A systematic eQTL study of cis–trans epistasis in 210 HapMap individuals

Abstract

We aimed at identifying transcripts whose expression is regulated by a SNP-SNP interaction. Out of 47 294 expression phenotypes we used 3107 transcripts that survived an extensive quality control and 86 613 linkage disequilibrium-pruned SNP markers that have been genotyped in 210 individuals. For each transcript we defined cis-SNPs, tested them for epistasis with all trans-SNPs, and corrected all observed cis-trans-regulated expression effects for multiple testing. We determined that the expression of about 15% of all included transcripts is regulated by a significant two-locus interaction, which is more than expected (P¼2.86Â10 À144 ). Our findings suggest further that cis-markers with so called 'marginal effects' are more likely to be involved in two-locus gene regulation than expected (P¼8.27Â10 À05 ), although the majority of interacting cis-markers showed no one-locus regulation. Furthermore, we found evidence that gene-mediated trans-effects are not a major source of epistasis, as no enrichment of genes has been found in close vicinity of trans-SNPs. In addition, our data support the notion that neither chromosomal regions nor cellular processes are enriched in epistatic interactions. Finally, some of the cis-trans regulated genes have been found in genome-wide association studies, which might be interesting for follow-up studies of the corresponding disorders. In summary, our results provide novel insights into the complex genome-transcriptome regulation.

Using next-generation RNA sequencing to identify imprinted genes

“…To identify novel candidate imprinted genes in humans, the first step is to check informative individuals for transcribed genes and exclude the possibility of RMAE. The next step is to genotype the informative SNP positions in the parents and distinguish the imprinting candidates from a large number of cis-eQTL (Montgomery et al, 2010;Majewski and Pastinen, 2011;Becker et al, 2012). One useful method is to search for a 'flipped' allelic expression pattern at SNP positions (Pollard et al, 2008).…”

Genomics and the Evolution of Phenotypic Traits

“…Functional assays can reveal that different mutations affecting the function of the same gene can have distinct functional consequences, for instance raising and lowering the level of expression (e.g., Babbitt et al 2010, Bickel et al 2011. The same mutation can have different effects on molecular function depending on genetic background or environmental condition (e.g., Becker et al 2012, Cubillos et al 2012). Mutations that extend the level, timing, or tissue distribution of expression are likely to be codominant (Stern 2010).…”

“…In some cases, the extent of trait variation that cannot be attributed to individual QTLs is extensive (e.g., Gaertner et al 2012, Zwarts et al 2011. Association methods can also be applied to thousands of eQTLs at once and have revealed evidence of extensive epistasis in molecular function (e.g., Becker et al 2012). A nongenetic test across deeper evolutionary time suggests that epistasis due to protein:protein interactions is extensive; 90% of amino acid substitutions are neutral or beneficial only in specific genetic backgrounds (Breen et al 2012).…”

Properties of Local Interactions and Their Potential Value in Complementing Genome-Wide Association Studies

“…The development of new methods and tools has greatly reduced the computational barrier and made the routine analysis of epistasis in GWAS data achievable [4], [5], [6], [7], [8]. Furthermore, progress has been made in dissecting the molecular mechanisms underlying epistasis [9], [10]. With these advances it is hoped that future studies will accumulate more evidence of epistasis and improve our understanding of the role of epistasis in the genetic regulation of complex traits [11].…”