Clusters proteins at each taxonomic rank - OMA

Question

[anusblat]

Hello everybody,

Anyone knows how to clusters proteins ( for example 90% of sequence identity) at each taxonomic rank (for example phylum) from an orthoXML file downloaded from OMA or other files. This will create non-redundant proteins in all taxomomic rank, but will avoid to loose a phylum that contain sequences.

Thanks! Alejandro

Answer 1

[Adrian Altenhoff]

Hi Alejandro,

can you elaborate what you're trying to do? I'm one of the OMA developers, but I have a hard time to understand what you're doing.

In brief, OMA HOGs are groups of genes that all started diverging from a certain ancestral gene through speciation, so that gives you already gene clusters. If you want to subcluster them according to sequence identity, you would have to do that based on the available protein sequences. you could use pyham (https://github.com/DessimozLab/pyham) to extract the extent genes of any cluster you're interested.