further improvements to charge and taut/stereo enumerators

Author: tdudgeon

src/python/pipelines/dimorphite/enumerate_charges.dsd.yml

@@ -66,9 +66,9 @@ serviceConfig:
       - mw
     defaultValue: hac
     visible: true
-  executorClassName: org.squonk.execution.steps.impl.ThinDatasetDockerExecutorStep
-thinDescriptors:
-- input: input
+  executorClassName: org.squonk.execution.steps.impl.DefaultDockerExecutorStep
+#thinDescriptors:
+#- input: input
 inputRoutes:
 - route: FILE
 outputRoutes:

src/python/pipelines/dimorphite/enumerate_charges.py

@@ -47,9 +47,9 @@ def writeEnumeratedMols(src_mol, enum_mols, writer, index):
 
 def add_src_mol_ref(src_mol, target_mol, index):
     """
-    Add the ID of the source molecule to the enumerated molecule as the field named EnumChargeSrcMol.
-    The ID is taken form the uuid field if it exists, if not form the _Name field if it exists and finally
-    from the index parameter (the index of the source molecule in the input) if neither of those fields are found.
+    Add the ID of the source molecule to the enumerated molecule as the field named EnumChargesSrcMolUUID.
+    The ID is taken form the uuid field if it exists, if not form the _Name field if it exists.
+    The EnumChargesSrcMolIdx field is always set with the index of the source molecule in the input..
     :param src_mol:
     :param target_mol:
     :param index:
@@ -59,11 +59,11 @@ def add_src_mol_ref(src_mol, target_mol, index):
         parent = src_mol.GetProp('uuid')
     elif src_mol.HasProp('_name_'):
         parent = src_mol.GetProp('_Name')
-    else:
-        parent = str(index)
 
     if parent:
-        target_mol.SetProp('EnumChargeSrcMol', parent)
+        target_mol.SetProp('EnumChargesSrcMolUUID', parent)
+
+    target_mol.SetIntProp('EnumChargesSrcMolIdx', index)
 
 ### start main execution #########################################
 
@@ -88,9 +88,12 @@ def main():
     source = "enumerate_charges.py"
     datasetMetaProps = {"source":source, "description": "Enumerate charges using Dimorphite-dl"}
     clsMappings = {
-        "EnumChargeSrcMol": "java.lang.String"}
+        "EnumChargesSrcMolUUID": "java.lang.String",
+        "EnumChargesSrcMolIdx": "java.lang.Integer"
+    }
     fieldMetaProps = [
-        {"fieldName":"EnumChargeSrcMol",   "values": {"source":source, "description":"ID of source molecule"}}
+        {"fieldName":"EnumChargesSrcMolUUID", "values": {"source":source, "description":"UUID of source molecule"}},
+        {"fieldName":"EnumChargesSrcMolIdx", "values": {"source":source, "description":"Index of source molecule"}}
     ]
 
     oformat = utils.determine_output_format(args.outformat)

src/python/pipelines/rdkit/sanifier.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python
 
-# Copyright 2017 Informatics Matters Ltd.
+# Copyright 2019 Informatics Matters Ltd.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -14,7 +14,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-### Use MolVS to do tautomer enumeration, sterochemistry enumeration, charge neutralisation.
+### Use MolVS to do tautomer enumeration, stereochemistry enumeration, charge neutralisation.
 
 import sys, argparse
 
@@ -73,7 +73,28 @@ def main():
     if args.standardize:
         getStandardMolecule = STANDARD_MOL_METHODS[args.standardize_method]
 
-    input ,output ,suppl ,writer ,output_base = rdkit_utils.default_open_input_output(args.input, args.informat, args.output, 'sanify', args.outformat)
+    # handle metadata
+    source = "sanifier.py"
+    datasetMetaProps = {"source":source, "description": "Enumerate tautomers and stereoisomers"}
+    clsMappings = {
+        "EnumTautIsoSourceMolUUID": "java.lang.String",
+        "EnumTautIsoSourceMolIdx": "java.lang.Integer"
+    }
+    fieldMetaProps = [
+        {"fieldName":"EnumTautIsoSourceMolUUID", "values": {"source":source, "description":"UUID of source molecule"}},
+        {"fieldName":"EnumTautIsoSourceMolIdx", "values": {"source":source, "description":"Index of source molecule"}}
+    ]
+
+    oformat = utils.determine_output_format(args.outformat)
+
+    input,output,suppl,writer,output_base = rdkit_utils. \
+        default_open_input_output(args.input, args.informat, args.output,
+                                  'sanifier', args.outformat,
+                                  thinOutput=False, valueClassMappings=clsMappings,
+                                  datasetMetaProps=datasetMetaProps,
+                                  fieldMetaProps=fieldMetaProps)
+
+
     i=0
     count=0
     errors=0
@@ -110,11 +131,13 @@ def main():
                 parentUuid = None
 
             results = []
-            results.append(mol)
+
 
             if args.enumerate_tauts:
                 utils.log("Enumerating tautomers")
                 results = enumerateTautomers(mol)
+            else:
+                results.append(mol)
 
             if args.enumerate_stereo:
                 utils.log("Enumerating steroisomers")
@@ -125,10 +148,14 @@ def main():
                     results.extend(enumerated)
 
             for m in results:
+                # copy the src mol props
+                for name in mol.GetPropNames():
+                    m.SetProp(name, mol.GetProp(name))
+                # add our new props
                 m.ClearProp("uuid")
-                m.SetIntProp("SourceMolNum", i)
+                m.SetIntProp("EnumTautIsoSourceMolIdx", i)
                 if parentUuid:
-                    m.SetProp("SourceMolUUID", parentUuid)
+                    m.SetProp("EnumTautIsoSourceMolUUID", parentUuid)
 
             count = write_out(results,count,writer,args.mol_format,args.outformat)
 
@@ -139,6 +166,11 @@ def main():
     input.close()
     output.close()
 
+    # re-write the metadata as we now know the size
+    if oformat == 'json':
+        utils.write_squonk_datasetmetadata(output_base, False, clsMappings, datasetMetaProps, fieldMetaProps, size=count)
+
+
     if args.meta:
         utils.write_metrics(output_base, {'__InputCount__':i, '__OutputCount__':count, '__ErrorCount__':errors , 'RDKitSanify':count })
 

src/python/pipelines/rdkit/sanifier_enumerator.dsd.yml

@@ -69,9 +69,7 @@ serviceConfig:
     minValues: 1
     maxValues: 1
     visible: true
-  executorClassName: org.squonk.execution.steps.impl.ThinDatasetDockerExecutorStep
-thinDescriptors:
-- input: input
+  executorClassName: org.squonk.execution.steps.impl.DefaultDockerExecutorStep
 inputRoutes:
 - route: FILE
 outputRoutes: