Integrative Multi‐Omics Analysis Prioritizes Candidate Genes for Essential Tremor and Reveals a Gap Between Computational Prediction and Experimental Validation

Abstract Background The genetic architecture of essential tremor (ET) remains incompletely understood. A key challenge is translating genome‐wide association study (GWAS) loci into specific effector genes to elucidate disease mechanisms and develop targeted therapies. Objective To implement a multistage computational framework to prioritize high‐confidence candidate genes for ET and to assess these predictions against publicly available, patient‐derived transcriptomic data. Methods We employed a convergent evidence strategy to prioritize genes, integrating cross‐tissue (UTMOST) and tissue‐specific (FUSION) transcriptome‐wide association studies (TWAS) with gene‐based association tests (MAGMA). Prioritized genes were subjected to causal inference analyses (summary‐data‐based Mendelian randomization [SMR] and colocalization), co‐expression network analysis (GeneMANIA), and pharmacogenomic analysis (DGIdb). We leveraged spatial transcriptomics to characterize gene expression patterns across cortical layers and cell types. Finally, we validated computational predictions using two independent post‐mortem brain datasets from ET patients and controls. Results Our prioritization pipeline identified 12 high‐confidence candidate genes. Co‐expression network analysis revealed 83.3% of candidates exhibit functional relationships, forming three modules centered on RNA processing ( NRBP1 ), metabolic regulation ( SLC5A6 ), and nucleotide synthesis ( CAD ). Pharmacogenomic analysis demonstrated 66.7% of candidates possess therapeutic target potential. Spatial transcriptomics revealed preferential expression in cortical Layer 5 pyramidal neurons. However, validation in post‐mortem cerebellar tissue showed no significant differential expression. Conclusions Our study provides a robust pipeline for ET gene prioritization and puts forward a novel cortical hypothesis for the disease. The discordance between strong computational predictions and their lack of validation in available patient tissue highlights a critical gap in the field. © 2025 International Parkinson and Movement Disorder Society.