search

biobakery / phylophlan

Precise phylogenetic analysis of microbial isolates and genomes from metagenomes
https://huttenhower.sph.harvard.edu/phylophlan
MIT License
128 stars 33 forks source link

The topology of the output trees #57

Closed MostafaYA closed 3 years ago

MostafaYA commented 3 years ago

Hi,

I have used Phylophlan on a subset of genomes. The output included a number of tree files. I am wondering what are the differences between the files *_resolved.tre and RAxML_bestTree.*_refined.tre? and which one should be the final? as in my case, the topology of both trees was different.

The topology of the faa_resolved.tre is similar to the core nucleotide analysis and also in agreement with the literature.

I am using Phylophlan version 0.43 installed using conda.

These are the two trees ...

Präsentation1

Thank you very much for your help,

fasnicar commented 3 years ago

Hi,

Please have a look at the Output section of the wiki that explains exactly that.

Briefly, in your case, I believe your config file contains both the [tree1] and [tree2] sections, where [tree1] specifies to use FastTree for building an initial phylogeny and [tree2] specifies RAxML for refining the phylogeny reconstructed with FastTree. In this case, as also detailed in the wiki, the final tree is the RAxML_bestTree.*.tre.

About topology, consider that those trees are not rerooted w.r.t. anything, so you might need to reroot them before evaluating them properly.

The output folder contains both phylogenies and the MSA as those can be used for other downstream analyses.

Thanks, Francesco

MostafaYA commented 3 years ago

Hi Francesco, thank you very much for your declaration.