We integrated genome wide association signals for AN with genetically regulated gene expression to identify novel risk genes and biological insights into the disorder. Critically, through the application of methods such as conditional analysis and finemapping, we refined a smaller set of genes with greater confidence of true association that can be subjected to future follow-up study. Leveraging proteomic and alternative splicing data also revealed association signals that were not seen using mRNA data alone. While previous studies have uncovered potential AN risk genes, our use of statistical finemapping and conditional analysis allowed us to prioritise confident associations such as WDR6 that more likely exert a causal effect. Upregulation of WDR6 exhibited a plausible causal relationship with AN across all methods with available data (TWAS, conditional analysis, finemapping). While the surrounding region is rich with genes predicted to have an association, a conditionally significant TWAS signal along with the finemapping evidence suggests that WDR6 is a causal gene at this locus, however, further analyses such as SNP based finemapping and in silico prediction of variants in this locus are warranted
