Extended Data Fig. 1: SNPs in the RIPK1 locus associate with obesity in humans.

a, eQTL analysis of SNPs rs2272990 and rs2064310 in the RIPK1 locus and levels of RIPK1 mRNA in adipose tissue from the METSIM cohort. b, Regional association plots of eQTLs for RIPK1 in subcutaneous adipose from GTEX (probe ENSG00000137275.9). SNPs are coloured based on their linkage disequilibrium (r²) with the labeled top eQTL (rs2064310), which has the smallest P value in the region. Linkage disequilibrium and recombination rate calculations are based on the European population from 1000 Genomes reference panel (Phase I; release 3). c, d, Individual SNPs from the GTEx database and their association with RIPK1 expression in visceral (c) and subcutaneous (d) adipose tissue. Number of subjects homozygous or heterozygous for the minor (Ref) or major (Alt) alleles are indicated. p-values as calculated by the GTEx Consortium detailed at: https://www.gtexportal.org. Box plot shows ranked normalized gene expression in median, 1st and 3rd quartiles, 1.5 interquartile range (IQR) of 1st and 3rd quartiles. e, Odds ratio for the index SNP rs6907943 and LD SNPs in the Ottawa OBLE cohort and comparison to rs9939609 and LD SNPs in the FTO locus. f–i, Mendelian Randomization analysis of SNPs that are independently (linkage disequilibrium r²<0.2) associated with expression of RIPK1 and significantly (P<0.05) associated with obesity metrics. The beta and standard error of the effect sizes of RIPK1 expression (eQTL, x-axis) and the effect sizes on obesity metrics (y-axis) are plotted. f- UKHLS& GTEX (Adipose tissue) g- UKHLS & METSIM (Adipose tissue) h- UK Biobank & GTEX (Adipose tissue), i- UK Biobank & Llyod-Jones, 2017 (Whole Blood).