From 22f6e132e0692da255453e5ad5087bdd59bdb75f Mon Sep 17 00:00:00 2001
From: Gordon Getzinger <16693333+gjgetzinger@users.noreply.github.com>
Date: Thu, 16 Apr 2020 14:14:42 -0400
Subject: [PATCH] Defines `super_parent_syn`

During compound standardization/registration, it can be helpful to preserve molecule identifiers for mapping synonyms or related molecules. This function mirrors the `super_parent` function, but captures InChI identifiers at each sanitization step and assigns them as properties of the sanitized molecule (either InChI or InChI Key).

```python
smi = '[13CH3][C@@H](C)C/C=C([O-])\O.[Na+]'
mol = Chem.MolFromSmiles(smi)
mol_san = super_parent_syn(mol, 'inchikey')
mol_san.GetPropsAsDict()
```

```
{'fragment_inchikey': 'GCBSTXJCFNXFGB-DQIXCRJTSA-M', 'charge_inchikey': 'VBJZLOGJBQCFRJ-BBAZYXFWSA-M', 'isotope_inchikey': 'VBJZLOGJBQCFRJ-BBAZYXFWSA-N', 'stereo_inchikey': 'VBJZLOGJBQCFRJ-UHFFFAOYSA-N', 'tautomer_inchikey': 'VBJZLOGJBQCFRJ-UHFFFAOYSA-N', 'standardize_inchikey': 'FGKJLKRYENPLQH-UHFFFAOYSA-N', 'inchikey': 'FGKJLKRYENPLQH-UHFFFAOYSA-N'}
```
---
 molvs/standardize.py | 38 +++++++++++++++++++++++++++++++++++++-
 1 file changed, 37 insertions(+), 1 deletion(-)

diff --git a/molvs/standardize.py b/molvs/standardize.py
index ca3f585..e3ae588 100644
--- a/molvs/standardize.py
+++ b/molvs/standardize.py
@@ -211,7 +211,43 @@ def super_parent(self, mol, skip_standardize=False):
         mol = self.tautomer_parent(mol, skip_standardize=True)
         mol = self.standardize(mol)
         return mol
-
+    
+    def super_parent_syn(self, mol, synonyms_as = 'inchikey', skip_standardize=False):
+        """Return the super parent of a given molecule with InChI(Key)s preserved as properties.
+        
+        :param mol: The input molecule.
+        :param synonyms_as: How should synonyms be kept? (inchi or inchikey)
+        :type mol: rdkit.Chem.rdchem.Mol
+        :param bool skip_standardize: Set to True if mol has already been standardized.
+        :returns: The super parent molecule.
+        :rtype: rdkit.Chem.rdchem.Mol
+        """
+        if not skip_standardize:
+            mol = self.standardize(mol)
+        
+        inchi = {}
+        inchi['fragment_inchi'] = Chem.MolToInchi(mol)
+        mol = self.fragment_parent(mol, skip_standarize=True)
+        inchi['charge_inchi'] = Chem.MolToInchi(mol)
+        mol = self.charge_parent(mol, skip_standardize=True)
+        inchi['isotope_inchi'] = Chem.MolToInchi(mol)
+        mol = self.isotope_parent(mol, skip_standardize=True)
+        inchi['stereo_inchi'] = Chem.MolToInchi(mol)
+        mol = self.stereo_parent(mol, skip_standardize=True)
+        inchi['tautomer_inchi'] = Chem.MolToInchi(mol)
+        mol = self.tautomer_parent(mol, skip_standardize=True)
+        inchi['standardize_inchi'] = Chem.MolToInchi(mol)
+        mol = self.standardize(mol)
+        inchi['inchi'] = Chem.MolToInchi(mol)
+        
+        if synonyms_as == 'inchi':
+            [mol.SetProp(key,inchi[key]) for key in inchi]
+        else:
+            inchikey = {key+'key':Chem.InchiToInchiKey(inchi[key]) for key in inchi}
+            [mol.SetProp(key,inchikey[key]) for key in inchikey]
+        
+        return mol
+    
     def standardize_with_parents(self, mol):
         """"""
         standardized = self.standardize(mol)