Can 'de_novo_wf' be placed by 'classify_wf'

[Zhichao Zhou]

Since that gtdb bacteria has 10s of thousands of MAGs, it will take a very long time to get the phylogeny even by FastTree. I am wondering can 'classify_wf' be used? Based on the description of classify_wf:

Finally, the classify step uses pplacer to find the maximum-likelihood placement of each genome in the GTDB-Tk reference tree. GTDB-Tk classifies each genome based on its placement in the reference tree, its relative evolutionary divergence, and/or average nucleotide identity (ANI) to reference genomes.

I think maybe it also provides a good phylogeny too

[Simon Roux]

Hi,

Unfortunately, de_novo_wf can not be replaced by classify_wf. This is because (as far as I’ve seen) classify_wf only provides trees with a single MAG, while iPHoP needs a tree including all references + additional MAGs, as we need to be able to get all-vs-all phylogenetic distances. If you find ways to get this kind of tree with classify_wf, that is quicker than de_novo_wf, please let me know !

Best,

Simon