Brady Ryan: FinemappingMR-- A novel Mendelian randomization method using statistical fine-mapping скачать в хорошем качестве

Brady Ryan: FinemappingMR-- A novel Mendelian randomization method using statistical fine-mapping

Abstract: Mendelian randomization (MR) is a statistical method that uses genetic variants as instrumental variables (IVs) to interrogate causal effects of modifiable exposures on health outcomes. There is a growing interest in using MR for drug target prioritization, using molecular traits such as circulating protein levels as exposures. Molecular traits often have multiple correlated causal variants in the same region of the genome. However, most MR methods use only approximately independent instruments. Additionally, many variants are associated with multiple molecular traits, resulting in horizontal pleiotropy. Including multiple correlated IVs while accounting for horizontal pleiotropy can boost power and improve control of type 1 error rates in MR analysis. We present finemappingMR, an MR method that allows for multiple correlated causal variants for the exposure while accounting for potential horizontal pleiotropy using sum-of-single effects (SuSiE) priors. We show in simulations that finemappingMR has better performance over traditional methods. We apply finemappingMR to test the causal effect of circulating plasma protein levels on cholesterol in the UK Biobank and show that finemappingMR avoids false positives due to horizontal pleiotropy identified using other methods. Speaker: Brady Ryan is a PhD student in the Department of Biostatistics at the University of Michigan. His dissertation focuses on two main areas of statistical genetics: 1) developing and applying methods for genome-wide association study (GWAS) summary statistics and 2) developing methods to identify causal variants and molecular pathways for human traits and disease. He has developed methods in the areas of rare-variant aggregation testing, GWAS meta-analysis, and Mendelian randomization. He also collaborates with the METabolic Syndrome in Men (METSIM) study.