Mol from rdkit

devtools/conda/meta.yaml

@@ -26,7 +26,7 @@ requirements:
     - networkx
     - rdkit
     - ambertools
-    - openforcefield==0.7.1
+    - openforcefield==0.8.0
 
 about:
   home: https://github.com/martimunicoy/peleffy

devtools/envs/standard.yaml

@@ -13,4 +13,4 @@ dependencies:
   - coverage < 5.0
   - ambertools
   - rdkit
-  - openforcefield ==0.7.1
+  - openforcefield ==0.8.0

docs/releasehistory.rst

@@ -18,6 +18,7 @@ New features
 - `PR #86 <https://github.com/martimunicoy/peleffy/pull/86>`_: New method to check the input PDB prior building the molecule.
 - `PR #94 <https://github.com/martimunicoy/peleffy/pull/94>`_: New method for the OPLS OBC parameters.
 - `PR #92 <https://github.com/martimunicoy/peleffy/pull/92>`_: New parameter to skip the stereochemistry assignment (and the checking from the OpenFF toolkit).
+- `PR #106 <https://github.com/martimunicoy/peleffy/pull/106>`_: New method to initialize a Molecule object directly from an RDKit and OpenFF molecular representations.
 
 API-breaking changes
 """"""""""""""""""""
@@ -31,6 +32,7 @@ Tests added
 - `PR #94 <https://github.com/martimunicoy/peleffy/pull/94>`_: Adds tests to validate the OPLS OBC parameters generator.
 - `PR #92 <https://github.com/martimunicoy/peleffy/pull/92>`_: New test to check the behaviour of the allow_undefined_stereo parameter.
 - `PR #97 <https://github.com/martimunicoy/peleffy/pull/97>`_: Includes tests for the new Topology container class.
+- `PR #106 <https://github.com/martimunicoy/peleffy/pull/106>`_: Adds tests to check the RDKit and OpenFF molecular initializers.
 
 1.0.0 - Full compatibility for OPLS2005 and OpenFF dihedrals and rotamer library improvements
 ---------------------------------------------------------------------------------------------

peleffy/tests/test_molecule.py

@@ -295,3 +295,51 @@ def test_undefined_stereo(self):
         with tempfile.TemporaryDirectory() as tmpdir:
             with temporary_cd(tmpdir):
                 mol.to_pdb_file('molecule.pdb')
+
+    def test_from_rdkit(self):
+        """
+        It checks the initialization of a peleffy Molecule from an RDKit
+        molecular representation.
+        """
+        from rdkit import Chem
+
+        pdb_path = get_data_file_path('ligands/malonate.pdb')
+        rdkit_molecule = Chem.MolFromPDBFile(pdb_path, removeHs=False)
+        molecule = Molecule.from_rdkit(rdkit_molecule)
+
+        assert molecule._rdkit_molecule is not None, \
+            'Unexpected molecule representation found, it is not initialized'
+        assert molecule._off_molecule is not None, \
+            'Unexpected molecule representation found, it is not initialized'
+
+        ref_pdb_atom_names = [' O1 ', ' C1 ', ' O2 ', ' C2 ', ' C3 ',
+                              ' O3 ', ' O4 ', ' H1 ', ' H2 ', ' H3 ']
+        pdb_atom_names = molecule.get_pdb_atom_names()
+
+        assert pdb_atom_names == ref_pdb_atom_names, \
+            'Unexpected PDB atom names found in the resulting Molecule ' \
+            + 'representation'
+
+    def test_from_openff(self):
+        """
+        It checks the initialization of a peleffy Molecule from an OpenFF
+        molecular representation.
+        """
+        from openforcefield.topology import Molecule as OpenFFMolecule
+
+        openff_molecule = OpenFFMolecule.from_smiles('C(C(=O)[O-])C(=O)[OH]')
+
+        molecule = Molecule.from_openff(openff_molecule)
+
+        assert molecule._rdkit_molecule is not None, \
+            'Unexpected molecule representation found, it is not initialized'
+        assert molecule._off_molecule is not None, \
+            'Unexpected molecule representation found, it is not initialized'
+
+        ref_pdb_atom_names = [' C1 ', ' C2 ', ' O1 ', ' O2 ', ' C3 ',
+                              ' O3 ', ' O4 ', ' H1 ', ' H2 ', ' H3 ']
+        pdb_atom_names = molecule.get_pdb_atom_names()
+
+        assert pdb_atom_names == ref_pdb_atom_names, \
+            'Unexpected PDB atom names found in the resulting Molecule ' \
+            + 'representation'

peleffy/topology/molecule.py

@@ -377,6 +377,161 @@ def get_conformer(self):
 
         return rdkit_toolkit.get_coordinates(self)
 
+    @staticmethod
+    def from_rdkit(rdkit_molecule, rotamer_resolution=30,
+                   exclude_terminal_rotamers=True, name='', tag='UNK',
+                   connectivity_template=None, core_constraints=[],
+                   allow_undefined_stereo=False):
+        """
+        It initializes and returns a peleffy Molecule representation
+        from an RDKit molecular representation.
+
+        Parameters
+        ----------
+        rdkit_molecule : an rdkit.Chem.rdchem.Mol object
+            The RDKit's Molecule object to use to initialize a peleffy
+            Molecule object
+        rotamer_resolution : float
+            The resolution in degrees to discretize the rotamer's
+            conformational space. Default is 30
+        exclude_terminal_rotamers : bool
+            Whether to exclude terminal rotamers when generating the
+            rotamers library  or not
+        name : str
+            The molecule name
+        tag : str
+            The molecule tag. It must be a 3-character string
+        connectivity_template : an rdkit.Chem.rdchem.Mol object
+            A molecule represented with RDKit to use when assigning the
+            connectivity of this Molecule object
+        core_constraints : list[int or str]
+            It defines the list of atoms to constrain in the core, thus,
+            the core will be forced to contain them. Atoms can be specified
+            through integers that match the atom index or strings that
+            match with the atom PDB name
+        allow_undefined_stereo : bool
+            Whether to allow a molecule with undefined stereochemistry
+            to be defined or try to assign the stereochemistry and
+            raise a complaint if not possible. Default is False
+
+        Returns
+        -------
+        molecule : an peleffy.topology.Molecule
+            The resulting peleffy's Molecule object
+
+        Examples
+        --------
+
+        Load a molecule from an RDKit molecular representation
+
+        >>> from rdkit import Chem
+
+        >>> rdkit_molecule = Chem.MolFromPDBFile(pdb_path, removeHs=False)
+
+        >>> from peleffy.topology import Molecule
+
+        >>> molecule = Molecule.from_rdkit(rdkit_molecule)
+
+        """
+        molecule = Molecule(
+            rotamer_resolution=30,
+            exclude_terminal_rotamers=exclude_terminal_rotamers,
+            name=name, tag=tag,
+            connectivity_template=connectivity_template,
+            core_constraints=core_constraints,
+            allow_undefined_stereo=allow_undefined_stereo)
+
+        logger = Logger()
+
+        logger.info(' - Initializing molecule from an RDKit '
+                    + 'molecular representation')
+        molecule._initialize()
+        molecule._rdkit_molecule = rdkit_molecule
+
+        logger.info('   - Representing molecule with the Open Force Field '
+                    + 'Toolkit')
+        openforcefield_toolkit = OpenForceFieldToolkitWrapper()
+        molecule._off_molecule = openforcefield_toolkit.from_rdkit(molecule)
+
+        return molecule
+
+    @staticmethod
+    def from_openff(openff_molecule, rotamer_resolution=30,
+                    exclude_terminal_rotamers=True, name='', tag='UNK',
+                    connectivity_template=None, core_constraints=[],
+                    allow_undefined_stereo=False):
+        """
+        It initializes and returns a peleffy Molecule representation
+        from an OpenForceField molecular representation.
+
+        Parameters
+        ----------
+        openff_molecule : an openforcefield.topology.Molecule object
+            The OpenForceField's Molecule to use to initialize a peleffy
+            Molecule object
+        rotamer_resolution : float
+            The resolution in degrees to discretize the rotamer's
+            conformational space. Default is 30
+        exclude_terminal_rotamers : bool
+            Whether to exclude terminal rotamers when generating the
+            rotamers library  or not
+        name : str
+            The molecule name
+        tag : str
+            The molecule tag. It must be a 3-character string
+        connectivity_template : an rdkit.Chem.rdchem.Mol object
+            A molecule represented with RDKit to use when assigning the
+            connectivity of this Molecule object
+        core_constraints : list[int or str]
+            It defines the list of atoms to constrain in the core, thus,
+            the core will be forced to contain them. Atoms can be specified
+            through integers that match the atom index or strings that
+            match with the atom PDB name
+        allow_undefined_stereo : bool
+            Whether to allow a molecule with undefined stereochemistry
+            to be defined or try to assign the stereochemistry and
+            raise a complaint if not possible. Default is False
+
+        Returns
+        -------
+        molecule : an peleffy.topology.Molecule
+            The resulting peleffy's Molecule object
+
+        Examples
+        --------
+
+        Load a molecule from an RDKit molecular representation
+
+        >>> from rdkit import Chem
+
+        >>> rdkit_molecule = Chem.MolFromPDBFile(pdb_path)
+
+        >>> from peleffy.topology import Molecule
+
+        >>> molecule = Molecule.from_rdkit(rdkit_molecule)
+
+        """
+        molecule = Molecule(
+            rotamer_resolution=30,
+            exclude_terminal_rotamers=exclude_terminal_rotamers,
+            name=name, tag=tag,
+            connectivity_template=connectivity_template,
+            core_constraints=core_constraints,
+            allow_undefined_stereo=allow_undefined_stereo)
+
+        logger = Logger()
+
+        logger.info(' - Initializing molecule from an OpenFF '
+                    + 'molecular representation')
+        molecule._initialize()
+        molecule._off_molecule = openff_molecule
+
+        logger.info('   - Generating RDKit molecular representation with '
+                    + 'the Open Force Field Toolkit')
+        molecule._rdkit_molecule = openff_molecule.to_rdkit()
+
+        return molecule
+
     @property
     def rotamer_resolution(self):
         """