To date, the aim of most genomic or transcriptomic-scale studies of splicing in AUD has been to reveal how alcohol consumption affects splicing [12, 66]. Multiple studies have indicated that specific RNA splicing events are important in the brain and in neurological disorders [2, 3]. However, our understanding of the causal role of splicing on the susceptibility of AUD is limited. Therefore, we implemented a Mendelian randomization (MR)-based strategy to systematically assess the causality of splicing events in AUD. MR typically utilizes a modeling approach to map genetic variants to the molecular trait of gene expression, such as PrediXcan, TWAS, or SMR [15–17]. The purpose of the modeling is not to simply predict the molecular trait; rather, it is to determine the extent to which genetic variants can explain the molecular trait. In this study, we adapted MR from a gene expression-centric approach to accommodate RNA splicing analysis.
