search

adamallo / SimPhy

SimPhy: A comprehensive simulator of gene family evolution
GNU General Public License v2.0
26 stars 1 forks source link

Bug in bounded-subtree output file #2

Open adamallo opened 6 years ago

adamallo commented 6 years ago

Locus-tree leaves coming from simulated species trees are incorrect.

vvoorr commented 6 years ago

Hello, I tried the simphy problem. I followed the mapsl file to try to map internal nodes of the locus tree to the internal nodes of the species tree, but I ended up with problems. Is it the bug that you are dealing with in this issue?

Yours sincerely, Tin

adamallo commented 6 years ago

This bug relates to the bounded-subtree output file (-ON option, see https://github.com/adamallo/SimPhy/wiki/Manual#532-bounded-locus-subtrees-output-file-format) but does not affect the more commonly used mapping files. This problem should be solved in SimPhy's development branch recphyloxml. I am working on improving the mapping of species trees, locus trees, and species trees in that branch, using the new recphyloxml format. The species-tree/locus-tree mapping is finished, but the recphyloxml is not compatible with gene-tree reconciliation just yet. Can you specify the problems you are facing?

vvoorr commented 6 years ago

Hi Diego

I was using the old version of simphy. I forced the program to add the clade names in the newick files using the tag -ol, but than I found that the mapping in the file mapsl cannot help to reconcile the locus tree and species tree.

Using the recphyloxml version of simphy, the mapsl file becomes an xml file, which contains the species tree, the locus tree and the reconciliation. However, the clade name of the species tree in the mapsl file does not correspond to the clade name of the species tree in the s_tree.trees. Here I think we will just need to ignore the clade names provided in the s_tree.trees file and everything will be fine.

By the way, have you used your program to test reconciliation algorithms like Amalgamated likelihood estimation (ALE), ranger or any others? Do you know which one works better?

Yours sincerely, Tin

2018-03-05 17:24 GMT+01:00 Diego Mallo notifications@github.com:

This bug relates to the bounded-subtree output file (-ON option, see https://github.com/adamallo/SimPhy/wiki/Manual#532- bounded-locus-subtrees-output-file-format) but does not affect the more commonly used mapping files. This problem should be solved in SimPhy's development branch recphyloxml https://github.com/adamallo/SimPhy/tree/recphyloxml. I am working on improving the mapping of species trees, locus trees, and species trees in that branch, using the new recphyloxml format http://phylariane.univ-lyon1.fr/recphyloxml/. The species-tree/locus-tree mapping is finished, but the recphyloxml is not compatible with gene-tree reconciliation just yet. Can you specify the problems you are facing?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/adamallo/SimPhy/issues/2#issuecomment-370475417, or mute the thread https://github.com/notifications/unsubscribe-auth/AC8Nb8VoUsnPxFeL0Itv8G-Stye_LcaAks5tbWatgaJpZM4P-vUs .

adamallo commented 6 years ago

Just for clarification, the old version should still allow you to map species and locus trees. However, that format is quite complicated, and a lot of people are having problems understanding and using it. I hope to release a new version of SimPhy implementing recphyloxml soon. In the meantime, feel free to contact me if you need additional help using the former reconciliation format.