rdworks 0.49.1__tar.gz → 0.51.1__tar.gz

{rdworks-0.49.1 → rdworks-0.51.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rdworks
-Version: 0.49.1
+Version: 0.51.1
 Summary: Routine tasks built on RDKit and other tools
 Author-email: Sung-Hun Bae <sunghun.bae@gmail.com>
 Maintainer-email: Sung-Hun Bae <sunghun.bae@gmail.com>

{rdworks-0.49.1 → rdworks-0.51.1}/src/rdworks/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '0.49.1'
+__version__ = '0.51.1'
 
 from rdworks.conf    import Conf
 from rdworks.mol     import Mol

{rdworks-0.49.1 → rdworks-0.51.1}/src/rdworks/conf.py

@@ -270,6 +270,62 @@ class Conf:
                 return self.sync(ase_atoms.get_positions())
  
     
+    def protonate(self, atom_indices: list[int]) -> Self:
+        """Protonate given non-hydrogen atoms.
+
+        Args:
+            atom_indices (list[int]): atom indices of non-hydrogen atoms to protonate.
+
+        Returns:
+            Self: self.
+        """
+        for idx in atom_indices:
+            atom = self.rdmol.GetAtomWithIdx(idx)
+            h = atom.GetNumExplicitHs()
+            c = atom.GetFormalCharge()
+            atom.SetNumExplicitHs(h+1)
+            atom.SetFormalCharge(c+1)            
+            Chem.SanitizeMol(self.rdmol)
+            self.rdmol = Chem.AddHs(self.rdmol, addCoords=True)
+            # The Chem.AddHs function in RDKit returns a new Mol object with hydrogens added to the molecule. 
+            # It modifies the input molecule by adding hydrogens, 
+            # but the original molecule remains unchanged.
+        
+        return self
+    
+
+    def deprotonate(self, atom_indices: list[int]) -> Self:
+        """Deprotonate given non-hydrogen atoms.
+
+        Args:
+            atom_indices (list[int]): atom indices of non-hydrogen atoms to deprotonate.
+
+        Returns:
+            Self: self.
+        """
+        for idx in atom_indices:
+            bonded_H_idx = None
+            atom = self.rdmol.GetAtomWithIdx(idx)
+            h = atom.GetNumExplicitHs()
+            if h-1 >= 0 :
+                atom.SetNumExplicitHs(h-1) # (h-1) must be unsigned int
+            c = atom.GetFormalCharge()
+            atom.SetFormalCharge(c-1)            
+            neighbors = atom.GetNeighbors()
+            
+            for neighbor in neighbors:
+                if neighbor.GetAtomicNum() == 1:
+                    bonded_H_idx = neighbor.GetIdx()
+                    break
+
+            if bonded_H_idx is not None:
+                edit_mol = Chem.EditableMol(self.rdmol)
+                edit_mol.RemoveAtom(bonded_H_idx)
+                self.rdmol = edit_mol.GetMol()
+                Chem.SanitizeMol(self.rdmol)
+        
+        return self
+
 
     ##################################################
     ### Endpoint methods

{rdworks-0.49.1 → rdworks-0.51.1}/src/rdworks/mol.py

@@ -1619,4 +1619,46 @@ class Mol:
         self.props = data['props']
         self.confs = [Conf().deserialize(_) for _ in data['confs']] # for 3D conformers (iterable)
         
+        return self
+    
+    def from_molblock(self, molblock: str) -> Self:
+        """Initialize a new Mol object from MolBlock.
+
+        Args:
+            molblock (str): MolBlock string
+
+        Raises:
+            ValueError: invalid MolBlock
+
+        Returns:
+            Self: self.
+        """
+        molecule = Chem.MolFromMolBlock(molblock)
+        try:
+            self.rdmol, _ = clean_2d(molecule, reset_isotope=True, remove_H=True)
+            self.smiles = Chem.MolToSmiles(self.rdmol)
+            self.confs = [Conf(x) for x in _]
+        except:
+            raise ValueError(f'Mol() Error: invalid MolBlock string')
+
+        assert self.smiles and self.rdmol, "Mol() Error: invalid molecule"
+        
+        name = self.rdmol.GetProp('_Name')
+
+        rdDepictor.Compute2DCoords(self.rdmol)
+
+        try:
+            self.name = str(name)
+        except:
+            self.name = 'untitled'
+        
+        self.rdmol.SetProp('_Name', self.name) # _Name can't be None        
+        self.InChIKey = generate_inchi_key(self.rdmol)
+        self.props.update({
+            'aka' : [], # <-- to be set by MolLibr.unique()
+            'atoms' : self.rdmol.GetNumAtoms(), # hydrogens not excluded?
+            'charge': rdmolops.GetFormalCharge(self.rdmol),
+            "nrb" : Descriptors.NumRotatableBonds(self.rdmol),
+            })
+        
         return self

{rdworks-0.49.1 → rdworks-0.51.1}/src/rdworks/xtb/wrapper.py

@@ -59,6 +59,43 @@ class GFN2xTB:
         return shutil.which('xtb') is not None
 
 
+    @staticmethod
+    def is_optimize_ready() -> bool:
+        try:
+            h2o = [
+                '$coord',
+                ' 0.00000000000000      0.00000000000000     -0.73578586109551      o',
+                ' 1.44183152868459      0.00000000000000      0.36789293054775      h',
+                '-1.44183152868459      0.00000000000000      0.36789293054775      h',
+                '$end',
+            ]
+
+            with tempfile.TemporaryDirectory() as temp_dir:
+                test_geometry = os.path.join(temp_dir, 'coord')
+                with open(test_geometry, 'w') as f:
+                    f.write('\n'.join(h2o))
+                proc = subprocess.run(['xtb', test_geometry, '--opt'], 
+                                      capture_output=True, 
+                                      text=True)
+                assert proc.returncode == 0
+
+            return True
+
+        except:
+            print("""                          
+Conda installed xTB has the Fortran runtime error in geometry optimization. 
+Please install xtb using the compiled binary:
+                    
+$ wget https://github.com/grimme-lab/xtb/releases/download/v6.7.1/xtb-6.7.1-linux-x86_64.tar.xz
+$ tar -xf xtb-6.7.1-linux-x86_64.tar.xz
+$ cp -r xtb-dist/bin/*      /usr/local/bin/
+$ cp -r xtb-dist/lib/*      /usr/local/lib/
+$ cp -r xtb-dist/include/*  /usr/local/include/
+$ cp -r xtb-dist/share      /usr/local/ """)
+            
+            return False
+
+
     @staticmethod
     def is_cpx_ready() -> bool:
         """Checks if the CPCM-X command-line tool, `cpx`, is accessible in the system.
@@ -70,7 +107,7 @@ class GFN2xTB:
     
 
     @staticmethod
-    def is_cpcmx_option_ready() -> bool:
+    def is_cpcmx_ready() -> bool:
         """Checks if xtb works with the `--cpcmx` option.
 
         xtb distributed by the conda does not include CPCM-X function (as of June 17, 2025). 
@@ -101,7 +138,8 @@ class GFN2xTB:
         """
         return all([GFN2xTB.is_xtb_ready(),
                     GFN2xTB.is_cpx_ready(),
-                    GFN2xTB.is_cpcmx_option_ready()])
+                    GFN2xTB.is_cpcmx_ready(),
+                    GFN2xTB.is_optimize_ready()])
     
 
     @staticmethod
@@ -115,10 +153,9 @@ class GFN2xTB:
             cmd = ['xtb', '--version']
             proc = subprocess.run(cmd, capture_output=True, text=True)
             assert proc.returncode == 0, "GFN2xTB() Error: xtb not available"
-            for line in proc.stdout.split('\n'):
-                line = line.strip()
-                if 'version' in line:
-                    return line
+            match = re.search('xtb\s+version\s+(?P<version>[\d.]+)', proc.stdout)
+            if match:
+                return match.group('version')
             
         return None
     
@@ -362,7 +399,7 @@ class GFN2xTB:
                 elif water == 'alpb':
                     options += ['--alpb', 'water']
                     # it does not provide Gsolv contribution to the total energy
-                elif water == 'cpcmx' and self.is_cpcmx_option_ready():
+                elif water == 'cpcmx' and self.is_cpcmx_ready():
                     options += ['--cpcmx', 'water']
 
             if verbose:

{rdworks-0.49.1 → rdworks-0.51.1}/src/rdworks.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rdworks
-Version: 0.49.1
+Version: 0.51.1
 Summary: Routine tasks built on RDKit and other tools
 Author-email: Sung-Hun Bae <sunghun.bae@gmail.com>
 Maintainer-email: Sung-Hun Bae <sunghun.bae@gmail.com>

{rdworks-0.49.1 → rdworks-0.51.1}/src/rdworks.egg-info/SOURCES.txt

@@ -76,6 +76,7 @@ src/rdworks/xtb/__init__.py
 src/rdworks/xtb/wrapper.py
 tests/test_basics.py
 tests/test_ionized.py
+tests/test_qupkake.py
 tests/test_round.py
 tests/test_torsion.py
 tests/test_xtb.py

rdworks-0.51.1/tests/test_qupkake.py

@@ -0,0 +1,19 @@
+import rdworks
+
+def test_readin_qupkake_output():
+    libr = rdworks.read_sdf('qupkake/output/qupkake_output.sdf', confs=True)
+    # libr contains one molecule because of confs=True
+    assert libr.count() == 1
+    for m in libr:
+        assert m.count() == 4
+        print(m.props)
+        print()
+        for c in m:
+            print(c.props)
+            print()
+
+        print(m.serialize(compressed=True))
+
+
+if __name__ == '__main__':
+    test_readin_qupkake_output()

rdworks-0.51.1/tests/test_xtb.py

@@ -0,0 +1,76 @@
+from rdworks import Mol
+from rdworks.xtb.wrapper import GFN2xTB
+from rdworks.testdata import drugs
+
+from pathlib import Path
+
+
+# In ASE, the default energy unit is eV (electron volt).
+# It will be converted to kcal/mol
+# CODATA 2018 energy conversion factor
+hartree2ev = 27.211386245988
+hartree2kcalpermol = 627.50947337481
+ev2kcalpermol = 23.060547830619026
+
+
+datadir = Path(__file__).parent.resolve() / "data"
+workdir = Path(__file__).parent.resolve() / "outfiles"
+
+workdir.mkdir(exist_ok=True)
+
+name = 'Atorvastatin'
+testmol = Mol(drugs[name], name).make_confs(n=50).optimize_confs()
+testmol = testmol.drop_confs(similar=True, verbose=True).sort_confs()
+
+
+def test_xtb_wrapper():
+    from rdworks.xtb.wrapper import GFN2xTB
+    assert GFN2xTB.is_xtb_ready() == True
+    assert GFN2xTB.is_cpx_ready() == True
+    assert GFN2xTB.is_cpcmx_ready() == True
+    assert GFN2xTB.is_ready() == True
+    assert GFN2xTB.version() is not None
+
+
+def test_singlepoint():        
+    mol = testmol.copy()
+
+    print("number of conformers=", mol.count())
+    print("number of atoms=", mol.confs[0].natoms)
+
+    gfn2xtb = GFN2xTB(mol.confs[0].rdmol)
+
+    print("GFN2xTB.singlepoint()")
+    outdict = gfn2xtb.singlepoint()
+    print(outdict)
+    print()
+
+    print("GFN2xTB.singlepoint(water='gbsa')")
+    outdict = gfn2xtb.singlepoint(water='gbsa')
+    print(outdict)
+    print()
+
+    print("GFN2xTB.singlepoint(water='alpb')")
+    outdict = gfn2xtb.singlepoint(water='alpb')
+    print(outdict)
+    print()
+
+    print("GFN2xTB.singlepoint(water='cpcmx')")
+    outdict = gfn2xtb.singlepoint(water='cpcmx')
+    print(outdict)
+    print()
+
+
+def test_optimize():
+    mol = testmol.copy()
+    print("number of conformers=", mol.count())
+    print("GFN2xTB.optimize()")
+    outdict = GFN2xTB(mol.confs[0].rdmol).optimize(verbose=True)
+    print(outdict)
+    print()
+
+
+if __name__ == '__main__':
+    test_xtb_wrapper()
+    test_singlepoint()
+    test_optimize()

rdworks-0.49.1/tests/test_xtb.py

@@ -1,160 +0,0 @@
-from rdworks import Mol
-from rdworks.xtb.wrapper import GFN2xTB
-from rdworks.testdata import drugs
-
-from pathlib import Path
-
-
-# In ASE, the default energy unit is eV (electron volt).
-# It will be converted to kcal/mol
-# CODATA 2018 energy conversion factor
-hartree2ev = 27.211386245988
-hartree2kcalpermol = 627.50947337481
-ev2kcalpermol = 23.060547830619026
-
-
-datadir = Path(__file__).parent.resolve() / "data"
-workdir = Path(__file__).parent.resolve() / "outfiles"
-
-workdir.mkdir(exist_ok=True)
-
-name = 'Atorvastatin'
-testmol = Mol(drugs[name], name).make_confs(n=50).optimize_confs()
-testmol = testmol.drop_confs(similar=True, verbose=True).sort_confs()
-
-
-def test_xtb_wrapper():
-    from rdworks.xtb.wrapper import GFN2xTB
-    assert GFN2xTB.is_xtb_ready() == True
-    assert GFN2xTB.is_cpx_ready() == True
-    assert GFN2xTB.is_cpcmx_option_ready() == True
-    assert GFN2xTB.is_ready() == True
-    assert GFN2xTB.version() is not None
-
-
-def test_singlepoint():        
-    mol = testmol.copy()
-
-    print("number of conformers=", mol.count())
-    print("number of atoms=", mol.confs[0].natoms)
-
-    gfn2xtb = GFN2xTB(mol.confs[0].rdmol, ncores=8)
-
-    print("GFN2xTB.singlepoint()")
-    outdict = gfn2xtb.singlepoint()
-    print(outdict)
-    print()
-
-    print("GFN2xTB.singlepoint(water='gbsa')")
-    outdict = gfn2xtb.singlepoint(water='gbsa')
-    print(outdict)
-    print()
-
-    print("GFN2xTB.singlepoint(water='alpb')")
-    outdict = gfn2xtb.singlepoint(water='alpb')
-    print(outdict)
-    print()
-
-    print("GFN2xTB.singlepoint(water='cpcmx')")
-    outdict = gfn2xtb.singlepoint(water='cpcmx')
-    print(outdict)
-    print()
-
-
-def test_optimize():
-    mol = testmol.copy()
-    print("number of conformers=", mol.count())
-    print("GFN2xTB.optimize()")
-    outdict = GFN2xTB(mol.confs[0].rdmol, ncores=8).optimize(verbose=True)
-    print(outdict)
-    print()
-
-
-def test_state_generate():
-    import rdworks
-    import numpy as np
-    import os
-
-    task_queue = 'xtb'
-
-    kT = 0.001987 * 300.0 # (kcal/mol K)
-
-    smiles =  'CCCCNC(=O)[C@@H]1CCCN1[C@@H](CC)c1nnc(Cc2ccc(C)cc2)o1'
-    n = 50
-    method = 'ETKDG'
-
-    standardized = rdworks.Mol(smiles)
-    libr = rdworks.complete_tautomers(standardized)
-
-    PE = []
-    for mol in libr:
-        mol = mol.make_confs(n=n, method=method, verbose=True)
-        mol = mol.optimize_confs(calculator='MMFF94', verbose=True)
-        mol = mol.drop_confs(similar=True, similar_rmsd=0.3, verbose=True)
-        mol = mol.sort_confs(calculator='xTB', verbose=True)
-        mol = mol.drop_confs(k=10, window=10.0, verbose=True) # enforcing both conditions
-        _PE = []
-        for conf in mol.confs:
-            conf = conf.optimize(calculator='xTB', verbose=True)
-            # GFN2xTB requires 3D coordinates
-            xtb = GFN2xTB(conf.rdmol).singlepoint(water='cpcmx', verbose=True)
-            _PE.append(xtb.PE)
-        # SimpleNamespace(
-        #             PE = datadict['total energy'] * hartree2kcalpermol,
-        #             Gsolv = Gsolv,
-        #             charges = datadict['partial charges'],
-        #             wbo = Wiberg_bond_orders,
-        #             )
-        PE.append(_PE)
-        print(_PE)
-    
-    # calculate population
-    PE = np.array(PE)
-    PE = PE - np.min(PE)
-    Boltzmann_factors = np.exp(-PE/kT)
-    # partition function
-    Z = np.sum(Boltzmann_factors)
-    # population
-    p = np.sum(Boltzmann_factors/Z, axis=1)
-
-    sorted_indices = sorted(list(enumerate(p)), key=lambda x: x[1], reverse=True) # [(0,p0), (1,p1), ...]
-
-    molecular_states = []
-    for idx, population in sorted_indices:
-        if population < 0.05:
-            continue
-
-        # state.keys() = ['rdmol','smiles','charge','population','pKa', 'qikprop']
-
-        state_mol = libr[idx].rename(f'state.{idx+1}').qed(
-            properties=['QED', 'MolWt', 'LogP', 'TPSA', 'HBD', 'HBA'])
-        
-        basic_properties = {
-            'QED'      : round(state_mol.props['QED'],   2),
-            'MolWt'    : round(state_mol.props['MolWt'], 2),
-            'LogP'     : round(state_mol.props['LogP'],  2),
-            'TPSA'     : round(state_mol.props['TPSA'],  2),
-            'HBD'      : state_mol.props['HBD'],
-            'HBA'      : state_mol.props['HBA'],
-            }
-        
-        state_props = {
-            'method': task_queue,
-            'PE(kcal/mol)': state_mol.confs[0].props['E_tot(kcal/mol)'],
-            'population' : round(float(population), 3),
-            'basic_properties': basic_properties,
-            'rdkit_version': rdworks.__rdkit_version__,
-            'rdworks_version': rdworks.__version__,
-        }
-        
-        molecular_states.append((state_mol.serialize(compressed=True), state_props))
-    
-    print(molecular_states)
-
-
-
-if __name__ == '__main__':
-    # test_xtb_wrapper()
-    # test_singlepoint()
-    # test_optimize()
-    test_state_generate()