A comparison of methods for estimating substitution rates from ancient DNA sequence data

Area of Science:

• Evolutionary biology
• Genetics
• Bioinformatics

Background:

• Phylogenetic analysis of modern and ancient DNA is crucial for reconstructing evolutionary history.
• Accurate estimation of evolutionary rates, calibrated by sample ages, is essential for these analyses.
• Among-lineage rate variation and non-random sampling can compromise the reliability of evolutionary rate estimates.

Purpose of the Study:

• To compare the performance of three phylogenetic methods for inferring evolutionary rates from time-structured data.
• To evaluate the impact of factors like rate variation and sampling on rate estimation accuracy.
• To apply these methods to empirical time-structured mitogenomic data from six vertebrate species.

Main Methods:

• Simulation study comparing regression of root-to-tip distances, least-squares dating, and Bayesian inference.
• Application of these three methods to time-structured mitogenomic data sets.
• Analysis of 12 simulation scenarios to assess method performance under varying conditions.

Main Results:

• All three methods provide reliable evolutionary rate estimates under conditions of high substitution rates, low rate variation, and minimal phylo-temporal clustering.
• The interplay between these factors significantly influences the accuracy of least-squares dating and Bayesian rate estimation.
• Consistent rate estimates were observed across most empirical data sets, with horse mitogenomic data showing an exception.

Conclusions:

• Phylogenetic studies utilizing ancient DNA sequences should employ multiple inference methods.
• It is recommended to rigorously test for the presence of temporal signal, among-lineage rate variation, and phylo-temporal clustering within the data.
• These practices enhance the robustness and reliability of evolutionary rate estimations from ancient DNA.