ReverseGWAS identifies combined phenotypes associated with a genotype in GWA studies

Area of Science:

• Genetics
• Computational Biology
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) traditionally link single genetic variants to single phenotypes.
• Modern GWAS often collect data on multiple phenotypes, creating opportunities to explore genetic variant-phenotype relationships from a new perspective.
• Elucidating the phenotypic architecture of a genetic variant involves identifying combinations of phenotypes associated with it.

Purpose of the Study:

• To introduce ReverseGWAS, an algorithmic platform designed for analyzing large-scale, multi-phenotype GWAS.
• To demonstrate the capability of ReverseGWAS in identifying complex phenotypic patterns associated with genetic variants.
• To apply ReverseGWAS to real-world datasets and validate findings in independent cohorts.

Main Methods:

• Development of the ReverseGWAS algorithmic platform for multi-phenotype GWAS analysis.
• Testing ReverseGWAS on simulated data to assess its performance in identifying phenotype combinations under varying noise levels.
• Application of ReverseGWAS to UK Biobank data, analyzing associations with autoimmune diseases and common ICD-10 codes.
• Replication analysis of identified associations using the FinnGen independent cohort.

Main Results:

• ReverseGWAS effectively identified logical combinations of phenotypes associated with genetic variants in simulated data, even with noise.
• Analysis of UK Biobank data yielded 719 candidate associations for autoimmune diseases and 205 for common ICD-10 codes.
• A significant majority of these candidate associations (546/719 and 111/205, respectively) were successfully replicated in the FinnGen cohort.

Conclusions:

• ReverseGWAS is a powerful tool for uncovering the phenotypic architecture of genetic variants in large-scale multi-phenotype GWAS.
• The platform demonstrates high performance in identifying and replicating complex genetic associations.
• The findings highlight the potential of ReverseGWAS for discovering novel genotype-phenotype relationships in human diseases.