recombClust

recombClust analyzes chromosome populations to identify and group chromosomes by historical recombination patterns using SNP-phased genotype data and mixture-model classification.

Key Features:

• Mixture Model Approach: Applies a mixture model that evaluates pairs of SNP-blocks to differentiate chromosomal subpopulations based on linkage disequilibrium (LD) and recombination rates.
• SNP-block Pair Evaluation: Evaluates pairs of SNP-blocks within chromosomes to detect signals of historical recombination.
• Chromosome Clustering: Aggregates classifications across multiple SNP-block pairs to cluster chromosomes with similar recombination histories.
• Input Data: Operates on SNP-phased genotype data to enable haplotype-resolved detection of recombination patterns.

Scientific Applications:

• Prediction of recombination modifier alleles: Predicts alleles associated with known recombination modifiers in simulations and real datasets.
• Inversion detection: Identifies common inversions in Drosophila melanogaster and humans.
• Selection scans at lactase locus: Detects chromosomes under selective pressure at the human lactase locus.
• Characterization of complex regions: Resolves distinct recombination histories in complex regions such as human 1q21.1, associating those histories with ANKRD35 expression in whole blood and with variants linked to hypertension and nonallelic homologous recombination (NAHR)-related deleterious phenotypes.

Methodology:

Applies a mixture model to pairs of SNP-blocks from SNP-phased genotype data to classify chromosomes into subpopulations defined by LD and recombination, then aggregates classifications across multiple SNP-block pairs to cluster chromosomes and detect differences in historic recombination patterns.