Maximum-likelihood inference of population size contractions from microsatellite data

Area of Science:

• Evolutionary biology
• Molecular ecology
• Population genetics

Background:

• Understanding demographic history is crucial for evolutionary biology and molecular ecology.
• Microsatellite data offers insights into past population dynamics.
• Existing methods for inferring population size changes have limitations.

Purpose of the Study:

• To develop a novel maximum-likelihood method for inferring past population size changes from microsatellite data.
• To extend existing methods with new mutation models, specifically the generalized stepwise mutation model (GSM).
• To evaluate the performance, accuracy, and robustness of the developed method.

Main Methods:

• Maximum-likelihood inference
• Importance sampling of gene genealogies
• Generalized stepwise mutation model (GSM)
• Simulation studies for performance testing

Main Results:

• The developed method demonstrates competitive performance compared to existing tools like MSVAR.
• Incorporating the GSM improves the accuracy of microsatellite data analysis, reducing false positive detection of population contractions.
• The method shows robustness to misspecification of mutation models but can be influenced by unaccounted population structure.

Conclusions:

• The new maximum-likelihood method provides a powerful and accurate tool for reconstructing demographic histories from microsatellite data.
• The inclusion of the GSM is critical for reliable analysis, mitigating biases caused by violations of simpler mutation models.
• While computationally intensive in some scenarios and sensitive to population structure, the method is a valuable addition to population genetics research, available in the MIGRAINE software.