NEW YORK – An international team led by investigators at Stanford University has identified genetic contributors to dozens of blood- or urine-based biomarkers in individuals of European descent, leading to a better understanding of causal variants behind these markers as well as risk scores for related conditions.
For a paper appearing in Nature Genetics, the researchers searched for common, low-frequency, and rare genetic contributors to 35 blood serum and urine biomarkers, using directly genotyped and imputed alterations for more than 360,000 UK Biobank participants with available laboratory measurements. In the process, they tracked down nearly 3,400 fine-mapped associations falling at 1,857 loci linked to one or more of the biomarkers considered, along with human leukocyte antigen (HLA) variants and copy number changes with large effects on biomarker genetics.
"Our results home in on the genetic basis of biomarkers, their causal influence on disease, and improve genetic risk stratification for common diseases," co-senior and co-corresponding author Manuel Rivas, a biomedical data science researcher at Stanford University, said in an email.
Expanding from these findings, the team searched for causal ties to the 40 clinical phenotypes related to the urine and blood biomarkers in question, reasoning that "the genetic predisposition to particular biomarker states and the factors that confound them may have implications for disease treatment."
From Mendelian randomization analyses, for example, the investigators focused in on 51 causal associations with disease conditions or features.
By bringing individual biomarker polygenic risk scores (PRS) together with the help of a population-scale modeling algorithm known as BASIL, meanwhile, they came up with "multi-PRS" scores for stratifying the European participants based on their genetic propensity for conditions ranging from type 2 diabetes and chronic kidney disease to gout and alcoholic cirrhosis — associations that they subsequently verified using data for another 135,500 individuals enrolled in the FinnGen study.
"Given that biomarker values are already routinely collected in structured data formats," the researchers reported, "we anticipate that multi-PRS methods could inform clinical practice in the coming years, as a larger fraction of the population is genotyped and sequenced."
Rivas cautioned that the risk scores developed so far were established with data for European participants, meaning they may not transfer directly to non-European populations.
"We hope that biomarker studies in non-European populations, and improved methodology to consider the mapping of genetic variants to causal variants, will improve the predictive performance across population groups," he said.
"The genome-wide resource made available with this study, including the association summary statistics, fine-mapped regions, and polygenic predictions models, provides a starting point for causal mapping of genetic variants affecting the 35 biomarkers and their relevance to medical phenotypes," he and his co-authors concluded. "These results highlight the benefits of direct measurements of biomarkers in population cohorts for interpreting the genetic basis of biomarkers and improved prediction of multiple common diseases."