Poster Presentation Society for Molecular Biology and Evolution Conference 2016

Measuring Natural Selection on Chromatin Accessibility in 1000 Humans From 10 Populations. (#553)

Ashley K. Tehranchi 1 , Hunter B. Fraser 1
  1. Department of Biology, Stanford University, Stanford, CA

Cis-regulatory elements such as transcription factor (TF) binding sites or open chromatin regions can be identified genome-wide, but it remains far more challenging to pinpoint genetic variants affecting these elements. We recently developed a pooling-based approach to mapping quantitative trait loci (QTLs) for molecular-level traits (Tehranchi et al., Cell, 2016). We applied this to chromatin immuno-precipitation followed by high-throughput sequencing (ChIP-seq) in five TFs and a histone modification and mapped thousands of cis-acting QTLs, with over 25-fold lower cost compared to standard QTL mapping. Thousands of these QTLs have been implicated in genome-wide association studies, providing candidate molecular mechanisms for many disease risk loci, and suggesting that TF binding variation may underlie a large fraction of human phenotypic variation. We are now applying our pooling method to ATAC-seq (a method to assay accessible chromatin) in cells from 1,000 human individuals, from 10 diverse populations, to identify chromatin-accessible QTLs (caQTLs). Our goal is to better understand how natural selection has shaped the landscape of cis-regulation in humans, and how regulatory differences may give rise to population-specific traits. Although many studies have used the genomes of diverse human populations to investigate patterns of natural selection, this approach has significant limitations—for instance, the statistical burden of testing the entire genome has severely limited the power of most selection scans. With our data, we will be able to focus the search on likely causal variants affecting chromatin accessibility. In addition, we will explore an entirely new type of population-genetic analysis not possible with DNA sequence data, comparing chromatin accessibility within populations to variation between populations. Loci with large differences between populations but little variation within populations will be top candidates for targets of positive selection.