A two-phase Bayesian methodology for the analysis of binary phenotypes in genome-wide association studies

Area of Science:

• Genetics
• Statistical Genetics
• Computational Biology

Background:

• Genome-wide association studies (GWAS) face the
• large p small n
• problem due to advances in sequencing technologies, where the number of genetic markers (p) far exceeds the number of subjects (n).
• Joint analysis of numerous genetic markers is computationally challenging, while marginal analysis alone is insufficient for identifying complex trait associations.

Purpose of the Study:

• To propose a novel Bayesian two-phase methodology for jointly analyzing genetic markers and binary traits.
• To effectively control for confounding factors in genetic association analyses.
• To develop a computationally efficient method for identifying sparse genetic marker associations.

Main Methods:

• A two-phase Bayesian approach is employed.
• Phase 1: Marginal scan to identify a reduced set of candidate genetic markers.
• Phase 2: Hierarchical modeling for joint analysis of candidate markers, utilizing a novel maximum a posteriori estimation technique for sparse selection.

Main Results:

• The proposed Bayesian method effectively handles the
• large p small n
• problem in GWAS.
• The two-phase approach successfully identifies relevant genetic markers for binary traits.
• Numerical studies and a genome-wide application for colorectal cancer demonstrate the method's performance and efficiency.

Conclusions:

• The developed Bayesian two-phase methodology provides an effective solution for joint genetic marker analysis in GWAS.
• The approach facilitates the identification of sparse genetic associations, improving the understanding of complex traits.
• This method offers a computationally efficient tool for genetic research, particularly in studies involving a large number of markers and limited sample sizes.