How robust are "isolation with migration" analyses to violations of the im model? A simulation study

Area of Science:

• Population Genetics
• Molecular Evolution
• Bioinformatics

Background:

• Advanced methods enable precise estimation of molecular demographic parameters like population size and gene flow.
• These methods rely on simplifying assumptions about species history and genetic data, raising questions about their robustness.

Purpose of the Study:

• To assess the impact of violations to the Isolation with Migration (IM) model on demographic parameter estimates.
• To evaluate the robustness of the IMA program to various deviations from model assumptions using simulated data.

Main Methods:

• Simulated genetic datasets were used to test the effects of intralocus recombination, population structure, unsampled gene flow, linkage, and divergent selection on IM model assumptions.
• The influence of nucleotide substitution model misspecification on parameter estimates was also examined.

Main Results:

• IMA demographic parameter estimates demonstrated considerable robustness to minor to moderate violations of IM model assumptions.
• Population structure within species and significant recombination (when data were analyzed in non-recombining blocks) had minimal impact on most estimates.
• However, using an incorrect nucleotide substitution model led to increased error rates, including predictable bias and higher variance.

Conclusions:

• Molecular demographic inference methods, like those in IMA, are generally reliable for real-world data with common assumption violations.
• Careful consideration of nucleotide substitution models is crucial to avoid significant biases and errors in demographic parameter estimation.