Instructions for downloading raw IMGT files:

• Go to the genedb page and click the first submit button
• Next, scroll down to the end of resulting page (loading can take a while) and mark Select all genes
• Select F+ORF+in-frame P nucleotide sequences with IMGT gaps format and click submit
• Copy-paste resulting FASTA records and use them as an input file

Get the compiled binaries and run the software as java -jar segmentparser.jar [options] imgt_raw_file output_prefix.

The following options can be selected:

• -n include non-functional segments into output (pseudogenes, etc)
• -m include minor alleles (segments with *02, *03, etc suffix)
• -s toggle species detalisation (e.g. BALBc and C57Bl6 for MusMusculus)
• -b report IMGT records that cannot be parsed properly (missing conserved residues, etc)

Output files include:

• A $output_prefix$.metadata.txt file with summary statistics.
• Files with erroneous/bad records: $output_prefix$.nojrefpoint.txt, $output_prefix$.novrefpoint.txt, $output_prefix$.othersegm.txt.
• Output file containing sequences, CDR3 reference points and CDR1,2,2.5 coordinates: $output_prefix$.txt.

SegmentParser generates a tab-delimited table with species name, gene and segment id, nucleotide sequence and the reference point position: 0-based coordinate of first nucleotide after conserved Cys for Variable segments and before first nucleotide before conserved Phe/Trp for Joining segments. The metadata table provided with results lists all species and genes and tells if there are any V/D/J segments associated with them (0 or 1 in corresponding row).

• A file with CDR1,2,2.5 nucleotide and amino acid sequences: $output_prefix$.txt (only includes V segments).

Note that CDR2.5 is a putative MHC-binding region of TCR V segment, defined in a recent work of Paul Thomas lab (Dash et al. Nature 2017).