Identification of Associations with Dermatologic Diseases through a Focused GWAS of the UK Biobank

Abstract

The UK Biobank includes genotype information for about 500,000 patients for over 7000 phenotypes. However, owing to multiple testing correction for approximately 200 billion tests, many clinically and statistically significant associations remain unappreciated. We perform a focused analysis of the UK Biobank for 13 dermatologic conditions, including malignant melanoma, melanoma in situ, squamous cell carcinoma, basal cell carcinoma, actinic keratosis, seborrheic keratosis, psoriasis, lichen planus, systemic lupus erythematosus, hyperhidrosis, pilonidal cyst, sebaceous cyst, and lipoma. We identify 447 sentinel variants, which are enriched for protein-coding variants and an elevated combined annotation-dependent depletion (CADD) score compared with background variants. Through gene ontology enrichment analysis, we identify known pathways involved in melanoma, actinic keratoses, and squamous cell carcinoma and uncover additional pathways. We also uncover 5 protein-coding variants, which, to our knowledge, have not been previously reported, including LRP3 for lipomas, PLCD1 for sebaceous cysts, EIF3CL for lichen planus, TTK for pilonidal cysts, and MAPK15 for systemic lupus erythematosus.