Locus and gene-based GWAS meta-analysis identifies new diabetic nephropathy genes

Area of Science:

• Genetics and Genomics
• Computational Biology
• Nephrology

Background:

• Genome-wide association studies (GWAS) are crucial for understanding complex disorders but often face challenges with false positives and negatives.
• Diabetic nephropathy (DN) genetics remain incompletely understood, with limited gene identification despite numerous GWAS.
• The Assimilation of SNPs Interacting in Synchrony (OASIS) algorithm offers a novel locus-based clustering approach to improve GWAS accuracy.

Purpose of the Study:

• To apply the OASIS algorithm to existing DN dbGaP GWAS datasets to identify novel candidate genes.
• To validate identified DN genes using established genetic association testing methods.
• To demonstrate the utility of OASIS in enhancing gene discovery for complex diseases using large-scale genetic data.

Main Methods:

• Application of the OASIS locus-based clustering algorithm to three DN dbGaP GWAS datasets (4725 subjects, 1.06 million SNPs).
• Verification of identified DN genes through single variant replication in standard association studies.
• Gene-based analysis using the GATES method for further validation and discovery.

Main Results:

• OASIS identified 19 potential DN genes.
• Confirmed CARS and FRMD3 as DN genes; discovered five known diabetes-associated genes (NLRP3, INPPL1, PIK3C2G, NRXN3, TBC1D4) not previously linked to DN in these datasets.
• Identified three novel DN genes (NTN1, EBF2, DNAH11) that replicated across multiple analyses.

Conclusions:

• The OASIS algorithm effectively identifies novel genes associated with diabetic nephropathy from existing GWAS data.
• A composite analysis strategy combining OASIS, gene-based, and single variant testing enhances the reliability of gene discovery.
• This integrated approach can be broadly applied to existing GWAS datasets for advancing the genetic understanding of complex disorders.