[Feature Proposal] Negative constraints

[HS6986]

Dear Professor Smith @ms609,

I'm trying reconstructing deep phylogeny of bilaterians, where the mutual relationships among the four major lineages Ambulacraria, Chordata, Xenacoelomorpha, and Protostomia are unclear, by analyzing my modified version of a morphological dataset of Juravel et al. (2023) with thorough positive constraints for the incontestable clades.

Maximum likelihood and parsimony analyses without corrections for inapplicable data recovered the typical morphology-based topology (Xenacoelomorpha, ((Ambulacraira, Chordata), Protostomia)) with almost maximal resampling support values. However, BGS parsimony recovered a very radical, idiosyncratic, unprecedented topology with almost maximal jackknife support values.

I'm planning to include this topology as one of the results in our paper, but I think that it would be beneficial to see the topology obtained from BGS parsimony with negative constraints for clades that have never been strongly supported in phylogenomics or phylogenetic analyses based on genomic information other than molecules.

The function for negative constraints is implemented, for example, in TNT and in MrBayes, and it should be useful in cases like this. I would deeply appreciate it if you could consider implementing this feature in TreeSearch, if the implementation was not so hard or time-consuming.

Thank you very much for your time and support.

[ms609]

Thanks for the suggestion; I'll think about whether there's an efficient way to implement this.

[ms609]

Sorry not to have got around to this yet – it's on the radar, but my to-do list is still growing faster than I can shrink it!

[HS6986]

Dear @ms609,

Thank you for paying attention to my proposal.

After posting it, I've surveyed the literature more thoroughly and found that some (but few) phylogenomic analyses had already supported topologies similar to the one in our study. Thus, I no longer consider our topology to be so peculiar.

Consequently, I'm not planning to additionally set negative constraints in the TreeSearch analysis of this study now, so you don't need to rush the implementation of them in TreeSearch. Thank you for your consideration. However, this function may sometimes be useful in other cases and should be indispensable for implementing the Bremer support, I think. It might therefore be worth implementing in the future. Of course, I'll leave the decision up to you.

Thank you very much for your time and support.