Background: Whereas the clinical world is familiar with genomic assays targeted to a limited number of mutations as a means to derive molecular insight to therapies, the power to deliver more comprehensive, non-assumptive, and stochastic molecular analysis is now available to guide treatment decision that is unbiased to traditional tissue-by-tissue assignment of therapeutics or a priori assumptions that a few hundred DNA mutations are drivers of the cancer. Methods: Over 10,000 tumor and germ line exomes and RNA-seq data representing over 5,000 patients across more than 20 tumor types were processed using Contrastor, a cancer genome differentiation algorithm designed to detect variants (germline and somatic), copy-number alterations (overall and allele-specific), and genomic rearrangements. RNA-seq data was used to confirm the presence of alterations in the transcript when available. Paradigm, an algorithm that integrates genomic and transcriptomics data and designed to identify key altered cell signaling pathways, was used to reveal shared pathways amongst independent tumor types. Results: Aggregating the top altered genes across tissues reveals very few genes are dominated with alterations present in single diseases. DNA repair and kinase signaling proteins appear to be the predominantly altered classes, with transcriptional proteins also highly altered (TP53, MLL, CREBBP). A surprising number of new targets for pharmaceutical discovery were uncovered, and particularly in patients with no clear therapeutic options based on the analysis of “actionable” genes. Despite patients having unique molecular profiles, Paradigm revealed shared pathways across tissue types driving tumor progression that are likely targets for therapeutic intervention. Conclusions: The vast landscape and ample number of mutations identified in this comprehensive analysis illustrates the complexity of cancer genomics and the need for transcriptomics driven pathway analysis to identify the focal points of therapeutic intervention.