HiMWA: A Hierarchical Multiple-wave Admixture Model for Reconstructing Complex Population Admixture Histories.

Area of Science:

• Population genetics
• Evolutionary biology
• Computational genomics

Background:

• Population admixture is crucial for genetic diversity but current methods oversimplify complex histories.
• Existing admixture models often assume sequential contributions, limiting their application to realistic scenarios.

Purpose of the Study:

• To introduce HiMWA, a novel computational framework for reconstructing complex, hierarchical, and multiple-wave admixture histories.
• To provide a flexible tool for disentangling intricate population genetic histories from genomic data.

Main Methods:

• HiMWA employs a hierarchical multiple-wave admixture model.
• Integrates model selection via ancestry switch counts and parameter estimation using ancestral tract lengths.
• Utilizes simulations to assess accuracy and robustness against genetic drift and local ancestry errors.

Main Results:

• HiMWA accurately reconstructs diverse admixture scenarios, even with errors.
• Kazakhs and Uyghurs exhibit a shared hierarchical admixture structure.
• Demonstrated complex admixture pathways involving intermediate West and East Eurasian populations in Central Asia.

Conclusions:

• Hierarchical multiple-wave admixture is prevalent in Central Asia, shaping complex demographic histories.
• HiMWA offers a powerful and flexible approach for realistic population genetic history reconstruction.
• The HiMWA software is publicly available for broader research application.