RT Journal Article SR Electronic T1 The impact of rare variation on gene expression across tissues JF bioRxiv FD Cold Spring Harbor Laboratory SP 074443 DO 10.1101/074443 A1 Xin Li A1 Yungil Kim A1 Emily K. Tsang A1 Joe R. Davis A1 Farhan N. Damani A1 Colby Chiang A1 Zachary Zappala A1 Benjamin J. Strober A1 Alexandra J. Scott A1 Andrea Ganna A1 Jason Merker A1 GTEx Consortium A1 Ira M. Hall A1 Alexis Battle A1 Stephen B. Montgomery YR 2016 UL http://biorxiv.org/content/early/2016/09/09/074443.abstract AB Rare genetic variants are abundant in humans yet their functional effects are often unknown and challenging to predict. The Genotype-Tissue Expression (GTEx) project provides a unique opportunity to identify the functional impact of rare variants through combined analyses of whole genomes and multi-tissue RNA-sequencing data. Here, we identify gene expression outliers, or individuals with extreme expression levels, across 44 human tissues, and characterize the contribution of rare variation to these large changes in expression. We find 58% of underexpression and 28% of overexpression outliers have underlying rare variants compared with 9% of non-outliers. Large expression effects are enriched for proximal loss-of-function, splicing, and structural variants, particularly variants near the TSS and at evolutionarily conserved sites. Known disease genes have expression outliers, underscoring that rare variants can contribute to genetic disease risk. To prioritize functional rare regulatory variants, we develop RIVER, a Bayesian approach that integrates RNA and whole genome sequencing data from the same individual. RIVER predicts functional variants significantly better than models using genomic annotations alone, and is an extensible tool for personal genome interpretation. Overall, we demonstrate that rare variants contribute to large gene expression changes across tissues with potential health consequences, and provide an integrative method for interpreting rare variants in individual genomes.