap requires new methods of data integration to identify a common set of pathways and molecular events. Using MAGNETIC, a new method to integrate molecular profiling data using functional networks, we identify 219 gene modules in TCGA breast cancers that capture recurrent alterations, reveal new roles for H3K27 tri-methylation and accurately quantitate various cell types within the tumor microenvironment. We show that a significant portion of gene expression and methylation in tumors is poorly reproduced in cell lines due to differences in biology and microenvironment and MAGNETIC identifies therapeutic biomarkers that are robust to these differences. This work addresses a fundamental challenge in pharmacogenomics that can only be overcome by the joint analysis of patient and cell line data.