TY - JOUR T1 - Dissecting the genomic heterogeneity of cancer hallmarks’ acquisition with SLAPenrich JF - bioRxiv DO - 10.1101/077701 SP - 077701 AU - Francesco Iorio AU - Luz Garcia-Alonso AU - Jonathan Brammeld AU - Iñigo Martincorena AU - David R Wille AU - Ultan McDermott AU - Julio Saez-Rodriguez Y1 - 2017/01/01 UR - http://biorxiv.org/content/early/2017/03/27/077701.abstract N2 - We present a computational method, implemented in an open source R package, to explore how cancers from different tissue types might have acquired the same cancer hallmark (evolutionary successful trait) via preferential genomically altering different biological pathways. To this aim, we have curated a collection of 374 orthogonal pathway gene-sets encompassing thousands of genes (from public available resources) mapped to 10 canonical cancer hallmarks.Using this curated data resource, we have characterised the landscape of pathway alterations putatively contributing to the acquisition of different cancer hallmarks via systematic analysis of somatic mutations in large cohorts of patients across 10 cancer types, from the Cancer Genome Atlas.We assume that the heterogeneity of each hallmark in terms of number of corresponding enriched pathway alterations is reflective of its evolutionary fitness to the cancer type under consideration. A systematic evaluation of this heterogeneity across hallmarks and cancer types has resulted into a set of cancer hallmark heterogeneity signatures. These signatures quantitatively confirm the established predominance of certain hallmarks in determined cancer types and their clinical relevance, and they allow an easy data-driven comparison of cancer hallmark heterogeneity across different lineages.We have found, as expected, that most of the pathway alteration enrichments and large hallmark heterogeneities are guided by somatic mutations in established, and highly frequently mutated, high-confidence cancer driver genes. However and most importantly, when excluding these variants from the analyses, we observe that the hallmark heterogeneity signatures, thus the level of predominance of the considered hallmarks, are strikingly preserved across cancer types. Therefore we propose to use the hallmark heterogeneity signatures as a ground truth to characterise long tails of infrequent genomic alterations, across cancer types, and we highlight a number of potential novel cancer driver genes and networks. ER -