search

COMBINE-lab / cuttlefish

Building the compacted de Bruijn graph efficiently from references or reads.
BSD 3-Clause "New" or "Revised" License
81 stars 9 forks source link

gfa visualization #2

Closed EA2106-Universite-Francois-Rabelais closed 3 years ago

EA2106-Universite-Francois-Rabelais commented 3 years ago

Hi! I successfully ran cuttlefish on 3 plant genomes (indeed in a very fast way). I would like to visualize the resulting pan-genome (could it be really named like this in this case?) from the gfa file (either gfa1 or gfa2). I tried SGTK (https://github.com/olga24912/SGTK) but without success. I will also try other https://github.com/rrwick/Bandage and https://github.com/ggonnella/gfaviz, but I wondered if you already have experience in this kind of task before going through it. Thanks and congrats for this impressive work.

jamshed commented 3 years ago

Hi!

Glad that you found Cuttlefish useful! We haven't yet tried visualizing the GFA outputs, but we intend to. We are a bit busy with the final draft of the paper right now, but we will get back to you shortly, hopefully within a few days. Let us know if you could sort out the stuff in the mean time, and / or need any other assistance using Cuttlefish!

Thanks!

EA2106-Universite-Francois-Rabelais commented 3 years ago

Hi Jamshed, ok I will perform several tests and keep you updated! Good luck with the writing.

jamshed commented 3 years ago

Hi @EA2106-Universite-Francois-Rabelais,

I've been able to visualize the GFA1 output using Bandage!

EA2106-Universite-Francois-Rabelais commented 3 years ago

Dear @jamshed-k thanks for the update! I am working with quite a large file and I have to download it from my computing grid. I will give a try!

EA2106-Universite-Francois-Rabelais commented 3 years ago

I indeed tried bandage with a toy example adapted from my own datasets (with some variations added manually). Worked perfectly with gfa1 format but not gfa2 (seems to be unable to detect edges with this version). Otherwise, it has nice features to extract sequences from nodes.

graph

jamshed commented 3 years ago

Hi @EA2106-Universite-Francois-Rabelais,

Indeed, the visualizations are good for small enough graphs. For example, the following two are for the compacted de Bruijn graph of the E. Coli genome:

k = 101: ecoli-k101

k = 255: ecoli-k255

I'm closing the issue for now. Please let us know if you do experiment more and find other suitable alternatives!