Are trait-associated genes clustered together in a gene network?

Area of Science:

• Genetics and Genomics
• Bioinformatics
• Computational Biology

Background:

• Genome-wide association studies (GWAS) identify genetic variants linked to complex traits and diseases.
• Pinpointing causal genes in GWAS is challenging due to linkage disequilibrium (LD) and noncoding regions.
• Gene network approaches assume trait-associated genes cluster, but this has not been empirically validated.

Purpose of the Study:

• To empirically test the assumption that trait-associated genes cluster within gene networks.
• To leverage whole exome sequencing (WES) data to directly identify trait-associated genes.
• To assess the utility of gene network proximity for prioritizing causal genes in genetic studies.

Main Methods:

• Utilized exome-based association statistics from the UK Biobank whole exome sequencing (WES) data.
• Employed two distinct gene network types to analyze gene proximity.
• Compared the proximity of trait-associated genes to randomly selected genes within the networks.

Main Results:

• Found that trait-associated genes were significantly more proximal to each other than random gene sets in both networks.
• Empirically validated the core assumption of gene network-based approaches for genetic association studies.
• Demonstrated that trait-associated genes exhibit significant clustering behavior within biological networks.

Conclusions:

• The study confirms that trait-associated genes are clustered within gene networks.
• This validated assumption can be leveraged to enhance the statistical power of genome-wide association studies (GWAS).
• Future GWAS analyses can potentially use less stringent p-value thresholds by incorporating gene network information.