mPopTag

mPopTag optimizes selection of linkage disequilibrium (LD) tag single nucleotide polymorphisms (SNPs) across multiple populations using HapMap LD data to enable efficient representation of genetic variation for whole-genome association and candidate gene studies.

Key Features:

• Multi-Population Analysis: Analyzes patterns of linkage disequilibrium across different ethnic groups using metrics such as composite LD (CLD) or r² between polymorphic sites.
• HapMap Data Utilization: Leverages HapMap LD data covering approximately 3.7 million SNPs for initial tag SNP selection.
• Algorithmic Efficiency: Implements a generalized greedy algorithm based on Carlson et al. (2004) to select near-minimal sets of tag SNPs that represent genetic variation across multiple populations.
• Resequencing Data Integration: Incorporates additional tag SNPs derived from resequencing data, including the Environmental Genome Project (EGP), which has resequenced over 500 genes, to improve gene-tagging proportions.
• Ethnic-Specific Optimization: Supports optimization that demonstrates ethnic-specific datasets yield superior tagging performance compared to ethnic-mixed datasets.

Scientific Applications:

• Whole-Genome Association Studies: Enables selection of representative tag SNPs to reduce genotyping burden while preserving coverage of genetic variation across populations.
• Candidate Gene Analyses: Improves gene-tagging by combining HapMap and resequencing-derived SNPs to approach near-complete SNP ascertainment in candidate genes.
• Population-Specific Association and Personalized Medicine: Facilitates identification of genetic associations across diverse ethnic groups to inform population-specific and personalized analyses.

Methodology:

Uses HapMap LD data (~3.7 million SNPs) for initial SNP sets; computes LD metrics such as composite LD (CLD) or r² between polymorphic sites across multiple populations; applies an improved greedy algorithm generalizing Carlson et al. (2004) to select near-minimal LD tag SNP sets across populations; integrates resequencing-derived SNPs (e.g., EGP resequenced >500 genes) to augment tagging.