Skip to main content
Premium Trial:

Request an Annual Quote

Genetically Predicted Gene Expression Analysis Provides Look at Plasma Proteome Regulation

NEW YORK – A team from Loyola University Chicago, the Broad Institute, and other US centers has used cross-tissue gene expression predictions to uncover regulatory features influencing the blood plasma proteome. Their study demonstrates that the genetic factors that influence gene expression can offer a window into the plasma proteome and related traits and conditions.

For a paper appearing in the American Journal of Human Genetics on Monday, the researchers relied on a transcriptome-wide association study to search for ties between genetically predicted gene expression and observed protein abundance in 3,301 genotyped participants in a UK-based study of blood donors of European ancestry, known as the INTERVAL study.

Using blood plasma measurements of 3,622 proteins with SomaLogic's SomaScan assay, together with prediction models from the Genotype-Tissue Expression project across more than four dozen tissues, the team considered potential associations between plasma protein levels and predicted gene expression within each GTEx tissue. Apparent associations were subsequently validated using data for 971 multi-ancestry participants in the Trans-omics for Precision Medicine (TOPMed) study.

"We applied a transcriptome-wide association study (TWAS) framework to proteomic data, testing the genetically predicted expression of genes across 49 tissues for association with the observed abundance of plasma proteins," the study's senior and corresponding author Heather Wheeler, a researcher at Loyola University Chicago, said in an email.

She and her colleagues highlighted 1,168 cis-acting associations between proteins and nearby regulatory sites. But the TWAS approach also made it possible to unearth 1,210 more distant trans-acting associations, including some that were missed by more conventional quantitative trait locus (QTL) analyses.

"We show that TWAS for protein levels is an effective method for identifying replicable trans-acting associations between predicted transcripts and proteins," the researchers wrote. "We also found a high expected proportion of true positives for associations between the predicted transcripts and protein products of the same underlying gene."

When they looked more closely at the proteins targeted by trans-acting factors, the investigators saw an apparent overlap with autoimmune disease associations found through prior genome-wide association studies and with regulatory elements such as transcription factor binding sites.

"Autoimmune disease enrichment is somewhat expected given the proteins in our TWAS were measured in blood plasma," they wrote, suggesting that expression shifts in pleiotropic genes linked to autoimmune conditions "could lead to the progression of autoimmune diseases."

More generally, the findings point to the potential advantages of using genetic predictions for gene expression to dig into the potential functional effects of genetic variants implicated in complex traits or conditions, Wheeler explained, particularly since actual blood protein levels appeared to coincide more closely with genetics-based expression predictions than with actual gene expression measurements.

"[T]he genetic component of gene expression more strongly correlates with protein levels than total observed expression, which includes genetic and environmental variation," she said, "and is therefore more useful in uncovering the functional mechanisms of SNPs associated with complex traits."