search

marschall-lab / panacus

Panacus is a tool for computing statistics for GFA-formatted pangenome graphs
MIT License
73 stars 4 forks source link

Request software updates #15

Closed Xuelei-Dai closed 9 months ago

Xuelei-Dai commented 9 months ago

Hello Daniel，

I found that the version of the panacus is no longer compatible with the results of Mingraph-Cactus (MC), and the final visualized graph is incorrect. I think this problem needs to be completely solved. It is necessary to use the current Mingraph-Cactus pipeline to construct the pan-genome and use the final results for visualization.

Best wishes Xuelei

danydoerr commented 9 months ago

Thanks for reporting! Can you provide a graph where I can debug the issue?

glennhickey commented 9 months ago

I don't have any ideas of mc changes that would affect Panacus. Here's two yeast graphs made with Cactus v2.6.4 (which was used for the v1.1 HPRC graphs) and Cactus v2.7.0 (which is the current release).

If there's a regression, I would think it would show up between these two graphs.

http://public.gi.ucsc.edu/~hickey/debug/panacus/

danydoerr commented 9 months ago

Thanks @glennhickey!

danydoerr commented 9 months ago

Panacus runs smoothly on these graphs (see attached files). @Xuelei-Dai could you describe the specific issue in more detail and provide a graph that does not work with the current release of panacus?

yeast-v2.6.4.histgrowth.all.html.gz yeast-v2.6.4.histgrowth.all.html.log.gz yeast-v2.6.4.histgrowth.all.pdf yeast-v2.6.4.histgrowth.all.tsv.gz yeast-v2.6.4.histgrowth.all.tsv.log.gz yeast-v2.7.0.histgrowth.all.html.gz yeast-v2.7.0.histgrowth.all.html.log.gz yeast-v2.7.0.histgrowth.all.pdf yeast-v2.7.0.histgrowth.all.tsv.gz yeast-v2.7.0.histgrowth.all.tsv.log.gz

Xuelei-Dai commented 9 months ago

Thanks @danydoerr and @glennhickey!

In the current version of Mingraph-Cactus (MC), the gfa file does not have P line, and the new W line was changed. Since I referred to the Pangenome Coverage and Growth Statistics pipeline (convert the 1.1 version of gfa to 1.0 (vg convert -gfW)), recently I directly used the 1.1 version of the gfafile as the input of panacus, it's okay. So I suggest that you can add some instructions to your pipeline, for example, how should the gfa1.1 version be run?

Best, Xuelei

danydoerr commented 9 months ago

That's a good point, @Xuelei-Dai, the current README does not contain an example for the MC graph. I have now added an example for the MC graph.

To be perfectly clear, panacus has always supported GFA 1.1 and therefore also the latest MC graphs. Neither any update is required, nor any bug to be fixed. The example in the README suggests to visualize pangenome growth only for a subset of samples (those corresponding to haplotypes). This preprocessing is not part of panacus and remains the responsibility of the users.