NGSremix

NGSremix estimates pairwise genetic relatedness between admixed individuals from low-depth next-generation sequencing (NGS) data by integrating genotype likelihoods and population admixture within a maximum likelihood framework.

Key Features:

• Maximum Likelihood Estimation: Employs a maximum likelihood framework that integrates genotype likelihoods to estimate relatedness coefficients.
• Handling Admixture: Explicitly models population admixture when estimating pairwise relatedness.
• Data Input Formats: Accepts called genotypes in PLINK format and genotype likelihoods in BEAGLE format.
• Performance Superiority: Demonstrates superior performance versus PLINK, KING, relateAdmix, and ngsRelate on simulated and empirical low-depth NGS datasets (average depth ~4x or below).
• Implementation: Implemented as a multithreaded C/C++ program with an R package wrapper that embeds the C++ code.

Scientific Applications:

• Population Genetics: Estimating pairwise relatedness in admixed and diverse populations for population-structure and demographic analyses.
• Evolutionary Biology: Assessing relatedness patterns relevant to evolutionary and comparative studies.
• Complex Trait Genetics: Accounting for relatedness in analyses of complex trait inheritance and association studies.

Methodology:

Uses maximum likelihood estimation incorporating genotype likelihoods and explicit admixture modeling; accepts PLINK and BEAGLE input formats; evaluated on simulated and real low-depth NGS data (~4x); implemented in multithreaded C/C++ with an R package wrapper.