<tool id="ANARCI" name="ANARCI" version="1.0">
  <description>Use ANARCUI tools to use HADDOCK-antibody-antigen after.</description>
  <command detect_errors="exit_code"><![CDATA[
    conda activate Ab-HADDOCK;
    python2.7 ImmunoPDB.py -i '${input}' -o '${output}' --scheme ${scheme} --splitscfv
    #if $receptor
      --receptor ${receptor}
    #end if;
    #if $fvonly
      --fvonly
    #end if;
    #if $rename
      --rename
    #end if;
    #if $splitscfv
      --splitscfv
    #end if;
    conda deactivate
  ]]></command>
  <inputs>
    <param format="pdb" name="input" type="data" label="Select your PDB file."/>
    <param name="scheme" type="select" label="Scheme to use - default is 'IMGT'" help="Which numbering scheme should be used.">
      <option value="imgt" selected="true">IMGT</option>
      <option value="kabat">Kabat</option>
      <option value="aho">Aho</option>
      <option value="wolfguy">Wolfguy</option>
      <option value="chothia">Chothia</option>
      <option value="martin">Martin</option>
      <option value="pdb">PDB</option>
    </param>
    <param name="receptor" type="select" optional="true" label="Choose ig or tr domains" help="Choose whether to number Antibody (ig) or TCR (tr) domains. (Optional)">
      <option value="ig" >Antibody domains</option>
      <option value="tr" >TCR domains</option>
    </param>
    <param name="fvonly" type="boolean" optional="true" label="Only output Fv regions." help="Only output Fv regions."/>
    <param name="rename" type="boolean" optional="true" label="Rename the receptor chains." help="Rename the receptor chains with H and L (ig) or B and A (tr). Only receptor chains output. First pair identified used."/>
    <param name="splitscfv" type="boolean" optional="true" label="Split chain." help="When they are found split single chain fvs into two seperate chains (fvonly becomes true)."/>
  </inputs>
  <outputs>
    <data name="output" format="pdb" />
  </outputs>
  <help>
usage: ImmunoPDB [-h] [-i INPUTSTRUCTURE] [-o OUTFILE]
                 [--scheme {kabat,aho,wolfguy,imgt,a,c,chothia,i,k,m,w,martin,pdb}]
                 [--receptor {ig,tr}] [--rename] [--fvonly] [--splitscfv]
                 [--warnings]

ANARCI - ImmunoPDB                                     \\    //
Antibody Numbering and Antigen Receptor ClassIfication  \\  //
                                                          ||
(c) Oxford Protein Informatics Group (OPIG). 2015-16      ||

Example script to number Antibody and TCR PDB structures.

ANARCI must be installed and in the path
opig.stats.ox.ac.uk/webapps/anarci

Requirements: Biopython (version >= 1.66)
              Muscle

This script extends the BioPython PDBParser and Structure classes so that a numbering scheme can be applied to the variable domain
of an antigen receptor chain. 

o Where available the Seqres record is used as the full sequence. Missing residues are recognised by comparing this to the residues
  with coordinates.

o *Only* variable domains are numbered consistently in the chosen scheme.

o Residues before the domain are numbered '0' with reverse alphabetical insertions if there are less than 28 (all 0 otherwise - this
  will break some PDB parsers...)

o Residues after the variable domain are numbered sequentially from 1001.

o By default when more than one variable domain is found on a single chain (e.g. single chain Fv, diabody...) the numbering will be
  with respect to the first domain identified.

o CDR recognition is performed and regions are annotated in the xtra dictionary attributes of residue objects.

o Pairing is performed using the distance between the interface cysteine positions (imgt 104).

Basic useage

Renumber antibody chains with imgt numbering scheme
python ImmunoPDB.py -i infile.pdb -o outfile.pdb -s imgt

Renumber tcr chains with imgt numbering scheme
python ImmunoPDB.py -i infile.pdb -o outfile.pdb -s imgt --receptor tr

optional arguments:
  -h, --help            show this help message and exit
  -i INPUTSTRUCTURE     A structure to be numbered
  -o OUTFILE            The output file to use. Default is stdout
  --scheme {kabat,aho,wolfguy,imgt,a,c,chothia,i,k,m,w,martin,pdb}, -s {kabat,aho,wolfguy,imgt,a,c,chothia,i,k,m,w,martin,pdb}
                        Which numbering scheme should be used. i, k, c, m, w
                        and a are shorthand for IMGT, Kabat, Chothia, Martin
                        (Extended Chothia), Wolfguy and Aho respectively.
                        Default IMGT. Use pdb to retain the numbering but get
                        the annotations as remarks
  --receptor {ig,tr}, -r {ig,tr}
                        Choose whether to number Antibody (ig) or TCR (tr)
                        domains.
  --rename              Rename the receptor chains with H and L (ig) or B and
                        A (tr). Only receptor chains output. First pair
                        identified used.
  --fvonly              Only output Fv regions.
  --splitscfv           When they are found split single chain fvs into two
                        seperate chains (fvonly becomes true)
  --warnings            Report warnings about missing residues

Author: James Dunbar (dunbar@stats.ox.ac.uk)
        Charlotte Deane (deane@stats.ox.ac.uk)

Copyright (C) 2016 Oxford Protein Informatics Group (OPIG)
Freely distributed under the GNU General Public License (GPLv3).
  </help>
</tool>