Transferability of the PRS estimates for height and BMI obtained from the European ethnic groups to the Western Russian populations
Elise Albert, Olga Kondratieva, Elena Baranova, O. V. Sagaydak, Maxim S. Belenikin, Gaukhar Zobkova, E. S. Kuznetsova, Andrei A. Deviatkin, A. A. Zhurov, E. A. Karpulevich, P. Y. Volchkov, Maria Vorontsova
Abstract
Genetic data plays an increasingly important role in modern medicine. Decrease in the cost of sequencing with subsequent increase in imputation accuracy, and the accumulation of large amounts of high-quality genetic data enable the creation of polygenic risk scores (PRSs) to perform genotype-phenotype associations. The accuracy of phenotype prediction primarily depends on the overall trait heritability, Genome-wide association studies cohort size, and the similarity of genetic background between the base and the target cohort. Here we utilized 8,664 high coverage genomic samples collected across Russia by "Evogen", a Russian biomedical company, to evaluate the predictive power of PRSs based on summary statistics established on cohorts of European ancestry for basic phenotypic traits, namely height and BMI. We have demonstrated that the PRSs calculated for selected traits in three distinct Russian populations, recapitulate the predictive power from the original studies. This is evidence that GWAS summary statistics calculated on cohorts of European ancestry are transferable onto at least some ethnic groups in Russia.