Systematic assessment of regulatory effects of human disease variants in pluripotent cells

Abstract

Identifying regulatory genetic effects in pluripotent cells provides important insights into disease variants with potentially transient or developmental origins. Combining existing and newlygenerated data, we established a population-scale resource of 1,367 induced pluripotent stem cell lines derived from 948 unique donors, with matched RNA-sequencing (RNA-seq) and genetic information. The sample size of our study allowed us to significantly expand our knowledge of quantitative trait loci (QTL) in pluripotent human cells and their broad trait and disease relevance.We identified cis-QTL for conventional gene expression levels but also transcript-ratio, exon expression levels, alternative-splicing and alternative polyadenylation usage, cumulatively yielding cis regulatory effects for 18,556 genes. We assessed the effects of rare variants identified from whole-genome sequencing data, by relating these to gene expression outliers, observing an enrichment for rare, deleterious variants in iPSCs with outlying expression profiles, which exceeds previous observations in differentiated tissues. Finally, we assessed distal QTL on a genomewide scale, identifying 193 trans-eQTL, linked to 191 trans-eGenes (FDR <10%), 38% of which were replicated in independent samples.By linking our regulatory map of rare and common QTL to comprehensive GWAS data, we identified high-confidence colocalization events for 4,336 individual GWAS loci, which includes physical traits such as height and coronary artery disease. In addition, we identify rare variant associations for metabolic rate and trans-eQTL linked to both cancer and height. Collectively, our data greatly expand the regulatory landscape in human pluripotent cells, and catalogues traitassociated variants that have potential developmental or transient contexts.

Population-scale single-cell RNA-seq profiling across dopaminergic neuron differentiation

“…For cell lines that were differentiated multiple times, their differentiation efficiency was taken as the mean of its differentiation efficiencies in the replicate experiments. Gene expression data was available from independent bulk RNA-seq experiments 34 . The feature set was all expressed genes (i.e.…”

“…To quantify the amount of sharing between each two pairs of eQTL maps (our cell type-condition maps to each of 13 eQTL maps of brain tissues from GTEx) we used the MASHR software 42 . Briefly, the effect sizes (betas) were calculated for each SNP-gene pair across the 14 cell type-condition eQTL maps from our study and extracted from the 13 brain tissues from the GTEx catalogue, as well as two iPSC eQTL maps, using scRNA-seq and bulk RNA-seq respectively 2,34 . Only genes expressed in all GTEx tissues, in iPSC 34 and all of our contexts were included (n=6,205).…”

“…Briefly, the effect sizes (betas) were calculated for each SNP-gene pair across the 14 cell type-condition eQTL maps from our study and extracted from the 13 brain tissues from the GTEx catalogue, as well as two iPSC eQTL maps, using scRNA-seq and bulk RNA-seq respectively 2,34 . Only genes expressed in all GTEx tissues, in iPSC 34 and all of our contexts were included (n=6,205). To calculate the sharing of effects we selected the top eQTL SNP per gene, based on the effect sizes in the tissues with more than 150 samples.…”

“…A gene expression signature in iPSCs is associated with neuronal differentiation efficiency. (a) Venn diagram indicating the overlap of cell lines included in this study and two recent iPSC studies, a bulk RNA-seq study34 and a single cell RNA-seq study 2 . (b) Histogram of Pearson correlation coefficients between variation in gene expression of individual genes (from bulk RNA-seq 34 ) and differentiation efficiency.…”

In-depth genetic analysis of 6p21.3 reveals insights into associations between HLA types and complex traits and disease

“…To address this gap, we developed a comprehensive map of regulatory variation in the MHC region using deep WGS from 419 individuals and RNA-seq data from matched iPSCs. We have previously shown that iPSCs are well-powered for eQTL mapping (Bonder et al, 2019;Jakubosky et al, 2019a;Jakubosky et al, 2019b), have a distinct regulatory landscape relative to somatic tissues , and thereby provide insights into regulatory variants that exert their effects during early development. While we demonstrate the feasibility of generating a map of regulatory variation for the MHC region and its utility for examining molecular mechanisms underlying disease associations in the interval, future studies using eQTLs from other cell types could increase the number of complex trait loci in the interval that are functionally annotated.…”