Estimating individual admixture proportions from next generation sequencing data

Area of Science:

• Genetics
• Bioinformatics
• Population Genetics

Background:

• Accurate inference of population structure and individual ancestry is crucial for population genetics and association studies.
• Next-generation sequencing (NGS) provides comprehensive genetic variation data but introduces uncertainty in genotype inference, especially at low depths.
• Existing admixture analysis methods require known genotypes, which are unreliable when inferred from low-depth NGS data.

Purpose of the Study:

• To develop a novel method for inferring individual ancestry that explicitly accounts for the uncertainty in genotype likelihoods from low-depth NGS data.
• To demonstrate the accuracy and robustness of the new method compared to existing approaches.
• To provide a software implementation for practical application in population genetic analyses.

Main Methods:

• Developed a new statistical method that directly utilizes genotype likelihoods, preserving information from unobserved genotypes.
• Employed simulations and analysis of publicly available low-depth sequencing data to evaluate method performance.
• Compared the proposed method against traditional approaches that rely on pre-called genotypes.

Main Results:

• The novel method accurately infers individual ancestry even with very low-depth sequencing data.
• Existing methods, when applied to genotypes called from low-depth data, introduce significant biases.
• The proposed method demonstrates superior accuracy and reduced bias in ancestry inference from NGS data.

Conclusions:

• The developed method effectively addresses the challenge of genotype uncertainty in low-depth NGS data for ancestry inference.
• This approach offers a more reliable alternative to existing methods for population genetics and association studies utilizing NGS data.
• The NGSadmix software provides a valuable tool for researchers working with large-scale genetic sequencing data.