Are mouse gene lists available?

[wyattmcdonnell]

Hi Mikhail and @antigenomics team!

Have really enjoyed using VDJTools so far for human data. Had to do a vanilla install of R 3.3.0 to get it working on my Mac but otherwise no issues until I tried to import IMGT alignments of mouse TRB sequences. The overwhelming majority of reads were rejected as bad lines (e.g. 38/18276 lines were retained after conversion from IMGT) -- is this because the mouse gene lists are currently unsupported in VDJTools? Thanks a ton for your help.

[mikessh]

Hello!

I don't think this is mouse-specific as VDJtools is neutral to species and receptor chain.

Can you please post a few good and bad examples here? It is better to attach them as file because IMGT table has lots of columns. Also, VDJtools normally reports the reason behind a line from clonotypes table is not parsed, e.g. bad_cdr3. Can you please post those.

[wyattmcdonnell]

Hi there!

I have a sneaking suspicion that it's not a VDJTools issue either; part of this is because we'd been uploading immunoSEQ to IMGT (we've found alignment issues in the past, and the immunoSEQ v3 parsers in both tcR and VDJTools don't seem to be working mostly because of header problems and Adaptive throwing metadata columns in randomly) and then importing into VDJTools. Apologies for the delay, the Thanksgiving holiday was being celebrated this past week since Wednesday!

Mouse files (not working) as follows, in order of IMGT file, VDJTools converted format, and terminal output: 3_Nt-sequences.txt VDJ_.3_Nt-sequences.txt vdjerror.txt

Working human examples as follows: good3_Nt-sequences.txt goodVDJ_.3_Nt-sequences.txt

[wyattmcdonnell]

Hi there! Just wondering if your team has had a chance to look over some of these? Is this an issue that would be resolved by changing "unknown" to "productive" in the 3_Nt-... files from IMGT?