Phenome-Wide Association Study Utilizing Imputation Identifies Common Genetic Variants Shared by Type 1 Diabetes and Multiple Diseases

Matthew Eidenschink
University of Minnesota Medical School

Matthew Eidenschink1, Jamie C. Fox1, John G. Mayer2, Zhan (Harold) Ye2, Brian A. Hoch2, and Scott Hebbring1
1Center for Human Genetics, 2Biomedical Informatics Research Center

Research area:  Genetics

Background: Type-1 diabetes mellitus (T1D) is an autoimmune disease with complex genetic influence. Genome-wide association studies (GWAS) have identified 118 single nucleotide polymorphisms (SNPs) associated with an increased risk for T1D. Using the complimentary phenome-wide association study (PheWAS) approach, T1D-associated SNPs were used to find common genetic etiology between T1D and other conditions. 

Methods: Personalized Medicine Research Project (PMRP) patients were previously genotyped on either the Illumina 660W-Quad SNP chip or the Illumina Expanded Human Core Exome SNP chip. PMRP genotype data was enhanced by imputation to test initially-untyped genetic variants for association with T1D and other conditions defined by International Classification of Diseases version 9 (ICD-9) codes. PheWAS data from 115 of the GWAS-significant T1D SNPs was compiled from approximately 12,000 PMRP patients. Significance threshold for top PheWAS associations was determined using the Bonferroni correction.

Results: PheWAS analysis found multiple statistically-significant relationships between T1D and other diseases. The following conditions were found to share a common genetic etiology with T1D: arthropathy (P=2.66*10-10), rheumatoid arthritis (P=1.37*10-8), hypothyroidism (P=6.87*10-10), thyrotoxicosis (P=2.31*10-9), urinary obstruction (P=2.54*10-8), blisters with epidermal loss due to second degree burn of upper arm (P=2.61*10-8) and congenital anomalies of limbs (P=2.68*10-8).

Conclusions: This comprehensive PheWAS identified multiple SNP-disease associations suggesting that T1D shares a common genetic etiology with thyroid disease, rheumatoid arthritis, urinary obstruction, blisters with epidermal loss, and congenital anomalies of limbs. Further research on the pathways shared through these genetic variants may provide targets for future drug-repurposing studies, which will ultimately improve how these diseases are predicted, diagnosed and treated.