Polygenic transcriptome risk scores enhance predictive accuracy in atopic dermatitis

• DOI: 10.1186/s12967-025-06570-8
• Publication Date: 2025-05-23T12:52:18Z
• Authors: Charalabos Antonatos; Ashley Budu-Aggrey; Alexandros Pontikas; Adam Akritidis; Efstathia Pasmatzi; Aikaterini Tsiogka; Stamatis Gregoriou; Katerina Grafanaki; Lavinia Paternoster; Yiannis Vasilopoulos
• Keywords: Polygenic risk score

Abstract Background Incorporation of gene expression when estimating polygenic risk scores (PRS) in atopic dermatitis (AD) may provide additional insights disease pathogenesis and enhance predictive accuracy. In this study, we developed transcriptome (PTRSs) derived from AD-enriched tissues evaluated their performance against traditional PRS models a baseline model incorporating eosinophil lymphocyte counts the prediction AD. Methods We conducted transcriptome-wide association studies (TWAS) using PrediXcan framework to construct tissue-specific PTRSs. Risk score was assessed 256,888 Europeans (10,816 cases) validated an independent cohort 64,152 (2669 UK Biobank. Results observed modest correlation between PTRS, exerting effects on AD risk. While demonstrated superior compared single-tissue PTRSs, combining both significantly enhanced accuracy, yielding c-statistic 0.646 (95% confidence intervals: 0.634–0.656). Notably, PTRSs revealed stronger associations with factors, where Eppstein-Bar virus (EBV)-transformed lymphocytes unexposed skin reported positive counts. Conclusions Our findings highlight value integrating transcriptome-based omics layer refine our understanding genetic architecture complex traits.