Manual Site Rates

[Paul Maier]

I hope this finds you well. I have a quick question about IQtree's implementation. Is there currently a way to manually specify site rates, and force the program to use those?

The reasoning. Often, in cases of phylogenetic conflict, I've observed that the program seems to down-weight stable mutations, and up-weight unstable mutations (i.e., those occurring more frequently). This causes topologies to form that are not very parsimonious. In some cases (but not all), this may be due to only including a small subset of the entire tree, hence, the program cannot measure the "global" mutation rate for a particular site. But I've observed this when including all sequences as well.

For my specific application in human mtDNA, it would be easier for me to force a set of site-specific rates, e.g. using --mlrate. Then, maybe there is an option (e.g., -fs) to force those rates in the model? I am using a partitioned analysis, with three partitions: SNPs, INDELs, and STRs.

Thank you!