Phenome-wide functional dissection of pleiotropic effects highlights key molecular pathways for human complex traits

Abstract

Over the recent decades, genome-wide association studies (GWAS) have dramatically changed the understanding of human genetics. A recent genetic data release by UK Biobank (UKB) has allowed many researchers worldwide to have comprehensive look into the genetic architecture of thousands of human phenotypes. In this study, we used GWAS summary statistics derived from the UKB cohort to investigate functional mechanisms of pleiotropic effects across the human phenome. We find that highly pleiotropic variants often correspond to broadly expressed genes with ubiquitous functions, such as matrisome components and cell growth regulators; and tend to colocalize with tissue-shared eQTLs. At the same time, signaling pathway components are more prevalent among highly pleiotropic genes compared to regulatory proteins such as transcription factors. our results suggest that proteinlevel pleiotropy mediated by ubiquitously expressed genes is the most prevalent mechanism of pleiotropic genetic effects across the human phenome.

LAVA: An integrated framework for local genetic correlation analysis

“…Results from just over a decade of genome-wide association studies (GWAS) have demonstrated that statistical pleiotropy across the genome is ubiquitous, meaning that particular genetic variants, genes, or genomic regions often show association with more than one trait 1–3 . Pleiotropy is valuable to study for a number of reasons, as it could elucidate biological pathways that are shared between traits 4–6 , help generate hypotheses about the functional significance of GWAS results 7–9 , and improve our understanding of the aetiology and overlap between complex traits and diseases 1,10,11 .…”

Highly pleiotropic variants of human traits are enriched in genomic regions with strong background selection

“…Here we carried out an analysis of variants recovered from the GWAS Catalog for 41 human traits and diseases to investigate the mean frequency, effect size, recombination rate and intensity of background selection associated with variants with different degrees of pleiotropy. In addition, we investigated the intensity of background selection associated with the data sets of pleiotropic variants analyzed by Pickrell et al (2016), Watanabe et al (2019) and Shikov et al (2020). Overall, the results suggest that more pleiotropic variants are located in regions with stronger background selection.…”

“…Wang et al 2010;Sivakumaran et al 2011;Pickrell et al 2016;Chesmore et al 2018;Jordan et al 2019;Watanabe et al 2019;Shikov et al 2020) and it has been suggested that complex traits are driven by an enormously large numbers of genes, implying that pleiotropy is the rule rather than the exception (Boyle et al 2017). The latest meta-analysis on pleiotropic variants carried out by Shikov et al (2020), and based on more than five hundred complex traits, concludes that about 180 Mbs of the human genome are covered by pleiotropic loci and about 50% of SNPs are associated with more than one phenotype. Another recent study (Watanabe et al 2019) suggests that this proportion is even larger (60 %).…”

“…An observation made by Shikov et al (2020) is that rare variants tend to be less pleiotropic than common ones. This result is coherent with the observation that natural selection against deleterious mutations has been shown to operate on complex trait variation (Gazal et al 2018;Zeng et al 2018).…”

“…Thus, if pleiotropic variants affecting human diseases have deleterious effects on them, more pleiotropic variants would be expected to be removed from the population (Paaby and Rockman 2013). Shikov et al (2020) showed that more pleiotropic variants have higher gene expression than less pleiotropic ones, but they did not compare the mean effect sizes of variants across different degrees of pleiotropy. A previous analysis of pleiotropy of human genes showed, however, a tendency for more pleiotropic variants to have larger effect sizes than less pleiotropic ones (Chesmore et al 2018).…”