Dcifer

Dcifer calculates genetic distances between polyclonal infections using an identity by descent (IBD) framework to quantify recent shared ancestry in recombining organisms such as malaria parasites.

Key Features:

• IBD-Based Framework: Employs an identity by descent (IBD) approach tailored for polyclonal infections and applicable to multiallelic data (microsatellites, microhaplotypes) and biallelic single nucleotide polymorphism (SNP) data.
• Unphased Data Compatibility: Operates on unphased genomic data without requiring prior phasing of haplotypes.
• Population Allele Frequencies: Explicitly incorporates population allele frequencies into genetic distance calculations.
• Combinatorial Approach: Uses combinatorics to account for unobserved phased haplotypes, enabling efficient processing of large datasets and providing pairwise relatedness estimates with measures of uncertainty.
• Statistical Power: Demonstrates superior statistical power for detecting related infections, outperforming identity by state (IBS)-based approaches by approximately 2–4×.
• Model Parameters for Interhost Relatedness: Defines model parameters representing interhost relatedness and proposes estimators with favorable statistical properties.

Scientific Applications:

• Transmission Reconstruction: Reconstructs pathogen transmission networks by quantifying recent shared ancestry between infections.
• Epidemiological Inference: Addresses epidemiologically relevant questions about relatedness and transmission dynamics among infections.
• Spatial Dynamics: Aligns relatedness structures with geographic locations to study the spatial dynamics of infectious diseases.

Methodology:

Defines interhost relatedness model parameters, applies an IBD framework, uses combinatorial techniques to handle unphased haplotypes, and incorporates population allele frequencies into calculations.