Changed galaxy xml file to use conda biotransformer package

tools/biotransformer/biotransformer.xml

@@ -1,8 +1,10 @@
-<tool id="biotransformer" name="BioTransformer" version="1.1.0">
-    <requirements>
-      <container type="docker">registry.gitlab.ics.muni.cz:443/recetox/mass-spectrometry/biotransformer:1.1.0</container>
-    </requirements>
-  <command><![CDATA[ cd /biotransformer && python3 wrapper_biotransformer.py
+<tool id="biotransformer" name="BioTransformer" version="1.1.5">
+  <requirements>
+     <requirement type="package" version="1.1.5"> biotransformer </requirement>
+     <requirement type="package" version="3.1.1"> openbabel </requirement>
+     <requirement type="package" version="1.1.1"> pandas </requirement>
+  </requirements>
+  <command><![CDATA[ python3 '${__tool_directory__}/wrapper_biotransformer.py'
  -k $k
  -b $b 
  -s $steps
@@ -127,3 +129,4 @@ reaction(Oxidation, reduction, (de-)conjugation) at each step.
     <citation type="doi">https://doi.org/10.1186/s13321-018-0324-5</citation>
   </citations>
 </tool>
+

tools/biotransformer/wrapper_biotransformer.py

@@ -0,0 +1,89 @@
+import sys
+import tempfile
+import subprocess
+import pandas
+from openbabel import pybel
+
+#function for translating inchi to smiles
+def InchiToSmiles(df):
+    sm = []
+    for item in df['InChI']:
+         tmp = pybel.readstring("inchi",item)
+         sm.append(tmp.write("smi"))
+    return(sm)
+
+executable = ["biotransformer"]
+#executable_r = ["Rscript", "inchi_to_smiles.r"]
+
+argv = sys.argv[1:]
+if "-icsv" in argv:
+    icsv = argv.pop(argv.index("-icsv") + 1)
+    argv.remove("-icsv")
+
+    if "-ocsv" not in argv:
+        sys.stderr.write("excpected -ocsv parameter\n")
+        sys.exit(1)
+    ocsv = argv.pop(argv.index("-ocsv") + 1)
+    argv.remove("-ocsv")
+    ocsv_dup = argv.pop(argv.index("-ocsvDup") + 1)
+    argv.remove("-ocsvDup")
+    ocsv_dup2 = argv.pop(argv.index("-ocsvDup2") + 1)
+    argv.remove("-ocsvDup2")
+
+    in_df = pandas.read_csv(icsv, header=None)
+    out_df1 = pandas.DataFrame() #file for all results
+    out_df2 = pandas.DataFrame() #file for filtered results based on 6 columns
+    out_df3 = pandas.DataFrame() #file for filtered results based on 3 columns
+
+    tmp2 = pandas.DataFrame()
+    tmp3 = pandas.DataFrame()
+
+    smList1 = [] #list with smiles string
+    smList2 = []
+    smList3 = []
+    for _, (smiles,) in in_df.iterrows():
+        with tempfile.NamedTemporaryFile() as out:
+            subprocess.run(executable + argv + ["-ismi", smiles] + ["-ocsv", out.name])
+            tmp2 = pandas.read_csv(out.name)
+            tmp3 = pandas.read_csv(out.name)
+            tmp2.drop_duplicates(inplace=True, subset=["InChI","InChIKey","Synonyms","Molecular formula","Major Isotope Mass","ALogP"])
+            tmp3.drop_duplicates(inplace=True, subset=["Molecular formula","Major Isotope Mass","ALogP"])
+            smList2.append([smiles]*tmp2.shape[0])
+            smList3.append([smiles]*tmp3.shape[0])
+            out_df1 = pandas.concat([out_df1, pandas.read_csv(out.name)])
+            out_df2 = pandas.concat([out_df2, tmp2])
+            out_df3 = pandas.concat([out_df3, tmp3])
+            smList1.append([smiles]*pandas.read_csv(out.name).shape[0])
+
+
+    smList1 = sum(smList1,[]) #merge sublists into one list
+    smList2 = sum(smList2,[])
+    smList3 = sum(smList3,[])
+
+    out_df1.insert(0,"SMILES query",smList1)
+    out_df1.drop_duplicates(inplace=True)
+    out_df1.insert(1,"SMILES target", InchiToSmiles(out_df1))
+    out_df1.to_csv(ocsv)
+
+    out_df2.insert(0,"SMILES query",smList2)
+    out_df3.insert(0,"SMILES query",smList3)
+    out_df2.drop_duplicates(inplace=True)
+    out_df3.drop_duplicates(inplace=True)
+    out_df2.insert(1,"SMILES target", InchiToSmiles(out_df2))
+    out_df3.insert(1,"SMILES target", InchiToSmiles(out_df3))
+    #out_df.drop_duplicates(inplace=True, subset=["InChI","InChIKey","Synonyms","Molecular formula","Major Isotope Mass","ALogP"])
+    out_df2.to_csv(ocsv_dup)
+    out_df3.to_csv(ocsv_dup2)    
+else:
+    #code = subprocess.run(executable + argv).returncode
+    #sys.exit(code)
+    subprocess.run(executable + argv)
+    smile = argv.pop(argv.index("-ismi") + 1)
+
+    tmp = pandas.DataFrame()
+    out = argv.pop(argv.index("-ocsv") + 1)
+    tmp = pandas.read_csv(out) #reads created output file
+    tmp.insert(0,"SMILES query",smile) #add SMILES string for query
+    tmp.insert(1,"SMILES target", InchiToSmiles(tmp)) #add SMILES string for target
+    tmp.to_csv(out)
+