Estimating admixture proportions with microsatellites: comparison of methods based on simulated data

Area of Science:

• Population Genetics
• Bioinformatics
• Evolutionary Biology

Background:

• Estimating parental contributions in admixed populations is crucial for understanding genetic dynamics.
• Previous comparisons of these methods have been limited in scope and scale.

Purpose of the Study:

• To systematically compare the performance of six widely-used methods for estimating parental contributions in admixed populations.
• To evaluate method performance under varying conditions using simulated data.
• To provide practical guidelines for selecting appropriate methods.

Main Methods:

• A simulation-based comparison of six prominent methods over the last two decades.
• Five methods utilized allele frequencies with different statistical approaches.
• One method incorporated molecular divergence and coalescence times.

Main Results:

• All methods performed comparably under optimal conditions (highly differentiated parents, recent hybridization).
• Performance varied significantly under non-optimal conditions, with no single method superior across all scenarios.
• Evaluation criteria included applicability, estimation bias, mean square error, and confidence interval accuracy.

Conclusions:

• Method selection for estimating parental contributions depends heavily on population-specific conditions.
• No universally superior method exists; performance is context-dependent.
• The study offers valuable insights for researchers choosing methods for admixed population genetic analyses.