rdworks 0.48.1__tar.gz → 0.50.1__tar.gz

{rdworks-0.48.1 → rdworks-0.50.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rdworks
-Version: 0.48.1
+Version: 0.50.1
 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-0.48.1 → rdworks-0.50.1}/src/rdworks/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '0.48.1'
+__version__ = '0.50.1'
 
 from rdworks.conf    import Conf
 from rdworks.mol     import Mol
@@ -25,10 +25,11 @@ __rdkit_version__ = rdkit.rdBase.rdkitVersion
 
 rdkit_logger = rdkit.RDLogger.logger().setLevel(rdkit.RDLogger.CRITICAL)
 
-main_logger = logging.getLogger()
-main_logger.setLevel(logging.INFO) # level: DEBUG < INFO < WARNING < ERROR < CRITICAL
-logger_formatter = logging.Formatter(fmt='%(asctime)s %(levelname)s %(message)s', 
-                                     datefmt='%Y-%m-%d %H:%M:%S')
-logger_ch = logging.StreamHandler()
-logger_ch.setFormatter(logger_formatter)
-main_logger.addHandler(logger_ch)
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.INFO) # level: DEBUG < INFO < WARNING < ERROR < CRITICAL
+
+logger_stream = logging.StreamHandler() # sys.stdout or sys.stderr
+logger_format = logging.Formatter(fmt='%(asctime)s %(levelname)s %(message)s', 
+                                  datefmt='%Y-%m-%d %H:%M:%S')
+logger_stream.setFormatter(logger_format)
+logger.addHandler(logger_stream)

{rdworks-0.48.1 → rdworks-0.50.1}/src/rdworks/ionized.py

@@ -130,6 +130,9 @@ class IonizedStates:
                     edmol.RemoveAtom(bonded_H_indices[0])
             
             elif acid_or_base == 'B': # protonate
+                # note that protonation at tertiary nitrogen may results in stereoisomers
+                # current implementation ignores the stereochemistry
+                # use rdworks.complete_stereoisomers() function to complete the stereoisomers
                 B = edmol.GetAtomWithIdx(i)
                 assert B.GetAtomicNum() > 1, f"Cannot protonate an atom (idx={i}; {B.GetAtomicNum()})"
                 charge = B.GetFormalCharge() + 1
@@ -138,7 +141,7 @@ class IonizedStates:
                 B.SetNumExplicitHs(nH+1)
                 edmol = Chem.AddHs(edmol)
             
-            # Clean up and save SMILES
+            # clean up and save SMILES
             ionized_smiles = IonizedStates.clean_smiles(edmol)
             ionized_mol = Chem.MolFromSmiles(ionized_smiles)
             ionized_mol = Chem.AddHs(ionized_mol)

{rdworks-0.48.1 → rdworks-0.50.1}/src/rdworks/mol.py

@@ -46,7 +46,8 @@ from rdworks.display import render_svg, render_png
 
 from scour.scour import scourString
 
-main_logger = logging.getLogger()
+
+logger = logging.getLogger(__name__)
 
 
 class Mol:
@@ -328,11 +329,7 @@ class Mol:
         return self
 
 
-    def make_confs(self, 
-                   n:int = 50, 
-                   method:str = 'ETKDG', 
-                   calculator:str | Callable = 'MMFF94',
-                   ) -> Self:
+    def make_confs(self, n: int = 50, method: str = 'ETKDG', **kwargs) -> Self:
         """Generates 3D conformers.
 
         Args:
@@ -348,12 +345,8 @@ class Mol:
             High-Quality Conformer Generation with CONFORGE: Algorithm and Performance Assessment. 
             J. Chem. Inf. Model. 63, 5549-5570 (2023).
         """
-        
-        # if n is None:
-        #     rot_bonds = rd_descriptor_f['RotBonds'](self.rdmol)
-        #     n = min(max(1, int(8.481 * (rot_bonds **1.642))), 1000)
-        # n = max(1, math.ceil(n * n_rel)) # ensures that n is at least 1
-        
+        verbose = kwargs.get('verbose', False)
+
         self.confs = []
 
         if method.upper() == 'ETKDG':
@@ -426,6 +419,7 @@ class Mol:
             os.remove(tmp_filename)
 
         # energy evaluations for ranking
+        calculator = kwargs.get('calculator', 'MMFF94')
         for conf in self.confs:
             conf.potential_energy(calculator) # default: MMFF94
         
@@ -436,8 +430,15 @@ class Mol:
         for conf in self.confs:
             conf.props.update({"E_rel(kcal/mol)": conf.props[sort_by] - lowest_energy})
 
+        # rename conformers
         self = self.rename()
 
+        if verbose:
+            rot_bonds = rd_descriptor_f['RotBonds'](self.rdmol)
+            nrb_suggested = int(8.481 * (rot_bonds **1.642))
+            logger.info(f"make_confs() rotatable bonds {rot_bonds} (suggested conformers {nrb_suggested}) generated {self.count()}")
+            logger.info(f"make_confs() updated potential energies E_tot(kcal/mol) and E_rel(kcal/mol) by {calculator}")
+
         return self
 
 
@@ -499,10 +500,15 @@ class Mol:
         Returns:
             Self: modified self.
         """
+        verbose = kwargs.get('verbose', False)
+
         if calculator is not None:
             # re-calculate potential energies
+            if verbose :
+                logger.info(f"sort_cons() calculate potential energy by {calculator}")
+
             for conf in self.confs:
-                PE = conf.potential_energy(calculator=calculator, **kwargs) # sets `E_tot(kcal/mol)`
+                PE = conf.potential_energy(calculator, **kwargs) # sets `E_tot(kcal/mol)`
 
         if all(['E_tot(kcal/mol)' in conf.props for conf in self.confs]):
             sort_by = 'E_tot(kcal/mol)'
@@ -767,7 +773,7 @@ class Mol:
                    cluster: bool | None =None,
                    k: int | None = None,
                    window: float | None = None,
-                   verbose: bool = False) -> Self:
+                   **kwargs) -> Self:
         """Drop conformers that meet some condition(s).
 
         Args:
@@ -792,6 +798,8 @@ class Mol:
             Self: modified self.    
         """
 
+        verbose = kwargs.get('verbose', False)
+
         reasons = [f'stereo flipped',  
                    f'unconverged', 
                    f'similar({similar_rmsd})',
@@ -805,13 +813,13 @@ class Mol:
             mask = [Chem.MolToSmiles(Chem.RemoveHs(_.rdmol)) == self.smiles for _ in self.confs]
             self.confs = list(itertools.compress(self.confs, mask))
             if verbose:
-                main_logger.info(f'drop_confs() {mask.count(False):3d} {reasons[0]:<{w}} -> {self.count()}')
+                logger.info(f'drop_confs() {mask.count(False):3d} {reasons[0]:<{w}} -> {self.count()}')
         
         if unconverged and self.count() > 0:
             mask = [_.props['Converged'] if 'Converged' in _.props else True for _ in self.confs]
             self.confs = list(itertools.compress(self.confs, mask))
             if verbose:
-                main_logger.info(f'drop_confs() {mask.count(False):3d} {reasons[1]:<{w}} -> {self.count()}')
+                logger.info(f'drop_confs() {mask.count(False):3d} {reasons[1]:<{w}} -> {self.count()}')
 
         if similar and self.count() > 1:
             # it is observed that there are essentially identical conformers 
@@ -831,7 +839,7 @@ class Mol:
             mask = [conf_idx in centroid_indices for conf_idx, conf in enumerate(self.confs)]
             self.confs = list(itertools.compress(self.confs, mask))
             if verbose:
-                main_logger.info(f'drop_confs() {mask.count(False):3d} {reasons[2]:<{w}} -> {self.count()}')
+                logger.info(f'drop_confs() {mask.count(False):3d} {reasons[2]:<{w}} -> {self.count()}')
 
             # note: it will retain the conformers with lower index
             # so, it should be sorted before dropping
@@ -856,9 +864,12 @@ class Mol:
             mask = [_.props['centroid'] if 'centroid' in _.props else True for _ in self.confs]
             self.confs = list(itertools.compress(self.confs, mask))
             if verbose:
-                main_logger.info(f'drop_confs() {mask.count(False):3d} {reasons[3]:<{w}} -> {self.count()}')
+                logger.info(f'drop_confs() {mask.count(False):3d} {reasons[3]:<{w}} -> {self.count()}')
 
         if (k or window) and self.count() > 0:
+            # confs must be sorted by energies
+            if not all(['E_rel(kcal/mol)' in _.props for _ in self.confs]):
+                self = self.sort_confs(**kwargs)
             if k:
                 mask_k = [i < k for i,_ in enumerate(self.confs)]
             else:
@@ -871,7 +882,7 @@ class Mol:
             mask = [(x and y) for (x,y) in zip(mask_k, mask_window)]
             self.confs = list(itertools.compress(self.confs, mask))
             if verbose:
-                main_logger.info(f'drop_confs() {mask.count(False):3d} {reasons[4]:<{w}} -> {self.count()}')
+                logger.info(f'drop_confs() {mask.count(False):3d} {reasons[4]:<{w}} -> {self.count()}')
         
         return self
 
@@ -974,7 +985,7 @@ class Mol:
     
 
     def torsion_energies(self,
-                         calculator: str | Callable,
+                         calculator: str | Callable = 'MMFF94',
                          torsion_key: int | None = None,
                          simplify: bool = True, 
                          fmax: float = 0.05,

{rdworks-0.48.1 → rdworks-0.50.1}/src/rdworks/xtb/wrapper.py

@@ -14,7 +14,8 @@ from rdkit import Chem
 from rdkit.Geometry import Point3D
 
 
-logger = logging.getLogger()
+logger = logging.getLogger(__name__)
+
 
 # In ASE, the default energy unit is eV (electron volt).
 # It will be converted to kcal/mol
@@ -25,7 +26,7 @@ ev2kcalpermol = 23.060547830619026
 
 
 class GFN2xTB:
-    def __init__(self, molecule: Chem.Mol, ncores: int = 8):
+    def __init__(self, molecule: Chem.Mol, ncores: int | None = None):
         assert isinstance(molecule, Chem.Mol), "molecule is not rdkit.Chem.Mol type"
         assert molecule.GetConformer().Is3D(), "molecule is not a 3D conformer"
         assert self.is_xtb_ready(), "xtb is not accessible"
@@ -35,6 +36,9 @@ class GFN2xTB:
         self.symbols = [ atom.GetSymbol() for atom in molecule.GetAtoms() ]
         self.positions = molecule.GetConformer().GetPositions().tolist()
 
+        if ncores is None:
+            ncores = os.cpu_count()
+
         # Parallelisation
         os.environ['OMP_STACKSIZE'] = '4G'
         os.environ['OMP_NUM_THREADS'] = f'{ncores},1'
@@ -55,6 +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.
@@ -66,7 +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). 
@@ -97,7 +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
@@ -107,14 +149,13 @@ class GFN2xTB:
         Returns:
             str | None: version statement.
         """
-        if GFN2xTB.is_ready():
+        if GFN2xTB.is_xtb_ready():
             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
     
@@ -338,12 +379,9 @@ class GFN2xTB:
 
         with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
             workdir = Path(temp_dir)
-            if verbose:
-                logger.info(f'xtb.singlepoint workdir= {temp_dir}')
             
             geometry_input_path = workdir / 'geometry.xyz'
             xtbout_path = workdir / 'xtbout.json'
-            stdout_path = workdir / 'fort.6'
             wbo_path = workdir / 'wbo'
             geometry_output_path = workdir / 'xtbtopo.mol'
             
@@ -354,8 +392,6 @@ class GFN2xTB:
             
             options = ['--gfn', '2', '--json']
             
-            Gsolv = None
-            
             if water is not None and isinstance(water, str):
                 if water == 'gbsa':
                     options += ['--gbsa', 'H2O']
@@ -363,9 +399,12 @@ 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:
+                logger.info(f"singlepoint() {' '.join(cmd+options)}")
+
             # 'xtbout.json', 'xtbrestart', 'xtbtopo.mol', 'charges', and 'wbo' files will be 
             # created in the current working directory.
             proc = subprocess.run(cmd + options, cwd=temp_dir, capture_output=True, text=True)
@@ -381,19 +420,20 @@ class GFN2xTB:
                 if xtbout_path.is_file():
                     with open(xtbout_path, 'r') as f:
                         datadict = json.load(f) # takes the file object as input
+
+                Gsolv = None
                                  
-                if (water is not None) and (Gsolv is None) and stdout_path.is_file():
+                if water is not None:
                     #  Free Energy contributions:                       [Eh]        [kcal/mol]
                     # -------------------------------------------------------------------------
                     #  solvation free energy (dG_solv):             -0.92587E-03    -0.58099
                     #  gas phase energy (E)                         -0.52068E+01
                     # -------------------------------------------------------------------------
                     #  total free energy (dG)                       -0.52077E+01
-                    with open(stdout_path, 'r') as f:
-                        for line in f:
-                            if 'solvation free energy' in line:
-                                m = re.search(r"solvation free energy \(dG_solv\)\:\s+[-+]?\d*\.?\d+E[-+]?\d*\s+(?P<kcalpermol>[-+]?\d*\.?\d+)", line)
-                                Gsolv = float(m.group('kcalpermol'))
+                    for line in proc.stdout.splitlines():
+                        if 'solvation free energy' in line:
+                            m = re.search(r"solvation free energy \(dG_solv\)\:\s+[-+]?\d*\.?\d+E[-+]?\d*\s+(?P<kcalpermol>[-+]?\d*\.?\d+)", line)
+                            Gsolv = float(m.group('kcalpermol'))
                 
                 Wiberg_bond_orders = self.load_wbo(wbo_path)
 
@@ -429,8 +469,6 @@ class GFN2xTB:
         """
         with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
             workdir = Path(temp_dir)
-            if verbose:
-                logger.info(f'xtb.optimize workdir= {temp_dir}')
             
             geometry_input_path = workdir / 'geometry.xyz'
             xtbout_path = workdir / 'xtbout.json'
@@ -452,6 +490,9 @@ class GFN2xTB:
                 elif water == 'cpcmx':
                     logger.warning('optimize with --cpcmx option is not implemented in xtb yet')
 
+            if verbose:
+                logger.info(f"optimize() {' '.join(cmd+options)}")
+
             proc = subprocess.run(cmd + options, cwd=temp_dir, capture_output=True, text=True)
 
             if proc.returncode == 0 and xtbout_path.is_file():
@@ -523,4 +564,4 @@ class GFN2xTB:
                 with open(xtb_esp_dat, 'r') as f:
                     pass
         
-        return None
+        return None

{rdworks-0.48.1 → rdworks-0.50.1}/src/rdworks.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rdworks
-Version: 0.48.1
+Version: 0.50.1
 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-0.48.1 → rdworks-0.50.1}/tests/test_ionized.py

@@ -13,6 +13,12 @@ def test_ionizedstate():
         print(k, v)
     print()
 
+    p = x.get_pairs()
+    print('pairs:')
+    for k in p:
+        print(k)
+    print()
+
     indices = d['CCCCNC(=O)[C@@H]1CCCN1[C@@H](CC)c1nnc(Cc2ccc(C)cc2)o1'][0]
     
     assert (11, 'B') in indices
@@ -30,5 +36,31 @@ def test_ionizedstate():
     assert set(expected).intersection(set(results)) == set(expected)
 
 
+def test_gypsum_dl():
+    import gypsum_dl
+    smiles = 'O=C(NCCCC)[C@H](CCC1)N1[C@@H](CC)C2=NN=C(CC3=CC=C(C)C=C3)O2'
+    state_smiles = list(
+        gypsum_dl.GypsumDL(smiles,
+            min_ph=6.4,
+            max_ph=8.4,
+            pka_precision=1.0,
+            thoroughness=3,
+            max_variants_per_compound=5,
+            second_embed=False,
+            skip_optimize_geometry=False,
+            skip_alternate_ring_conformations=False,
+            skip_adding_hydrogen=False,
+            skip_making_tautomers=False,
+            skip_enumerate_chiral_mol=False,
+            skip_enumerate_double_bonds=False,
+            let_tautomers_change_chirality=False,
+            use_durrant_lab_filters=True,
+            job_manager='serial',
+            num_processors=1,
+            ))
+    for smi in state_smiles:
+        print(smi)
+    
 if __name__ == '__main__':
-    test_ionizedstate()
+    test_ionizedstate()
+    test_gypsum_dl()

rdworks-0.50.1/tests/test_xtb.py

@@ -0,0 +1,160 @@
+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()
+
+
+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=15.0, 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()

rdworks-0.48.1/tests/test_xtb.py

@@ -1,76 +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()
-
-
-if __name__ == '__main__':
-    test_xtb_wrapper()
-    test_singlepoint()
-    test_optimize()