@article {Lamaze126359, author = {Fabien C. Lamaze and Aurelien Chateigner and Hilary A. Edgington and Marie-Julie Fave and Armande Ang Houle and PCAWG3 and Philip Awadalla}, title = {Motif Disruption Domains Lead To Cancer Gene Expression Rewiring}, elocation-id = {126359}, year = {2017}, doi = {10.1101/126359}, publisher = {Cold Spring Harbor Laboratory}, abstract = {Somatic mutations accumulate in non-coding regions of the genome during tumorigenesis, but their functional characterization presents a challenge. Somatic non-coding mutations rarely overlap among patients, which necessitates large sample sizes to detect associations. We analysed somatic mutations called from whole-genome sequencing (WGS) and RNA sequencing (RNAseq) from 3000 tumors across the Pan-Cancer Analysis of Whole Genomes to identify and functionally characterize mutation accumulation and its impact on gene dysregulation in cancer. We identified 1.5 million motif disruption domains (MDDs) across 40 cancer types, which we characterized as pan-cancer targets for recurrent mutation accumulation. These MDDs deregulate gene expression in cancer-specific and pan-cancer patterns by disrupting transcription factor binding sites in regulatory and insulator elements. Disruption is most recurrent across individuals at MDDs in conserved open chromatin, revealing potential drivers. This accumulation of somatic variants targeting regulatory and structural elements in MDDs generates gene expression dysregulation during tumorigenesis.MDDMotif Disruption DomainPCAWGthe Pan-Cancer Analysis of Whole Genomes projectSNVSomatic single Nucleotide VariantsTFBSTranscription Factor Binding Site}, URL = {https://www.biorxiv.org/content/early/2017/04/11/126359}, eprint = {https://www.biorxiv.org/content/early/2017/04/11/126359.full.pdf}, journal = {bioRxiv} }