DSSS - Haplotype effects on gene expression and the evolution of complex traits

The regulation of gene expression is typically mediated by multiple independent associations each parsimoniously assumed to be due to the activity of a single causal variant. In this talk I will challenge this assumption by reporting results of experiments using expression CROPseq to fine-map regulatory potential for 4,382 SNPs in 87 loci associated with immune disease. Multiple variants are observed to contribute to each of 66 fine-mapped eQTL, with some cell-state specificity at different states of myeloid differentiation. Base editing also confirms that allelic substitution validates effects initially identified as indels. These results imply that haplotype effects are likely to explain many of the associations, and computational modeling confirms that cell-state specific polymorphisms embedded in haplotypes best explain complex GWAS-eQTL relationships. Relatively rapid evolution of haplotype structure also suggests a novel mechanism for divergence of genetic effect sizes. I will discuss broad implications for interpretation of GWAS findings, transferability of polygenic scores, and the evolution of quantitative traits.