Single-Cell cis-Mendelian Randomization Reveals Cell-Specific Genetic Mechanisms Underlying Atopic Dermatitis

Atopic dermatitis (AD) is a chronic inflammatory skin disease with a complex and highly polygenic genetic architecture, in which immune-mediated mechanisms play a central role. Here, we integrated single-cell cis-expression quantitative trait loci from 14 immune cell types with AD GWAS summary statistics using a two-sample Mendelian Randomization (MR) framework to resolve cell-specific genetically mediated transcriptional effects. We identified 303 significant cell-specific gene–trait associations with limited overlaps across cell types. A multi-step prioritization strategy refined these findings to 35 genes across all 14 cell types. A comparison with whole blood cis-eQTLs revealed a limited concordance, suggesting an attenuation of cell-specific regulatory effects in bulk transcriptomic approaches. Intersecting single-cell and bulk evidence identified 22 high-confidence genes with a relatively independent mechanism of action. Integrative annotation implicated several immune-relevant and druggable genes, including IL2RA, with distinct cell-specific effects. Our findings demonstrate diverse mechanisms of risk genes for AD at the single-cell level that act across immune cell states and pathways, with implications for therapeutic interventions.