New multilocus linkage disequilibrium measure for tag SNP selection

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Selecting optimal tag single-nucleotide polymorphism (SNP) sets is crucial for genetic studies.
• Existing methods primarily rely on pairwise linkage disequilibrium (LD) measures (e.g., D', r²), which are insufficient for multi-marker analysis.
• Current LD measures often require haplotype data, increasing costs and complexity.

Purpose of the Study:

• To propose a novel cluster algorithm for grouping SNPs based on a multi-locus LD measure.
• To develop an optimized tag SNP selection strategy within identified SNP clusters.
• To evaluate the performance of the proposed method using genotype data.

Main Methods:

• Development of a multi-locus LD measure based on information theory.
• Application of a cluster algorithm to group SNPs using the novel LD measure.
• Optimization of tag SNP selection within each cluster based on SNP count and prediction accuracy.

Main Results:

• The proposed multi-locus LD measure effectively clusters SNPs.
• The method can be directly applied to genotype data, eliminating the need for costly haplotyping.
• Significantly improved efficiency and prediction accuracy in tag SNP selection were observed.

Conclusions:

• The novel information-theory-based multi-locus LD measure provides a robust approach for SNP clustering.
• The proposed tag SNP selection method enhances accuracy and efficiency while reducing experimental costs.
• This approach offers a valuable alternative for genome-wide association studies and genetic marker selection.