search

qmarcou / IGoR

IGoR is a C++ software designed to infer V(D)J recombination related processes from sequencing data. Find full documentation at:
https://qmarcou.github.io/IGoR/
GNU General Public License v3.0
47 stars 25 forks source link

IGoR crashes when reading mouse V CDR3 anchors #20

Closed ingewortel closed 6 years ago

ingewortel commented 6 years ago

Hi @qmarcou!

I am trying to run IGoR with the standard mouse beta chain model but have some problems with the alignment:

igor -set_wd . -batch foo -read_seqs myseqs.fasta -species mouse -chain beta -align --all

yields the following output:

Batch name set to: foo_ FASTA extension detected for the input sequence file Species parameter set to: mouse Chain parameter set to: beta Working directory set to: "./" [IGoR] ERROR: Exception caught while reading V CDR3 anchors. [IGoR] ERROR: stoi [IGoR] ERROR: Use "man igor", "igor -help" or visit https://bitbucket.org/qmarcou/igor to see available commands and their effects. [IGoR] ERROR: Please report any bug to: quentin.marcou@lpt.ens.fr [IGoR] ERROR: Terminating IGoR...

Note that I don't have this problem when I just read the sequences with igor -set_wd . -batch foo -read_seqs myseqs.fasta or try the alignment on the human model with igor -set_wd . -batch foo -read_seqs myseqs.fasta -species human -chain beta -align --all But I do get the same error about V CDR3 anchors even when I try to align only the J genes: igor -set_wd . -batch foo -read_seqs myseqs.fasta -species mouse -chain beta -align --J

So the issue seems to be in the mouse model itself somewhere and probably has something to do with the _V_gene_CDR3anchors.csv file.

I looked into the _V_gene_CDR3anchors.csv file provided, and noticed that some of the TRBV genes that occur in the genomicVs.fasta file and the _modelparms.txt file are missing from the anchor file (for example, TRBV11*01). Could this be why IGoR crashes? Or am I doing something wrong?

Thanks!

Inge

qmarcou commented 6 years ago

Hi @ingewortel , Sorry about that this is a silly mistake on my side, the genomic anchors file should be semicolon separated and not coma separated. You can hardfix this locally by replacing comas by semicolons in the gene anchors file of your interest (either in the directory IGoR models have been installed into, or before installation) This is something I should have fixed earlier, I will try and release a new version of IGoR containing this fix along other bug fixes in the next few days. Again my apologies for this silly mistake...

ingewortel commented 6 years ago

Hi @qmarcou

Thanks for the quick reply, the semicolons indeed fixed the problem!

Best wishes, Inge

qmarcou commented 6 years ago

Hi @ingewortel , I have uploaded IGoRv1.3.0 release in which your problem has been addressed, along many others and new features were added. Check it out! I'll thus close this issue since the problem is solved. Best wished