RT Journal Article
SR Electronic
T1 Identification of putative effector genes across the GWAS Catalog using molecular quantitative trait loci from 68 tissues and cell types
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 808444
DO 10.1101/808444
A1 Cong Guo
A1 Karsten B. Sieber
A1 Jorge Esparza-Gordillo
A1 Mark R. Hurle
A1 Kijoung Song
A1 Astrid J. Yeo
A1 Laura M. Yerges-Armstrong
A1 Toby Johnson
A1 Matthew R. Nelson
YR 2019
UL http://biorxiv.org/content/early/2019/10/17/808444.abstract
AB Identifying the effector genes from genome-wide association studies (GWAS) is a crucial step towards understanding the biological mechanisms underlying complex traits and diseases. Colocalization of expression and protein quantitative trait loci (eQTL and pQTL, hereafter collectively called “xQTL”) can be effective for mapping associations to genes in many loci. However, existing colocalization methods require full single-variant summary statistics which are often not readily available for many published GWAS or xQTL studies. Here, we present PICCOLO, a method that uses minimum SNP p-values within a locus to determine if pairs of genetic associations are colocalized. This method greatly expands the number of GWAS and xQTL datasets that can be tested for colocalization. We applied PICCOLO to 10,759 genome-wide significant associations across the NHGRI-EBI GWAS Catalog with xQTLs from 28 studies. We identified at least one colocalized gene-xQTL in at least one tissue for 30% of associations, and we pursued multiple lines of evidence to demonstrate that these mappings are biologically meaningful. PICCOLO genes are significantly enriched for biologically relevant tissues, and 4.3-fold enriched for targets of approved drugs.