Exploring regulatory and loss-of-function variation in personalized multi-tissue transcriptomes using allele-specific expression. T. Lappalainen1,2,3,4, M. A. Rivas5, M. Lek6,7, M. Pirinen5,8, J. Maller6,7, K. Kukurba1, E. Tsang9, D. DeLuca7, M. Sammeth10,11,12, . Geuvadis Consortium2,11, M. I. McCarthy5, C. D. Bustamante1, S. B. Montgomery1,13, K. Ardlie7, D. G. MacArthur6,7, E. T. Dermitzakis2,3,4, GTEx Consortium 1) Department of Genetics, Stanford University, Stanford, CA; 2) Department of Genetic Medicine and Development, University of Geneva Medical School, Switzerland; 3) Institute for Genetics and Genomics in Geneva (1G3), University of Geneva, Switzerland; 4) Swiss Institute of Bioinformatics, Geneva, Switzerland; 5) Wellcome Trust Centre for Human Genetics, University of Oxford, UK; 6) Analytic and Translational Genetics Unit, Massachusetts General Hospital, Boston, MA; 7) Broad Institute of Harvard and MIT, Cambridge, MA; 8) Institute for Molecular Medicine Finland, University of Helsinki, Finland; 9) Program in Biomedical Informatics, Stanford University School of Medicine, Stanford, CA; 10) Centro Nacional d'AnÓlisi Gen˛mica, Barcelona, Spain; 11) Center for Genomic Regulation CRG, Barcelona, Spain; 12) National Center for Scientific Computation, Petropolis, Rio de Janeiro, Brazil; 13) Department of Pathology, Stanford University, Stanford, CA.

   Analysis of genetic regulatory variation is advancing towards comprehensive analysis of common and rare regulatory and loss-of-function variants across a wide variety of tissues. A powerful approach to characterize these effects both in populations and at the level of an individual is analysis of allelic imbalance between the two haplotypes of an individual. In this study, we analyzed allele-specific expression (ASE) and transcript structure (ASTS) in 1582 samples of the GTEx pilot data from 171 individuals and 45 tissues with RNA-seq, genotype and partial exome-seq data. First, we measured the activity of eQTLs discovered from 9 major tissues across all the 45 tissues, and quantified how well these eQTLs capture systemic regulatory effects and assigned the best eQTL proxy for diverse tissues. Quantifying common and rare regulatory variation genome-wide with ASE analysis shows that 9% of effects in one tissue are captured by analyzing another tissue. However, the low sharing is mostly due to tissue-specificity of gene expression, as allelic imbalance alone is highly shared (median 43%) between different tissues within an individual, and ASTS analysis captures a similar pattern for splicing variation. We quantified transcriptome effects of a total of 11,518 putative loss-of function variants - stop-gained SNPs, splice variants, and frameshift indels - from both GTEx and the Geuvadis project (RNA-seq data of 462 individuals from 1000 Genomes). We predict the likelihood for a stop-gained variant to trigger nonsense-mediated decay based on its properties, which will be crucial for interpreting loss-of-function variants discovered in future studies. Additionally, as much as 22% of variants show variable NMD between tissues. Taking population-level assessment of functional variants to the level of an individual is a challenge in personalized genetics applications. With ASE analysis we uncover substantial variation in eQTL effects between individuals, suggesting that modifiers of common variants are widespread even at the cellular level. Furthermore, transcriptome sequencing of carriers of rare putative loss-of-function variants allows interpretation of their functional effects across different tissues. Altogether, this study demonstrates the power of transcriptome sequencing to understand systemic effects of functional variation and to interpret personalized genomes.

You may contact the first author (during and after the meeting) at