Uncovering the role of rare and low-frequency mutations in stroke using a polygenic risk score informed approach – implications for risk prediction

• Applicants: Professor Dr. Martin Dichgans; Dr. Rainer Malik
• Subject Area: Clinical Neurology; Neurosurgery and Neuroradiology ; General Genetics and Functional Genome Biology ; Molecular and Cellular Neurology and Neuropathology
• Term: since 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 497256604

Stroke is the leading cause of long-term disability, the second most common cause of death, and a major cause of dementia worldwide and thus represents a major global health problem in need for novel therapeutic strategies. Genetics has started to transform vascular medicine by providing novel drug targets and insights into the etiology and pathophysiology of cardiovascular disease. Up to now, genome-wide association studies (GWAS) revealed 35 risk loci harboring common risk variants for stroke and stroke subtypes. However, these loci collectively only explain a small proportion of the phenotypic variance. Hence, there is a requirement for high-resolution fine-mapping studies of variants at and beyond previously established risk loci to identify high-impact risk variants for stroke and enable the functional exploration of such variants. The overarching goal of this proposal is to identify genes harboring rare and low-frequency variants with large effects on ischemic stroke risk. To this end, we have assembled a previously untapped sample of ~4,000 ischemic stroke cases with available GWAS data for whole exome sequencing (WES) and have access to ~6,000 WES controls. While many WES experiments suffer from low power for discovery, we will increase power through a polygenic risk score (PRS) informed approach. Specifically, sampling from an extreme phenotype distribution will increase power in our WES setting. We recently showed that individuals in the top 0.25% of a PRS distribution for ischemic stroke have a >3x increased risk for ischemic stroke. This enables us to calculate the PRS for each individual in our dataset, putting all individuals on a continuous risk spectrum. We will select cases at the lower half (2000 / 4000) of the polygenic risk distribution and controls at the upper half (3000 / 6000) of the polygenic risk distribution, thus enhancing the power to identify novel low-frequency and rare risk variants and genes. We will perform both single-variant and gene-based tests for enrichment of rare variants. To further increase power we will add external, publicly available controls to our analysis. Findings will be replicated in international large-scale sequencing studies (UK Biobank Exome Project, deCode Genetics, TOPMed) and integrated into a novel improved PRS for the prediction of ischemic stroke. Individual variants and genes will be selected for genome-editing in human induced pluripotent stem cells and functional exploration in relevant cellular models. The outputs of this project will define novel molecular targets and pathways, provide formidable starting points for functional studies, and facilitate the development of novel therapeutics for stroke as well as and improve risk prediction for stroke and cardiovascular disease based on polygenic scores.

• DFG Programme: Research Grants