SNPSpD

SNPSpD applies spectral decomposition to pairwise linkage disequilibrium matrices to correct for non-independence among single-nucleotide polymorphisms (SNPs) and adjust multiple-testing in genetic association analyses.

Key Features:

• Spectral Decomposition (SpD): Performs spectral decomposition of matrices containing pairwise linkage disequilibrium (LD) between SNPs to capture correlation structure.
• Multiple Testing Correction: Uses the SpD-derived component structure to adjust significance thresholds for correlated SNPs, reducing false positives due to LD.
• Computational Efficiency: Provides an analytical correction approach that contrasts with computationally intensive permutation tests.
• Eigenvalue Output: Computes eigenvalues that quantify variance explained by each principal component of the LD matrix.
• Principal-Component Coefficients: Produces principal-component coefficients describing each SNP's contribution to components derived from the LD matrix.
• Factor Loadings with Varimax Rotation: Generates factor "loadings" after varimax rotation to aid identification of SNP subsets that maximize regional informational content.

Scientific Applications:

• Genome-wide Association Studies (GWAS): Adjusts multiple-testing corrections for correlated SNPs to improve the reliability of association signals.
• Informative SNP Selection: Identifies subsets of SNPs that capture regional variation for targeted analysis or follow-up studies.
• Study Design Optimization: Informs marker selection and significance thresholding when planning genetic association analyses.
• LD Structure Characterization: Provides component and loading outputs that support interpretation of local linkage disequilibrium architecture.

Methodology:

SNPSpD performs spectral decomposition of pairwise LD matrices, computes eigenvalues and principal-component coefficients, applies varimax rotation to obtain factor loadings, and uses the SpD-derived component structure to adjust for multiple testing among correlated SNPs.