A composite-likelihood approach for identifying polymorphisms that are potentially directly associated with disease

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Linkage analysis identifies regions associated with disease, but causal variants require further investigation.
• Candidate single nucleotide polymorphisms (SNPs) are tested individually to explain linkage signals, which can be inefficient.

Purpose of the Study:

• To develop a novel statistical method for genetic analysis that combines multiple candidate SNPs.
• To improve the power of detecting disease-associated loci compared to single SNP approaches.

Main Methods:

• A composite-likelihood approach is proposed to analyze two or more candidate SNPs simultaneously.
• The method tests whether the combined association of multiple SNPs explains the observed linkage signal.

Main Results:

• Simulations demonstrate that the proposed composite-likelihood method significantly increases statistical power.
• The new method outperforms traditional single SNP analysis in detecting causal variants.

Conclusions:

• Combining multiple candidate SNPs using a composite-likelihood approach is a more powerful strategy for genetic association studies.
• This method enhances the ability to identify disease-related genetic variants in complex regions.