Merge branch 'targetp' into 'develop'

Targetp

See merge request SOaAS/psot!1

README.md

@@ -18,6 +18,7 @@ Supported bioinformatic tools:
 * blastp
 * signalp
 * ghostx
+* targetp
 
 Install nextflow:
 

config.yaml

 tools:
     signalp: '/vol/biotools/bin/signalp'
     ghostx: '/vol/biotools/bin/ghostx'
+    targetp: '/vol/software/bin/targetp'
     tmhmm: '/vol/biotools/bin/tmhmm'

modules/targetp.yaml

+# Module manifest for the targetp analysis 
+
+# The name of the module. Is needed for the list-analyses option, for custom
+# configurations and custom profiles
+name: 'targetp'
+
+# Short description of the analysis
+info: 'predict subcellular peptide location'
+
+# The configuration of the script for the analysis step. 
+analysis:
+    # script must take a --fasta parameter
+    script: 'run_targetp.py'
+    # specify additional default configuration here
+    parameters:
+
+# The configuration of the script for the json conversion step. 
+converter:
+    # script must take a --result parameter, which is the result from the analysis step
+    script: 'convert_targetp.py'
+    # specify additional default configuration here
+    parameters:

profiles/complete.yaml

@@ -5,3 +5,5 @@ modules:
   blastp_swissprot:
       alignment:
   hmmer_pfam_a:
+  targetp:
+    organism_group: 'non-plant'

profiles/debug.yaml

@@ -6,3 +6,5 @@ modules:
   blastp_swissprot:
   ghostx_swissprot:
   hmmer_pfam_a:
+  targetp:
+    organism_group: 'non-plant'

scripts/convert_targetp.py

+#!/usr/bin/python3
+import sys
+import json
+import argparse
+
+parser = argparse.ArgumentParser(description = 'Convert targetp results to json documents')
+parser.add_argument('--result', '-r', required = True, help = 'The targetp result file')
+parser.add_argument('--output', '-o', required = True, help = 'The converted results json file')
+args = parser.parse_args()
+
+loc_dict = {"C": "chloroplast", "M": "mitochondrion", "S": "secretory pathway", "_": "other", "*": "undetermined"}
+
+filename = args.result
+documents = {}
+with open(filename) as f:
+    tool = None
+    field_index = {"Name": None, "cTP": None, "mTP": None, "SP": None, "Loc": None}
+    is_datasection = False
+
+    for line in f:
+        if line.startswith("---------"):
+            is_datasection = not is_datasection
+        elif is_datasection:
+            split = line.split()
+            if not split[0] in documents:
+                documents[split[0]] = {"id": split[0], "computations": [{'tool': tool, 'results':[{}]}]}
+                results = documents[split[0]]['computations'][0]["results"][0]
+                for field in field_index:
+                    if field_index[field] is not None:
+                        if field in ["cTP", "mTP", "SP"]:
+                            results[field.lower()] = float(split[field_index[field]])
+                        elif field == "Loc":
+                            results[field.lower()] = loc_dict[split[field_index[field]]]
+                        else:
+                            results[field.lower()] = split[field_index[field]]
+        else:
+            if line.startswith('Name   '):
+                split = line.split()
+                for field in field_index:
+                    try:
+                        field_index[field] = split.index(field)
+                    except ValueError:
+                        print('TargetP Converter: No column "' + field + '" found in table!')
+            elif line.startswith('### targetp '):
+                split = line.split()
+                tool = {'name': 'TargetP',
+                        'version': split[2].replace("v", ""),
+                        'organism_group': 'plant'}
+            elif ' NON-PLANT ' in line:
+                tool['organism_group'] = 'non-plant'
+
+output_filename = args.output
+with open(output_filename, 'w') as o:
+    json.dump(documents, o)

scripts/run_targetp.py

+#!/usr/bin/env python3
+import env
+import argparse
+from psot import config
+from os import system
+
+targetp_tool = config.load_config()['tools'].get('targetp', 'targetp')
+
+org_flags = {'plant': '-P', 'non-plant': '-N'}
+
+parser = argparse.ArgumentParser(description = 'Determine subcellular locations of eukaryotic amino acid sequences')
+parser.add_argument('--fasta', '-f', required = True, help = 'A fasta file with amino acid sequences')
+parser.add_argument('--organism_group', choices = org_flags.keys(), required = True, help = 'Define wether to use plant/non-plant networks')
+parser.add_argument('--output', required = True, help = 'The output file')
+args = parser.parse_args()
+
+system(targetp_tool + " " + org_flags[args.organism_group]  + " " + args.fasta + " > " + args.output)