hmmIBD

hmmIBD detects segments of identity by descent (IBD) in haploid genetic data using a hidden Markov model to characterize shared ancestry in organisms such as the malaria parasite Plasmodium falciparum.

Key Features:

• Hidden Markov Model framework: Implements an HMM tailored for analyzing haploid genomes, including Plasmodium falciparum.
• Pairwise IBD estimation: Estimates pairwise identity by descent between sampled haploid genomes from one or two populations.
• Cross-population detection: Treats different populations distinctly within the analytical framework to enhance detection accuracy and support identification of imported cases.
• Performance verification and benchmarking: Performance verified using simulated data and benchmarked against existing IBD methods, demonstrating superior sensitivity and specificity.
• Efficiency and scalability: Reports an average runtime of ~70 seconds to analyze 50 whole genome sequences on a standard laptop and scales linearly with the number of pairwise comparisons.
• Robustness to parameter misspecification: Demonstrates robustness to misspecified genotyping error rates and recombination rates, with a recommendation to exclude genomic regions where these rates substantially deviate from the genome-wide average.

Scientific Applications:

• Malaria epidemiology: Detects IBD segments in Plasmodium falciparum to inform population-level epidemiological analyses.
• Transmission and spread tracking: Supports reconstruction of transmission dynamics and tracking of pathogen spread across and within populations.
• Imported case identification and elimination strategies: Aids identification of imported malaria cases and supports targeted elimination interventions based on genetic data.

Methodology:

Uses a hidden Markov model tailored for haploid genomes to estimate pairwise IBD between sampled haploid genomes from one or two populations, with performance assessed using simulated data and benchmarking against existing methods.