Locus Tags vs Gene_ids about ppanggolin HOT 4 CLOSED

Hi !

If i'm not mistaken, the /gene from genbank files often corresponds to gene names, and it is often not unique within genbank (genes in different genomes can have the same gene name). it is parsed and saved in the .h5 file though, but not reported in many of the files.

In which files or figures are you looking for those gene identifiers ?

Adelme

from ppanggolin.

I was looking at the families and the presence absence matrix. It would be much better if the gene names is placed rather than the locus tags.

from ppanggolin.

I understand. Indeed it would be much more practical to know what we are manipulating directly, as gene names are more explicit than locus tags.
It is not currently doable with the current version of ppanggolin, but it is definitely possible to code it. I will ask the other members of the project what they think about it and see if I can code that at some point in the near futur.

In the mean time I'm afraid there are no easy way to do it with ppanggolin currently, sorry.

from ppanggolin.

After discussing with other members of the project, there's actually a file that can help you, the 'matrix.csv' file which indicates, among many other information, the gene family id and the most common gene name in the cluster.

I do agree that have gene names directly would make it much more practical to know what you are looking at.

Using the 'matrix.csv' is not the same as naming clusters with the gene names, but for many reasons we will not do it, the first one being that gene names are sadly not unique.

While it will not be a ppanggolin default behavior, I guess that a script that generates the presence_absence matrix with informations from the matrix.csv file could be a nice add-on at some point in the futur, but for now this won't be done.

from ppanggolin.