Allele Specific Expression Analysis Using Transcriptome Sequencing in Three Tissues of a Twin Cohort Reveals Large Effect of Gene-by-Gene and Gene-by-Environment Interactions. A. Buil1,2,3, A. A. Brown4, A. Vi˝uela5, M. N. Davies5, H. F. Zheng6, J. B. Richards5,6, K. S. Small5, R. Durbin4, T. D. Spector5, E. T. Dermitzakis1,2,3 1) Genetics and Development, University of Geneva, Geneva, Switzerland; 2) Institute of Genetics and Genomics in Geneva, University of Geneva, Geneva, Switzerland; 3) Swiss Institute of Bioinformatics, Switzerland; 4) Wellcome Trust Sanger Institute, Hinxton, United Kingdom; 5) Department of Twin Research and Genetic Epidemiology, King's College London, London, United Kingdom; 6) Departments of Medicine, Human Genetics, Epidemiology and Biostatistics McGill University, Canada.
While allele specific expression is expected to result from genetic regulatory variants, a proper estimation and dissection of the causes has not been performed to date. In this study we used RNA-seq data from fat, LCLs and skin from ~400 female MZ and DZ twin pairs (2330 RNA-seq samples in total) to quantify allelic specific expression and to dissect its underlying causes . First, we performed eQTL analysis and we found 9861 genes with at least a significant cis eQTLs for fat, 10015 for LCL and 9243 for skin (FDR=1%). Then, we estimated ASE at every heterozygous site for every individual. At a 10% FDR, we observed a significant ASE effect in 9.5% of the transcript heterozygous sites in fat, 9.3% in LCL and 9.1% in skin. ASE may be caused by genetic or epigenetic /environmental factors. To measure the relative contribution of the underlying causes of allelic expression we estimated the variance components of the ASE ratios using the identity-by-descended status (IBD) of the twin pairs at the ASE site and the identity-by-state status (IBS) at the best eQTL. We found that about 35% of the variance in ASE is due to the effect of the best eQTL , 14% to the additive effect of the other genetic variants in cis, 33% to the interaction between cis and trans variants and 18% to the individual environment. The additive trans and the shared environmental effects were negligible. There were small differences among tissues. The sum of all the genetic effects gives an average heritability estimate of 73% for fat, 86% for LCL and 83% for skin. Our results show a complex genetic architecture for allelic expression that identifies GxG and putative GxE effects. We took advantage of the twin structure of our sample to look for examples of GxE interactions. Since MZ twins are genetically identical, differences in allelic expression in a MZ pair have to be caused by non genetic effects. For every site, we calculated the association between allelic expression differences within MZ pairs and SNPs around the site and found examples of potential GxE interactions. One example in fat tissue was found for ADIPOQ, a gene that codifies for adiponectin, whose expression has been observed to be affected by environmental factors such as diet and physical exercise. We are exploring further to find putative GxG interactions affecting allelic expression.