Gene-gene and gene-environment interactions in ulcerative colitis

Area of Science:

• Genetics
• Genomics
• Epidemiology

Background:

• Genome-wide association studies (GWAS) have identified numerous ulcerative colitis (UC) risk loci, but their clinical utility and the role of complex genetic interactions remain unclear.
• Understanding gene-gene and gene-environment interactions is crucial for elucidating UC pathogenesis and improving risk prediction.

Purpose of the Study:

• To evaluate the predictive value of combining 133 UC risk loci with genetic interactions in GWAS data.
• To identify and replicate high-order genetic interactions influencing UC risk using logic regression.
• To explore genetic associations with UC sub-phenotypes.

Main Methods:

• Applied logic regression (LR) to analyze a GWAS dataset for high-order gene-gene and gene-environment interactions in UC.
• Utilized the Wellcome Trust Case-Control Consortium (WTCCC) IBD GWAS as a replication cohort.
• Calculated cumulative allele scores and assessed model predictability using area under the curve (AUC).

Main Results:

• The combination of 133 UC loci achieved good UC risk predictability (AUC = 0.86).
• Logic regression identified and replicated significant genetic interactions, including a gene-smoking interaction, which improved model predictability (AUC = 0.89).
• Inclusion of interaction terms increased explained UC variance from 37% to 42% and suggested a genetic association with primary sclerosing cholangitis (PSC).

Conclusions:

• Logic regression is an effective methodology for identifying and replicating high-order genetic interactions in UC GWAS.
• Combining established UC risk loci with identified genetic interactions significantly enhances UC risk prediction.
• These findings underscore the importance of considering complex genetic interactions for a comprehensive understanding of UC etiology and clinical management.