Analysis of associations between polygenic risk score and COVID-19 severity in a Russian population using low-pass genome sequencing

Abstract

ABSTRACTThe course of COVID-19 is characterized by wide variability, with genetics playing a contributing role. Through large-scale genetic association studies, a significant link between genetic variants and disease severity was established. However, individual genetic variants identified thus far have shown modest effects, indicating a polygenic nature of this trait. To address this, a polygenic risk score (PRS) can be employed to aggregate the effects of multiple single nucleotide polymorphisms (SNPs) into a single number, allowing practical application to individuals within a population. In this work, we investigated the performance of a PRS model in the context of COVID-19 severity in 1,085 Russian participants using low-coverage NGS sequencing. By developing a genome-wide PRS model based on summary statistics from the COVID-19 Host Genetics Initiative consortium, we demonstrated that the PRS, which incorporates information from over a million common genetic variants, can effectively identify individuals at significantly higher risk for severe COVID-19. The findings revealed that individuals in the top 10% of the PRS distribution had a markedly elevated risk of severe COVID-19, with an odds ratio (OR) of 2.1 (95% confidence interval (CI): 1.4–3.2, p-value = 0.00046). Furthermore, incorporating the PRS into the prediction model significantly improved its accuracy compared to a model that solely relied on demographic information (p-value < 0.0001). This study highlights the potential of PRS as a valuable tool for identifying individuals at increased risk of severe COVID-19 based on their genetic profile.