Extended Data Fig. 8: Characterization and internal replications of trans-eQTLs.

(a) Comparison of imputation quality (DR²) of trans-eQTLs and all tested SNPs. Dashed lines are median values of DR². (b) Inter-chromosomal linkage disequilibrium (LD) (the genotype correlation in absolute values) between trans-eQTLs and cis-eQTLs of the same genes in muscle. (c) Comparison of LD of cis-eQTLs vs. trans-eQTLs of the same genes and that of cis-eQTLs vs. random SNPs with matched minor allele frequency (MAF) and chromosomes. The comparisons of all tissues are statistically significant (P < 0.05, the two-sided Student’s t-test). Box plots depict the interquartile range (IQR), whiskers depict 1.5 × IQR. (d) Tissue-sharing patterns (π₁ statistics) of trans-eQTLs across tissues and replicates. Muscle_G1 (n = 435) and Muscle_G2 (n = 435) are two replicates of muscle samples by dividing the whole muscle samples randomly into two groups. Similarly, Blood_G1 (n = 349) and Blood_G2(n = 349) are two replicates of blood samples. (e) Pearson correlations of effect sizes (beta values) of trans-eQTLs in one tissue (x-axis) and those of matched SNPs in another tissue (y-axis). (f) Pearson correlation of effect sizes (beta values) of trans-eQTLs (n = 5,782) in blood tissue in Group1 (G1, n = 349) and those of matched SNPs in Group2 (G2, n = 349) (r = 0.66, the two-sided Student’s t-test: P < 2.2 × 10⁻³⁰⁸). (g) Pearson correlation of effect sizes (beta values) of trans-eQTLs (n = 4,344) in muscle tissue in Group1 (G1, n = 435) and those of matched SNPs in Group2 (G2, n = 435) (r = 0.75, the two-sided Student’s t-test: P < 2.2 × 10⁻³⁰⁸).