Genetic control of the human brain proteome

Abstract

We generated an online brain pQTL resource for 7,376 proteins through the analysis of genetic and proteomic data derived from post-mortem samples of the dorsolateral prefrontal cortex of 330 older adults. The identified pQTLs tend to be non-synonymous variation, are over-represented among variants associated with brain diseases, and replicate well (77%) in an independent brain dataset. Comparison to a large study of brain eQTLs revealed that about 75% of pQTLs are also eQTLs. In contrast, about 40% of eQTLs were identified as pQTLs. These results are consistent with lower pQTL mapping power and greater evolutionary constraint on protein abundance. The latter is additionally supported by observations of pQTLs with large effects' tending to be rare, deleterious, and associated with proteins that have evidence for fewer protein-protein interactions. Mediation analyses using matched transcriptomic and proteomic data provided additional evidence that pQTL effects are often, but not always, mediated by mRNA. Specifically, we identified roughly 1.6 times more mRNA-mediated pQTLs than mRNA-independent pQTLs (550 versus 341). Our pQTL resource provides insight into the functional consequences of genetic variation in the human brain and a basis for novel investigations of genetics and disease.

Keywords: brain; eQTL; expression; gene regulation; pQTL; proteomics; quantitative trait locus.

Figure 1

Enrichment of proximal pQTLs (A) Enrichment of proximal pQTLs by genomic context. (B) Enrichment of proximal pQTLs in exonic coding regions. For each annotation, enrichment was evaluated on the basis of a Fisher’s exact tests assessing the overlap between pQTLs and the annotated SNVs. The estimated odds ratio (OR) and 95% confidence interval (CI) is shown for each test.

Figure 2

Comparisons between genetic effects on mRNA and protein abundance (A) Comparison of the effect of each variant on mRNA and protein abundance. Each point represents one SNV tested against the abundance of the mRNA and protein of a single gene. The eQTLs (defined on the basis of False Discovery Rate [FDR] < 0.05) are shown in green, the pQTLs (defined on the basis of FDR < 0.05) are shown in blue, and the sites that are both an eQTL and a pQTL are shown in red. The shown effects are t-statistics. (B) Replication rates (π₁) of pQTLs and eQTLs at FDR < 0.05. These analyses included the lead discovery QTLs for each gene tested in both studies. (C) Comparison of the distribution of the size of genetic effects on mRNA and protein abundance for variants influencing both. The effect size is the absolute value of the pQTL or eQTL t-statistic. The boxes reflect the values corresponding to the first and third quartile, the horizontal line within the box reflects the median, the lines extending from the box represents the range of values within 1.5 times the interquartile range, and points beyond the line are plotted individually. (D) Comparison of the genetic effects on mRNA and protein abundance for genes with protein expression mediated by mRNA. (E) Comparison of the genetic effects on mRNA and protein abundance for genes with protein expression not mediated by mRNA. For all analyses the genetic effects on protein and mRNA abundance were from the discovery ROSMAP pQTL results and Sieberts et al. (2020), respectively.