Brain region-specific effects of nearly fixed sapiens-derived alleles

Abstract

The availability of high-coverage genomes of our extinct relatives, the Neanderthals and Denisovans, and the emergence of large, tissue-specific databases of modern human genetic variation, offer the possibility of probing the effects of modern-derived alleles in specific tissues, such as the brain, and its specific regions. While previous research has explored the effects of introgressed variants in gene expression, the effects of Homo sapiens-specific gene expression variability are still understudied. Here we identify derived, Homo sapiens-specific high-frequency (≥90%) alleles that are associated with differential gene expression across 15 brain structures derived from the GTEx database. We show that regulation by these derived variants targets regions under positive selection more often than expected by chance, and that high-frequency derived alleles lie in functional categories related to transcriptional regulation. Our results highlight the role of these variants in gene regulation in specific regions like the cerebellum and pituitary.

Keywords: Brain; Gene regulation; Human evolution; cis-eQTL.

Fig. 1

A Hierarchical clustering analysis of eQTL normal effect size, not controlled for linkage disequilibrium (LD). Color denotes hierarchical distance. B Number of tissue-specific eQTLs after clumping. Adrenal gland and Amygdala do not contain tissue-specific eQTL in our dataset. C Brain region sample size and eQTL count correlate in our dataset. The blue line marks a polynomial regression line fit, with regression’s standard error confidence intervals (95%) in gray

Fig. 2

Derived, HF eQTLs are present more than expected by chance in selective sweeps from [13] (A) and [25] (B). C shows the count of eQTL overlapping with regions under putative positive selection per region

Fig. 3

Distribution of up and down-regulating ancestral variants across different subsets of the data, in all eGenes. We include here data before (A) and after (B) controlling for linkage disequilibrium in minor alleles (≥10% frequency). A control using major ancestral alleles (at ≥90% frequency) is included (C)