SA score

python/local/SyntheticAccessibilityScore.yaml

@@ -0,0 +1,87 @@
+id: 5ef62e8f-e15f-4675-9a2c-3b13c3265e43
+name: Synthetic Accessibility Score
+description: Calculates the Synthetic Accessibility Score using the RDKit implementation.
+category: Chemistry
+version: 1.0.0
+serviceName: Script
+serviceUri: glysade.python
+executorId: Glysade.CPythonDataFxn
+inputFields:
+- control:
+    id: structureColumn
+    label: Select structure column
+    type: columnselect
+    filters:
+    - dataType: string
+      contentType:
+      - chemical/x-mdl-molfile
+      - chemical/x-mdl-molfile-v3000
+      - chemical/x-smiles
+      - chemical/x-daylight-smiles
+    - dataType: binary
+      contentType:
+      - chemical/x-mdl-molfile
+      - chemical/x-mdl-molfile-v3000
+    validationRules:
+    - type: required
+      message: Must select column of 2D structures
+  request:
+    id: structureColumn
+    dataType: string
+    selectorType: column
+tags:
+- color: '#50AF28'
+  text: chemistry
+- color: '#c6fc00'
+  text: calcprop
+updateBehavior: automatic
+maximumOutputColumns: !!int 1
+maximumOutputTables: !!int 0
+chemistryFunction: !!bool false
+script: |
+  from df.sascorer import calculateScore
+  from df.chem_helper import column_to_molecules
+  from df.data_transfer import (DataFunctionRequest, DataFunctionResponse,
+                                DataType, ColumnData, string_input_field)
+  from rdkit.Chem import AllChem
+  from rdkit.Chem.MolStandardize import rdMolStandardize
+
+  from ruse.rdkit.rdkit_utils import standardize_mol
+
+
+  def execute(request: DataFunctionRequest) -> DataFunctionResponse:
+      column_id = string_input_field(request, 'structureColumn')
+      input_column = request.inputColumns[column_id]
+      mols = column_to_molecules(input_column)
+
+      # The objects used for standardizing the molecules when creating
+      # the scores database.
+      # SureChEMBL has some odd azides that the normalizer doesn't touch.
+      fix_azide = AllChem.ReactionFromSmarts('[N-:1]=[N:2]#[N+:3]>>[N+0:1]#[N+:2][N-:3]')
+      normer = rdMolStandardize.Normalizer()
+      uncharger = rdMolStandardize.Uncharger()
+      metal_disconnector = rdMolStandardize.MetalDisconnector()
+
+      scores = []
+      for m in mols:
+          if m is None:
+              scores.append(None)
+          else:
+              sm = standardize_mol(m, normer, uncharger, metal_disconnector,
+                                   fix_azide)
+              scores.append(calculateScore(sm))
+      output_column = ColumnData(name=f'{input_column.name} SA Score',
+                                 dataType=DataType.DOUBLE, values=scores)
+      response = DataFunctionResponse(outputColumns=[output_column])
+      return response
+outputFields:
+- id: column1
+  source: column
+  type: filter
+  name: Output SA Score column
+allowedClients:
+- Analyst
+- WebPlayer
+demoUrl: 
+limitBy: none
+minimumChartsVersion: