PIP-SNP

PIP-SNP performs linkage disequilibrium-based SNP binning, k-nearest neighbors (kNN) genotype imputation, and synthetic marker synthesis to reduce genotype data dimensionality while preserving genetic information for GWAS and other genetic analyses.

Key Features:

• Linkage Disequilibrium Bin Mapping: Uses linkage disequilibrium (LD) and a stochastic process to model SNP signals and compute autocorrelation coefficients to construct LD bins that group genetically linked SNPs for dimension reduction.
• Genotype Imputation: Imputes missing genotype data using the k-nearest neighbors (kNN) algorithm based on similarity between SNP profiles.
• Marker Synthesizing: Identifies optimal synthetic markers that represent each LD bin as dimension-reduced proxies of original genome-wide markers.
• Information Conservation Evaluation: Assesses conservation of genetic information when replacing original SNP markers with synthetic markers to ensure retention of critical signals.

Scientific Applications:

• Genome-wide association studies (GWAS): Facilitates GWAS by reducing SNP dimensionality while preserving association signals for phenotype-genotype analysis.
• Population-scale genotype analyses: Enables analysis of large and diverse datasets, with demonstrated applicability to rice recombinant inbred lines (RIL) and HapMap project data.

Methodology:

Models SNP signals with a stochastic process and computes autocorrelation coefficients to form LD bins; performs k-nearest neighbors (kNN) imputation for missing genotypes; derives synthetic markers representing LD bins; and evaluates information conservation between original and synthetic markers.