Merge pull request #91 from molssi-seamm/dev

Bugfix: more issues with property handling.

* The types of properties were not kept when using Open Babel or RDKit, so when properties were reread from an SDF file the JSON properties were converted into strings, causing various errors. This is fixed.

HISTORY.rst

@@ -1,6 +1,11 @@
 =======
 History
 =======
+2024.12.14 -- Bugfix: more issues with property handling.
+    * The types of properties were not kept when using Open Babel or RDKit, so when
+      properties were reread from an SDF file the JSON properties were converted into
+      strings, causing various errors. This is fixed.
+
 2024.12.11 -- Bugfix: Properties in SDF files
     * Transferring properties to the Open Babel and RDKit molecules was incorrect after
       recent changes to the handling of properties. This fixes the problem, and now SDF

molsystem/data/standard_properties.csv

@@ -1 +1 @@
-Property,Type,Units,Description,URLHOMO energy#experiment,float,eV,"The {model} energy of the highest occupied molecular orbital (HOMO)",https://en.wikipedia.org/wiki/HOMO_and_LUMOLUMO energy#experiment,float,eV,"The {model} energy of the lowest unoccupied molecular orbital (LUMO)",https://en.wikipedia.org/wiki/HOMO_and_LUMOband gap#experiment,float,eV,"The band gap in a solid.",https://en.wikipedia.org/wiki/Band_gapchemical formula,string,,"The chemical formula of the molecule or conventional unit cell.",constant pressure heat capacity#experiment,float,J/mol/K,"The constant-pressure heat capacity, Cp.",https://en.wikipedia.org/wiki/Heat_capacityconstant volume heat capacity#experiment,float,J/mol/K,"The constant-volume heat capacity, Cv.",https://en.wikipedia.org/wiki/Heat_capacitydensity,float,g/mL,https://en.wikipedia.org/wiki/Densitydensity#experiment,float,g/mL,https://en.wikipedia.org/wiki/Densitydipole moment#experiment,float,D,"The electric dipole moment, which is the magnitude of the dipole vector.",https://en.wikipedia.org/wiki/Electric_dipole_momentempirical formula,string,,"The empirical formula of the molecule or conventional unit cell.",enthalpy of formation#experiment,float,kJ/mol,"The enthalpy of formation, DHf.",https://en.wikipedia.org/wiki/Standard_enthalpy_of_formationentropy#experiment,float,J/mol/K,"The standard entropy.",https://en.wikipedia.org/wiki/Entropyionization energy#experiment,float,eV,"The ionization energy or potential.",https://en.wikipedia.org/wiki/Ionization_energymolecular weight,float,Da,"The molecular weight in Daltons.",nuclear repulsion energy,float,eV,"The repulsion energy between the nuclei, based on atomic numbers and distances",unit cell volume,float,Å^3,"The volume of the unit cell.",
+Property,Type,Units,Description,URLHOMO energy#experiment,float,eV,"The {model} energy of the highest occupied molecular orbital (HOMO)",https://en.wikipedia.org/wiki/HOMO_and_LUMOLUMO energy#experiment,float,eV,"The {model} energy of the lowest unoccupied molecular orbital (LUMO)",https://en.wikipedia.org/wiki/HOMO_and_LUMOband gap#experiment,float,eV,"The band gap in a solid.",https://en.wikipedia.org/wiki/Band_gapchemical formula,str,,"The chemical formula of the molecule or conventional unit cell.",constant pressure heat capacity#experiment,float,J/mol/K,"The constant-pressure heat capacity, Cp.",https://en.wikipedia.org/wiki/Heat_capacityconstant volume heat capacity#experiment,float,J/mol/K,"The constant-volume heat capacity, Cv.",https://en.wikipedia.org/wiki/Heat_capacitydensity,float,g/mL,https://en.wikipedia.org/wiki/Densitydensity#experiment,float,g/mL,https://en.wikipedia.org/wiki/Densitydipole moment#experiment,float,D,"The electric dipole moment, which is the magnitude of the dipole vector.",https://en.wikipedia.org/wiki/Electric_dipole_momentempirical formula,str,,"The empirical formula of the molecule or conventional unit cell.",enthalpy of formation#experiment,float,kJ/mol,"The enthalpy of formation, DHf.",https://en.wikipedia.org/wiki/Standard_enthalpy_of_formationentropy#experiment,float,J/mol/K,"The standard entropy.",https://en.wikipedia.org/wiki/Entropyionization energy#experiment,float,eV,"The ionization energy or potential.",https://en.wikipedia.org/wiki/Ionization_energymolecular weight,float,Da,"The molecular weight in Daltons.",nuclear repulsion energy,float,eV,"The repulsion energy between the nuclei, based on atomic numbers and distances",unit cell volume,float,Å^3,"The volume of the unit cell.",

molsystem/openbabel.py

@@ -3,6 +3,7 @@
 """Interface to openbabel."""
 
 import logging
+import json
 from pathlib import Path
 
 try:
@@ -70,32 +71,30 @@ def to_OBMol(self, properties=None):
         pair = ob.OBPairData()
 
         if self.__class__.__name__ == "_Configuration":
-            pair.SetAttribute("net charge")
+            pair.SetAttribute("SEAMM|net charge|int|")
             pair.SetValue(str(self.charge))
             ob_mol.CloneData(pair)
 
-            pair.SetAttribute("spin multiplicity")
+            pair.SetAttribute("SEAMM|spin multiplicity|int|")
             pair.SetValue(str(self.spin_multiplicity))
             ob_mol.CloneData(pair)
 
         if properties is not None:
             data = self.properties.get(properties, include_system_properties=True)
-            for key, value in data.items():
+            for _property, value in data.items():
+                _type = self.properties.type(_property)
+                units = self.properties.units(_property)
                 value = value["value"]
+                key = f"SEAMM|{_property}|{_type}|"
+                if units is not None and units != "":
+                    key += units
+
+                if _type == "json":
+                    value = json.dumps(value)
+
                 pair.SetAttribute(key)
                 pair.SetValue(str(value))
                 ob_mol.CloneData(pair)
-
-                # Units, if any
-                units = self.properties.units(key)
-                if units is not None and units != "":
-                    tmp = key.split("#", maxsplit=1)
-                    if len(tmp) > 1:
-                        pair.SetAttribute(tmp[0] + ",units" + "#" + tmp[1])
-                    else:
-                        pair.SetAttribute(key + ",units")
-                    pair.SetValue(units)
-                    ob_mol.CloneData(pair)
         return ob_mol
 
     def from_OBMol(
@@ -159,23 +158,27 @@ def from_OBMol(
         data = {}
         for item in ob_mol.GetData():
             value = item.GetValue()
-            try:
-                value = int(value)
-            except Exception:
+            key = item.GetAttribute()
+            if key.startswith("SEAMM|"):
+                data[key] = value
+            else:
                 try:
-                    value = float(value)
+                    value = int(value)
                 except Exception:
-                    pass
-            data[item.GetAttribute()] = value
+                    try:
+                        value = float(value)
+                    except Exception:
+                        pass
+                data[key] = value
 
         # Check for property items for charge and multiplicity
         if self.__class__.__name__ == "_Configuration":
             self.charge = ob_mol.GetTotalCharge()
             self.spin_multiplicity = ob_mol.GetTotalSpinMultiplicity()
-            if "net charge" in data:
-                self.charge = int(data["net charge"])
-            if "spin multiplicity" in data:
-                self.spin_multiplicity = int(data["spin multiplicity"])
+            if "SEAMM|net charge|int|" in data:
+                self.charge = int(data["SEAMM|net charge|int|"])
+            if "SEAMM|spin multiplicity|int|" in data:
+                self.spin_multiplicity = int(data["SEAMM|spin multiplicity|int|"])
 
         if atoms:
             if any([i != 0.0 for i in qs]):
@@ -199,21 +202,26 @@ def from_OBMol(
         # Record any properties in the database if desired
         if properties == "all":
             for key, value in data.items():
-                if ",units" not in key and key not in [
-                    "net charge",
-                    "spin multiplicity",
-                ]:
-                    if not self.properties.exists(key):
-                        tmp = key.split("#", maxsplit=1)
-                        if len(tmp) > 1:
-                            units_key = tmp[0] + ",units" + "#" + tmp[1]
-                        else:
-                            units_key = key + ",units"
-                        _type = value.__class__.__name__
-                        if units_key in data:
-                            units = data[units_key]
-                            self.properties.add(key, _type, units=units)
+                if key not in (
+                    "SEAMM|net charge|int|",
+                    "SEAMM|spin multiplicity|int|",
+                ):
+                    if key.startswith("SEAMM|"):
+                        _, _property, _type, units = key.split("|", 4)
+                        units = None if units.strip() == "" else units
+                        if not self.properties.exists(_property):
+                            self.properties.add(_property, _type=_type, units=units)
+                        if _type == "int":
+                            value = int(value)
+                        elif _type == "float":
+                            value = float(value)
+                        elif _type == "json":
+                            value = json.dumps(value)
                         else:
+                            pass
+                    else:
+                        if not self.properties.exists(key):
+                            _type = value.__class__.__name__
                             self.properties.add(key, _type)
                     self.properties.put(key, value)
         return self

molsystem/rdkit_.py

@@ -3,6 +3,7 @@
 """Interface to RDKit."""
 
 import logging
+import json
 import pprint
 
 
@@ -97,29 +98,28 @@ def to_RDKMol(self, properties=None):
 
         # Add the net charge and spin multiplicity as properties for configurations
         if self.__class__.__name__ == "_Configuration":
-            rdk_mol.SetIntProp("net charge", self.charge)
-            rdk_mol.SetIntProp("spin multiplicity", self.spin_multiplicity)
+            rdk_mol.SetIntProp("SEAMM|net charge|int|", self.charge)
+            rdk_mol.SetIntProp("SEAMM|spin multiplicity|int|", self.spin_multiplicity)
 
         if properties is not None:
             data = self.properties.get(properties, include_system_properties=True)
-            for key, value in data.items():
+            for _property, value in data.items():
+                _type = self.properties.type(_property)
+                units = self.properties.units(_property)
                 value = value["value"]
-                if isinstance(value, int):
+                key = f"SEAMM|{_property}|{_type}|"
+                if units is not None and units != "":
+                    key += units
+
+                if _type == "int":
                     rdk_mol.SetIntProp(key, value)
-                elif isinstance(value, float):
+                elif _type == "float":
                     rdk_mol.SetDoubleProp(key, value)
+                elif _type == "json":
+                    rdk_mol.SetProp(key, json.dumps(value))
                 else:
                     rdk_mol.SetProp(key, str(value))
 
-                # Units, if any
-                units = self.properties.units(key)
-                if units is not None and units != "":
-                    tmp = key.split("#", maxsplit=1)
-                    if len(tmp) > 1:
-                        rdk_mol.SetProp(tmp[0] + ",units" + "#" + tmp[1], units)
-                    else:
-                        rdk_mol.SetProp(key + ",units", units)
-
         return rdk_mol
 
     def from_RDKMol(
@@ -210,31 +210,34 @@ def from_RDKMol(
             self.spin_multiplicity = n_electrons + 1
 
             # Check for property items for charge and multiplicity
-            if "net charge" in data:
-                self.charge = int(data["net charge"])
-            if "spin multiplicity" in data:
-                self.spin_multiplicity = int(data["spin multiplicity"])
+            if "SEAMM|net charge|int|" in data:
+                self.charge = int(data["SEAMM|net charge|int|"])
+            if "SEAMM|spin multiplicity" in data:
+                self.spin_multiplicity = int(data["SEAMM|spin multiplicity|int|"])
 
         # Record any properties in the database if desired
         if properties == "all":
             for key, value in data.items():
-                if ",units" not in key and key not in [
-                    "net charge",
-                    "spin multiplicity",
-                ]:
-                    if not self.properties.exists(key):
-                        tmp = key.split("#", maxsplit=1)
-                        if len(tmp) > 1:
-                            units_key = tmp[0] + ",units" + "#" + tmp[1]
-                        else:
-                            units_key = key + ",units"
-                        _type = value.__class__.__name__
-                        if units_key in data:
-                            units = data[units_key]
-                            self.properties.add(key, _type, units=units)
-                        else:
+                if ",units" not in key and key not in (
+                    "SEAMM|net charge|int|",
+                    "SEAMM|spin multiplicity|int|",
+                ):
+                    if key.startswith("SEAMM|"):
+                        _, _property, _type, units = key.split("|", 4)
+                        units = None if units.strip() == "" else units
+                        if not self.properties.exists(_property):
+                            self.properties.add(_property, _type=_type, units=units)
+
+                        if _type == "json":
+                            value = json.dumps(value)
+
+                        self.properties.put(key, value)
+                    else:
+                        if not self.properties.exists(key):
+                            _type = value.__class__.__name__
                             self.properties.add(key, _type)
-                    self.properties.put(key, value)
+
+                        self.properties.put(key, value)
         return self
 
     def debug_print(self):

tests/test_openbabel.py

@@ -501,12 +501,11 @@ def test_all_residue_search(configuration):
 def test_to_OBMol(Acetate):
     """Test creating an OBMol object from a structure."""
     correct = {
-        "float property": 3.14,
-        "float property,units": "kcal/mol",
-        "int property": 2,
-        "net charge": -1,
-        "spin multiplicity": 1,
-        "str property": "Hi!",
+        "SEAMM|float property|float|kcal/mol": 3.14,
+        "SEAMM|int property|int|": 2,
+        "SEAMM|net charge|int|": -1,
+        "SEAMM|spin multiplicity|int|": 1,
+        "SEAMM|str property|str|": "Hi!",
     }
 
     mol = Acetate.to_OBMol(properties="*")

tests/test_rdkit.py

@@ -20,12 +20,11 @@ def test_version():
 def test_to_RDKMol(Acetate):
     """Test creating a RDKMol object from a structure."""
     correct = {
-        "float property": 3.14,
-        "float property,units": "kcal/mol",
-        "int property": 2,
-        "net charge": -1,
-        "spin multiplicity": 1,
-        "str property": "Hi!",
+        "SEAMM|float property|float|kcal/mol": 3.14,
+        "SEAMM|int property|int|": 2,
+        "SEAMM|net charge|int|": -1,
+        "SEAMM|spin multiplicity|int|": 1,
+        "SEAMM|str property|str|": "Hi!",
     }
 
     mol = Acetate.to_RDKMol(properties="*")