Python wrapper around SymD executable and xml tool description for running behind galaxy

symd_tool.xml

<tool id="symd_tool" name="Perform SymD calculation on a PDB file">
  <description>Use this tool to run symd on a PDB file</description>
  <command interpreter="python">symd_wrapper.py $input $trfm_file $info_file ${input.name} </command>
  <inputs>
    <param format="txt" name="input" type="data" label="PDB file"/>
  </inputs>
  <outputs>
    <data format="txt" name="trfm_file" />
    <data format="txt" name="info_file" />
  </outputs>
  <help>
This tool simply runs SymD on the input file.
  </help>
</tool>

symd_wrapper.py

import os
import subprocess
import sys
import tempfile
import shutil


# This is the name from galaxy for the input .pdb file                                                                                                                                      
# Unfortunately, it has a name that ends in .dat                                                                                                                                            
fname = sys.argv[1]
# These are the names of the datafiles supplied by galaxy                                                                                                                                   
# for storing the output.                                                                                                                                                                   
outfname1 = sys.argv[2]
outfname2 = sys.argv[3]

# this is the original filename                                                                                                                                                             
origfname = sys.argv[4]

# We simply symlink the galaxy system filename to the original                                                                                                                              
# filename to get SymD the protein name and the .pdb extension                                                                                                                              
os.symlink(fname,origfname)

# These are the names of the output files as created by                                                                                                                                     
# SymD.  We will need to copy these files back to                                                                                                                                           
# outfname1 and outfname2 above.                                                                                                                                                            
symdinfofname = os.path.splitext(origfname)[0] + "-info.txt"
symdpdbfname = os.path.splitext(origfname)[0] + "-trfm.pdb"

# Call SymD with the symlinked file ending with .pdb                                                                                                                                        
symdcmd = "SymD %s" % origfname
res = subprocess.call(symdcmd,stderr=subprocess.PIPE,stdout=subprocess.PIPE,shell=True)
if(res<>0):
    print(stderr)
else:
    # Finally, copy the output files from SymD back to the data files.                                                                                                                      
    shutil.copy(symdinfofname,outfname1)
    shutil.copy(symdpdbfname,outfname2)

To use these two files:

1. Make sure that SymD is in the PATH
2. make a directory in tools called "local_tools"
3. Place both of these files inside "local_tools"
4. Edit the tool_data.xml to include a section like:

    <section name="Local Tools" id="local_tools">
      <tool file="local_tools/symd_tool.xml" />
    </section>

Unfortunately ${input.name} is a job name and ONLY in case of file upload happens to be the original file name. If you feed an output from ANOTHER tool as input for this tool - it will be something like "<TOOL-NAME on data XX".

So this script only works on uploaded data.