and my perspective piece on this: www.nature.com/articles/s41...

We suggest this is inherent to the very distinct nature of these datatypes and the way they are collected, and we provide a conceptual framework from what can be concluded by evolutionary rates studies using each datatype.

Lack of comparability across existing methods makes definite conclusions more limited than what we wanted, but the biggest difference in our comparisons is always across datatypes.

We included discrete (phylogenetic characters) and continuous data (from 2D GMM) from the exact same species and, in most cases, from the exact same specimens. That was a lot of work…

We tried to assess whether rates of phenotypic evolution based on multivariate data (discrete and continuous), using popular methods, and which influence how we measure phenotypic change and innovation across lineages provide comparable results. The answer is a very big NO.

Don't forget to cite it if you use it!;)
Simões, T. R., Greifer, N., Barido-Sottani, J., & Pierce, S. E. (2023). EvoPhylo: an R package for pre- and postprocessing of morphological data from relaxed clock Bayesian phylogenetics. Methods in Ecology and Evolution. doi.org/10.1111/2041...

New stuff: combine_log now also combines posterior log files from BEAST2 and MCMCTREE; write_partitioned_alignments2 writes partitioned data as separate alignments for any data type; clade_membership designates clade membership for each tip for in the tree used by other functions

Post-processing Mr.Bayes and BEAST2 outputs: mode of selection upon morphological characters based on their rates of evolution + other stas and plots for phenotypic rates; Summary stats and plots for rates of diversification and fossilization from FBD analyses

EvoPhylo pre-processing: detects morphological clusters, and exporting datasets already formatted for partitioned phylo analyses.

That's why his books are mandatory reading for my students :)