Abstract
The causal contributions of specific mitochondrial genes to common pain phenotypes remain unclear. We employed a multi-omics Mendelian randomization (SMR) approach, integrating QTL data (expression, methylation, protein) for mitochondrial genes with GWAS summary statistics for seven pain phenotypes. We identified 18 candidate genes with robust SMR associations across omics layers. However, strong colocalization evidence (PP.H4 > 0.7) was largely absent, pointing towards complex genetic architectures. A notable exception was a strong signal for a shared causal variant found at the methylation level for the MCL1 gene in hip pain (PP.H4 = 0.962), nominating it as a high-confidence candidate. Additionally, genetically predicted higher protein levels of Glycine amidinotransferase (GATM) showed consistent protective associations with neck or shoulder, back, and knee pain. This study provides novel evidence for mitochondrial gene regulation in pain, highlighting the GATM pathway as protective and identifying MCL1 methylation as a potential causal mechanism in hip pain.