Evaluating the use of Monte Carlo simulation for statistically assessing topological congruence of phylogenetic trees

Many questions in evolutionary biology involve assessing the topological congruence between phylogenetic trees. However, quantification of tree congruence provides a single measurement that is often difficult to contextualize and use to make inferences. One approach to statistically inform measurements of tree congruence is to generate null congruence distributions using Monte Carlo simulations, which can then be used for hypothesis testing. Although this approach has previously been used in empirical studies, its validity and effectiveness have yet to be tested. Here, I evaluate this method in combination with a variety of commonly used congruence metrics. I first explore how the number of simulations conducted influences the stability of null model generation. I then explore the error and power associated with different congruence metrics and tree sizes. These findings reveal nuances to this approach that are important for researchers to consider when using it to make evolutionary inferences. Finally, I introduce manticore , an R library that provides a simple interface for researchers to utilize this approach.