Use rdkit for SSSR and RCs (bug fix + Python upgrade)

.github/workflows/CI.yml

@@ -56,7 +56,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.9"]
+        python-version: ["3.9", "3.10", "3.11", "3.12"]
         os: [macos-13, macos-latest, ubuntu-latest]
         include-rms: ["", "with RMS"]
         exclude:

environment.yml

@@ -83,7 +83,6 @@ dependencies:
   # bug in quantities, see:
   # https://github.com/ReactionMechanismGenerator/RMG-Py/pull/2694#issuecomment-2489286263
   - conda-forge::quantities !=0.16.0,!=0.16.1
-  - conda-forge::ringdecomposerlib-python
 
 # packages we maintain
   - rmg::pydas >=1.0.3

rmgpy/molecule/graph.pyx

@@ -35,7 +35,7 @@ are the components of a graph.
 
 import itertools
 
-import py_rdl
+from rdkit import Chem
 
 from rmgpy.molecule.vf2 cimport VF2
 
@@ -971,68 +971,57 @@ cdef class Graph(object):
 
     cpdef list get_smallest_set_of_smallest_rings(self):
         """
-        Returns the smallest set of smallest rings as a list of lists.
-        Uses RingDecomposerLib for ring perception.
-
-        Kolodzik, A.; Urbaczek, S.; Rarey, M.
-        Unique Ring Families: A Chemically Meaningful Description
-        of Molecular Ring Topologies.
-        J. Chem. Inf. Model., 2012, 52 (8), pp 2013-2021
-
-        Flachsenberg, F.; Andresen, N.; Rarey, M.
-        RingDecomposerLib: An Open-Source Implementation of
-        Unique Ring Families and Other Cycle Bases.
-        J. Chem. Inf. Model., 2017, 57 (2), pp 122-126
-        """
-        cdef list sssr
-        cdef object graph, data, cycle
-
-        graph = py_rdl.Graph.from_edges(
-            self.get_all_edges(),
-            _get_edge_vertex1,
-            _get_edge_vertex2,
-        )
-
-        data = py_rdl.wrapper.DataInternal(graph.get_nof_nodes(), graph.get_edges().keys())
-        data.calculate()
-
-        sssr = []
-        for cycle in data.get_sssr():
-            sssr.append(self.sort_cyclic_vertices([graph.get_node_for_index(i) for i in cycle.nodes]))
-
+        Returns the smallest set of smallest rings (SSSR) as a list of lists of atom indices.
+        Uses RDKit's built-in ring perception (GetSymmSSSR).
+
+        References:
+            Kolodzik, A.; Urbaczek, S.; Rarey, M.
+            Unique Ring Families: A Chemically Meaningful Description
+            of Molecular Ring Topologies.
+            J. Chem. Inf. Model., 2012, 52 (8), pp 2013-2021
+
+            Flachsenberg, F.; Andresen, N.; Rarey, M.
+            RingDecomposerLib: An Open-Source Implementation of
+            Unique Ring Families and Other Cycle Bases.
+            J. Chem. Inf. Model., 2017, 57 (2), pp 122-126
+        """
+        cdef list sssr = []
+        cdef object ring_info, ring
+        # Get the symmetric SSSR using RDKit
+        ring_info = Chem.GetSymmSSSR(self)
+        for ring in ring_info:
+            # Convert ring (tuple of atom indices) to sorted list
+            sorted_ring = self.sort_cyclic_vertices(list(ring))
+            sssr.append(sorted_ring)
         return sssr
 
     cpdef list get_relevant_cycles(self):
         """
-        Returns the set of relevant cycles as a list of lists.
-        Uses RingDecomposerLib for ring perception.
-
-        Kolodzik, A.; Urbaczek, S.; Rarey, M.
-        Unique Ring Families: A Chemically Meaningful Description
-        of Molecular Ring Topologies.
-        J. Chem. Inf. Model., 2012, 52 (8), pp 2013-2021
-
-        Flachsenberg, F.; Andresen, N.; Rarey, M.
-        RingDecomposerLib: An Open-Source Implementation of
-        Unique Ring Families and Other Cycle Bases.
-        J. Chem. Inf. Model., 2017, 57 (2), pp 122-126
-        """
-        cdef list rc
-        cdef object graph, data, cycle
-
-        graph = py_rdl.Graph.from_edges(
-            self.get_all_edges(),
-            _get_edge_vertex1,
-            _get_edge_vertex2,
-        )
-
-        data = py_rdl.wrapper.DataInternal(graph.get_nof_nodes(), graph.get_edges().keys())
-        data.calculate()
-
-        rc = []
-        for cycle in data.get_rcs():
-            rc.append(self.sort_cyclic_vertices([graph.get_node_for_index(i) for i in cycle.nodes]))
-
+        Returns the set of relevant cycles as a list of lists of atom indices.
+        Uses RDKit's RingInfo to approximate relevant cycles.
+
+        References:
+            Kolodzik, A.; Urbaczek, S.; Rarey, M.
+            Unique Ring Families: A Chemically Meaningful Description
+            of Molecular Ring Topologies.
+            J. Chem. Inf. Model., 2012, 52 (8), pp 2013-2021
+
+            Flachsenberg, F.; Andresen, N.; Rarey, M.
+            RingDecomposerLib: An Open-Source Implementation of
+            Unique Ring Families and Other Cycle Bases.
+            J. Chem. Inf. Model., 2017, 57 (2), pp 122-126
+        """
+        cdef list rc = []
+        cdef object mol = self
+        cdef object ring_info = mol.GetRingInfo()
+        cdef object atom_rings = ring_info.AtomRings()
+        cdef object ring
+        for ring in atom_rings:
+            # Convert ring (tuple of atom indices) to sorted list
+            sorted_ring = self.sort_cyclic_vertices(list(ring))
+            # Filter for "relevant" cycles (e.g., rings up to size 7)
+            if len(sorted_ring) <= 7:
+                rc.append(sorted_ring)
         return rc
 
     cpdef list sort_cyclic_vertices(self, list vertices):

setup.py

@@ -146,7 +146,7 @@
     author='William H. Green and the RMG Team',
     author_email='[email protected]',
     url='http://reactionmechanismgenerator.github.io',
-    python_requires='>=3.9,<3.10',
+    python_requires='>=3.9,<3.13',
     packages=find_packages(where='.', include=["rmgpy*"]) + find_packages(where='.', include=["arkane*"]),
     scripts=scripts,
     include_package_data=True,

utilities.py

@@ -50,7 +50,6 @@ def check_dependencies():
     missing = {
         'openbabel': _check_openbabel(),
         'pydqed': _check_pydqed(),
-        'pyrdl': _check_pyrdl(),
         'rdkit': _check_rdkit(),
         'symmetry': _check_symmetry(),
     }
@@ -104,22 +103,6 @@ def _check_pydqed():
     return missing
 
 
-def _check_pyrdl():
-    """Check for pyrdl"""
-    missing = False
-
-    try:
-        import py_rdl
-    except ImportError:
-        print('{0:<30}{1}'.format('pyrdl', 'Not found. Necessary for ring perception algorithms.'))
-        missing = True
-    else:
-        location = py_rdl.__file__
-        print('{0:<30}{1}'.format('pyrdl', location))
-
-    return missing
-
-
 def _check_rdkit():
     """Check for RDKit"""
     missing = False