iAdmix

iAdmix estimates individual admixture proportions from exome, targeted, low-coverage whole-genome sequencing, and genotyping data to infer genetic ancestry for applications such as disease-association studies and population-history analyses.

Key Features:

• Computationally Efficient Ancestry Inference: iAdmix uses allele frequencies from known reference populations and performs maximum likelihood estimation with the BFGS optimization algorithm to estimate the relative contribution of reference populations.
• Utilization of Genotype Likelihoods: iAdmix incorporates genotype likelihoods to account for genotype uncertainty in sequence data and eliminates the need for external reference genotype panels.
• Versatility Across Data Types: iAdmix handles exome sequencing, targeted sequencing, low-coverage whole-genome sequencing, and genotyping data and produces consistent genome-wide average ancestry estimates validated using 1000 Genomes data.
• Application to Pooled Sequence Data: iAdmix can estimate admixture proportions from pooled sequencing data to support control of population stratification in sequencing-based association studies that utilize DNA pooling.

Scientific Applications:

• Disease Association Studies: Estimating individual ancestry to control for population stratification in genetic association analyses.
• Population Genetics Research: Analyzing admixture proportions to investigate human population history and migration patterns.
• Personal Genomics: Providing ancestry estimates to aid interpretation of personal genomic variation.

Methodology:

iAdmix uses allele frequencies from reference populations with individual genotype or sequence data, incorporates genotype likelihoods to model genotype uncertainty, and derives maximum likelihood estimates of global admixture proportions using the BFGS optimization algorithm.