rdworks 0.51.1__tar.gz → 0.52.1__tar.gz

{rdworks-0.51.1 → rdworks-0.52.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rdworks
-Version: 0.51.1
+Version: 0.52.1
 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-0.51.1 → rdworks-0.52.1}/src/rdworks/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '0.51.1'
+__version__ = '0.52.1'
 
 from rdworks.conf    import Conf
 from rdworks.mol     import Mol

{rdworks-0.51.1 → rdworks-0.52.1}/src/rdworks/conf.py

@@ -355,6 +355,15 @@ class Conf:
         return True
     
 
+    def charge(self) -> int:
+        """Returns molecular formal charge.
+
+        Returns:
+            int: charge
+        """
+        return self.charge
+    
+
     def positions(self) -> np.array:
         """Returns the coordinates.
 

rdworks-0.52.1/src/rdworks/microstates.py

@@ -0,0 +1,138 @@
+import numpy as np
+import math
+import itertools
+
+from types import SimpleNamespace
+from rdworks import Conf, Mol
+from rdworks.xtb.wrapper import GFN2xTB
+
+
+kT = 0.001987 * 298.0 # (kcal/mol K), standard condition
+C = math.log(10) * kT
+
+
+
+class Microstates():
+
+    def __init__(self, origin: Mol, calculator: str = 'xTB'):
+        self.origin = origin
+        self.calculator = calculator
+        self.basic_sites = []
+        self.acidic_sites = []
+        self.states = []
+        self.mols = []
+        self.reference = None
+    
+        # read QupKake results
+        for conf in self.origin:
+            print(conf.props)
+            pka = conf.props.get('pka', None)
+            if pka is None:
+                # no protonation/deprotonation sites
+                continue
+            if isinstance(pka, str) and pka.startswith('tensor'):
+                # ex. 'tensor(9.5784)'
+                pka = float(pka.replace('tensor(','').replace(')',''))
+            if conf.props.get('pka_type') == 'basic':
+                self.basic_sites.append(conf.props.get('idx'))
+            elif conf.props.get('pka_type') == 'acidic':
+                self.acidic_sites.append(conf.props.get('idx'))
+
+        # enumerate protonation/deprotonation sites to generate microstates
+
+        np = len(self.basic_sites)
+        nd = len(self.acidic_sites)
+        P = [c for n in range(np+1) for c in itertools.combinations(self.basic_sites, n)]
+        D = [c for n in range(nd+1) for c in itertools.combinations(self.acidic_sites, n)]
+        
+        PD = list(itertools.product(P, D))
+        
+        for (p, d) in PD:
+            conf = self.origin.confs[0].copy()
+            conf = conf.protonate(p).deprotonate(d).optimize(calculator=calculator)
+            charge = len(p) - len(d)
+            self.states.append(SimpleNamespace(
+                charge=charge, 
+                protonation_sites=p, 
+                deprotonation_sites=d,
+                conf=conf,
+                smiles=Mol(conf).smiles,
+                delta_m=None,
+                PE=None))
+            
+        # sort microstates by ascending charges
+        self.states = sorted(self.states, key=lambda x: x.charge)
+
+
+    @staticmethod
+    def Boltzmann_weighted_average(potential_energies: list) -> float:
+        """Calculate Boltzmann weighted average potential energy at pH 0.
+
+        Args:
+            potential_energies (list): a list of potential energies.
+
+        Returns:
+            float: Boltzmann weighted average potential energy.
+        """
+        pe_array = np.array(potential_energies)
+        pe = pe_array - min(potential_energies)
+        Boltzmann_factors = np.exp(-pe/kT)
+        Z = np.sum(Boltzmann_factors)
+        p = Boltzmann_factors/Z
+
+        return float(np.dot(p, pe_array))
+
+
+    def potential_energy(self) -> None:
+        for microstate in self.states:
+            mol = Mol(microstate.conf).make_confs(n=4).optimize_confs()
+            # mol = mol.drop_confs(similar=True, similar_rmsd=0.3, verbose=True)
+            # mol = mol.optimize_confs(calculator=calculator)
+            # mol = mol.drop_confs(k=10, window=15.0, verbose=True)
+            PE = []
+            for conf in mol.confs:
+                conf = conf.optimize(calculator=self.calculator, verbose=True)
+                # GFN2xTB requires 3D coordinates
+                # xtb = GFN2xTB(conf.rdmol).singlepoint(water='cpcmx', verbose=True)
+                xtb = GFN2xTB(conf.rdmol).singlepoint(verbose=True)
+                PE.append(xtb.PE)
+                # SimpleNamespace(
+                #             PE = datadict['total energy'] * hartree2kcalpermol,
+                #             Gsolv = Gsolv,
+                #             charges = datadict['partial charges'],
+                #             wbo = Wiberg_bond_orders,
+                #             )
+            print("PE=", PE)
+            microstate.PE = self.Boltzmann_weighted_average(PE)
+            print("Boltzmann weighted=", microstate.PE)
+            
+            self.mols.append(mol)
+            print("microstate.energy", microstate)
+
+
+    def populations(self, pH: float) -> list[tuple]:
+        # set the lowest dG as the reference
+        self.reference = self.states[np.argmin([microstate.PE for microstate in self.states])]
+        for microstate in self.states:
+            microstate.delta_m = microstate.charge - self.reference.charge
+        dG = []
+        for microstate in self.states:
+            dG.append((microstate.PE - self.reference.PE) + microstate.delta_m * C * pH)
+        dG = np.array(dG)
+
+        print("dG=", dG)
+        Boltzmann_factors = np.exp(-dG/kT)
+        Z = np.sum(Boltzmann_factors)
+        p = Boltzmann_factors/Z
+        idx_p = sorted(list(enumerate(p)), key=lambda x: x[1], reverse=True)
+        # [(0, p0), (1, p1), ...]
+
+        return idx_p
+    
+
+    def count(self) -> int:
+        return len(self.states)
+    
+
+    def get_mol(self, idx: int) -> Mol:
+        return self.mols[idx]

{rdworks-0.51.1 → rdworks-0.52.1}/src/rdworks/readin.py

@@ -232,6 +232,7 @@ def read_sdf(path:str | Path, std:bool=False, confs:bool=False, props:bool=True,
                 # start a new molecule
                 rdmol_2d = Chem.RemoveHs(rdmol)
                 AllChem.Compute2DCoords(rdmol_2d)
+                # initialize a new molecule with the H-removed 2D
                 new_mol = Mol(rdmol_2d, isomer_name, std=False) # atom indices remain unchanged.
             new_conf = Conf(rdmol)
             new_conf.props.update(props)

{rdworks-0.51.1 → rdworks-0.52.1}/src/rdworks/xtb/wrapper.py

@@ -11,6 +11,7 @@ from pathlib import Path
 from types import SimpleNamespace
 
 from rdkit import Chem
+from rdkit.Chem import rdmolops
 from rdkit.Geometry import Point3D
 
 
@@ -32,6 +33,7 @@ class GFN2xTB:
         assert self.is_xtb_ready(), "xtb is not accessible"
 
         self.rdmol = molecule
+        self.charge = rdmolops.GetFormalCharge(self.rdmol)
         self.natoms = molecule.GetNumAtoms()
         self.symbols = [ atom.GetSymbol() for atom in molecule.GetAtoms() ]
         self.positions = molecule.GetConformer().GetPositions().tolist()
@@ -74,7 +76,8 @@ class GFN2xTB:
                 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'], 
+                proc = subprocess.run(['xtb', test_geometry, '--opt'],
+                                      cwd=temp_dir,
                                       capture_output=True, 
                                       text=True)
                 assert proc.returncode == 0
@@ -117,14 +120,15 @@ $ cp -r xtb-dist/share      /usr/local/ """)
             bool: True if the --cpcmx option is working, False otherwise.
         """
         if GFN2xTB.is_xtb_ready():
-            cmd = ['xtb', '--cpcmx']
-            proc = subprocess.run(cmd, capture_output=True, text=True)
-            # we are expecting an error because no input file is given
-            assert proc.returncode != 0
-            for line in proc.stdout.split('\n'):
-                line = line.strip()
-                if 'CPCM-X library was not included' in line:
-                    return False
+            with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+                cmd = ['xtb', '--cpcmx']
+                proc = subprocess.run(cmd, cwd=temp_dir, capture_output=True, text=True)
+                # we are expecting an error because no input file is given
+                assert proc.returncode != 0
+                for line in proc.stdout.split('\n'):
+                    line = line.strip()
+                    if 'CPCM-X library was not included' in line:
+                        return False
         
         return True
 
@@ -150,12 +154,13 @@ $ cp -r xtb-dist/share      /usr/local/ """)
             str | None: version statement.
         """
         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"
-            match = re.search('xtb\s+version\s+(?P<version>[\d.]+)', proc.stdout)
-            if match:
-                return match.group('version')
+            with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+                cmd = ['xtb', '--version']
+                proc = subprocess.run(cmd, cwd=temp_dir, capture_output=True, text=True)
+                assert proc.returncode == 0, "GFN2xTB() Error: xtb not available"
+                match = re.search('xtb\s+version\s+(?P<version>[\d.]+)', proc.stdout)
+                if match:
+                    return match.group('version')
             
         return None
     
@@ -265,109 +270,56 @@ $ cp -r xtb-dist/share      /usr/local/ """)
             return Wiberg_bond_orders
 
 
-    def cpx(self, verbose: bool = False) -> float | None:
-        """Runs cpx and returns Gsolv (kcal/mol)
-
-        Warning: 
-            Solvation energy obtained from `xtb --cpcmx water` differs from
-            `cpx --solvent water` (difference between gas.out and solv.out in terms of total energy).
-            There are other correction terms not clearly defined in the output files.
-            So, this method is not reliable and should be discarded
-
-        Returns:
-            float or None: Gsolv energy in kcal/mol or None.
-        """
-        with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
-            workdir = Path(temp_dir)
-            if verbose:
-                logger.info(f'xtb.cpx workdir= {temp_dir}')
-            
-            geometry_input_path = workdir / 'coord'
-            geometry_output_path = workdir / 'xtbtopo.mol'
-            gas_out_path = workdir / 'gas.out'
-            solv_out_path = workdir / 'solv.out'
-            wbo_path = workdir / 'wbo'
-         
-            with open(geometry_input_path, 'w') as f:
-                f.write(self.to_turbomole_coord())
-        
-            cmd = ['cpx']
-            options = ['--solvent', 'water']
-            
-            proc = subprocess.run(cmd + options, cwd=temp_dir, capture_output=True, text=True)
-            # cpx creates the following files:
-            # charges  gas.energy    solute_sigma.txt   solvent_sigma.txt   xtbtopo.mol
-            # coord    gas.out       solute_sigma3.txt  solvent_sigma3.txt
-            # error    solute.cosmo  solv.out           wbo
-
-            # example of solv.out
-            #  :::::::::::::::::::::::::::::::::::::::::::::::::::::
-            #  ::                     SUMMARY                     ::
-            #  :::::::::::::::::::::::::::::::::::::::::::::::::::::
-            #  :: total energy            -119.507131639760 Eh    ::
-            #  :: w/o Gsasa/hb/shift      -119.494560363045 Eh    ::
-            #  :: gradient norm              0.084154442395 Eh/a0 ::
-            #  :: HOMO-LUMO gap              2.966157362876 eV    ::
-            #  ::.................................................::
-            #  :: SCC energy              -121.121278922798 Eh    ::
-            #  :: -> isotropic ES            0.180705208303 Eh    ::
-            #  :: -> anisotropic ES          0.003924951393 Eh    ::
-            #  :: -> anisotropic XC          0.040710819025 Eh    ::
-            #  :: -> dispersion             -0.088336282215 Eh    ::
-            #  :: -> Gsolv                  -0.039236762590 Eh    ::
-            #  ::    -> Gelec               -0.026665485874 Eh    ::
-            #  ::    -> Gsasa               -0.012571276716 Eh    ::
-            #  ::    -> Ghb                  0.000000000000 Eh    ::
-            #  ::    -> Gshift               0.000000000000 Eh    ::
-            #  :: repulsion energy           1.614147283037 Eh    ::
-            #  :: add. restraining           0.000000000000 Eh    ::
-            #  :: total charge              -0.000000000000 e     ::
-            #  :::::::::::::::::::::::::::::::::::::::::::::::::::::
-
-            # example gas.out
-            # :::::::::::::::::::::::::::::::::::::::::::::::::::::
-            #  ::                     SUMMARY                     ::
-            #  :::::::::::::::::::::::::::::::::::::::::::::::::::::
-            #  :: total energy            -119.473726280382 Eh    ::
-            #  :: gradient norm              0.085445002241 Eh/a0 ::
-            #  :: HOMO-LUMO gap              2.562893747102 eV    ::
-            #  ::.................................................::
-            #  :: SCC energy              -121.087873563419 Eh    ::
-            #  :: -> isotropic ES            0.152557320965 Eh    ::
-            #  :: -> anisotropic ES          0.007343156635 Eh    ::
-            #  :: -> anisotropic XC          0.039625076440 Eh    ::
-            #  :: -> dispersion             -0.088605122696 Eh    ::
-            #  :: repulsion energy           1.614147283037 Eh    ::
-            #  :: add. restraining           0.000000000000 Eh    ::
-            #  :: total charge              -0.000000000000 e     ::
-            #  :::::::::::::::::::::::::::::::::::::::::::::::::::::
-
-            if proc.returncode == 0:
-                total_energy_solv = None
-                total_energy_gas = None
-                
-                with open(solv_out_path, 'r') as f:
-                    for line in f:
-                        if 'total energy' in line:
-                            m = re.search(r"total energy\s+(?P<solv>[-+]?\d*\.?\d+)\s+Eh", line)
-                            total_energy_solv = float(m.group('solv'))
-                with open(gas_out_path, 'r') as f:
-                    for line in f:
-                        if 'total energy' in line:
-                            m = re.search(r"total energy\s+(?P<gas>[-+]?\d*.?\d+)\s+Eh", line)
-                            total_energy_gas = float(m.group('gas'))
-                
-                if total_energy_solv and total_energy_gas:
-                    return (total_energy_solv - total_energy_gas) * hartree2kcalpermol
-        
-        return None
-
-
     def singlepoint(self, water: str | None = None, verbose: bool = False) -> SimpleNamespace:
         """Calculate single point energy.
         
         Total energy from xtb output in atomic units (Eh, hartree) is converted to kcal/mol.
 
+        Options:
+            ```sh
+            -c, --chrg INT
+                specify molecular charge as INT, overrides .CHRG file and xcontrol option
+            
+            --scc, --sp
+                performs a single point calculation
+            
+            --gfn INT
+                specify parametrisation of GFN-xTB (default = 2)
+            
+            --json
+                write xtbout.json file
+            
+            --alpb SOLVENT [STATE]
+                analytical linearized Poisson-Boltzmann (ALPB) model,
+                available solvents are acetone, acetonitrile, aniline, benzaldehyde,
+                benzene, ch2cl2, chcl3, cs2, dioxane, dmf, dmso, ether, ethylacetate, furane,
+                hexandecane, hexane, methanol, nitromethane, octanol, woctanol, phenol, toluene,
+                thf, water.
+                The solvent input is not case-sensitive. The Gsolv
+                reference state can be chosen as reference, bar1M, or gsolv (default).
+
+            -g, --gbsa SOLVENT [STATE]
+                generalized born (GB) model with solvent accessable surface (SASA) model,
+                available solvents are acetone, acetonitrile, benzene (only GFN1-xTB), CH2Cl2,
+                CHCl3, CS2, DMF (only GFN2-xTB), DMSO, ether, H2O, methanol,
+                n-hexane (only GFN2-xTB), THF and toluene.
+                The solvent input is not case-sensitive.
+                The Gsolv reference state can be chosen as reference, bar1M, or gsolv (default).
+
+            --cosmo SOLVENT/EPSILON
+                domain decomposition conductor-like screening model (ddCOSMO),
+                available solvents are all solvents that are available for alpb.
+                Additionally, the dielectric constant can be set manually or an ideal conductor
+                can be chosen by setting epsilon to infinity.
+
+            --tmcosmo SOLVENT/EPSILON
+                same as --cosmo, but uses TM convention for writing the .cosmo files.
+
+            --cpcmx SOLVENT
+                extended conduction-like polarizable continuum solvation model (CPCM-X),
+                available solvents are all solvents included in the Minnesota Solvation Database.
+            ```
+
         Args:
             water (str, optional) : water solvation model (choose 'gbsa' or 'alpb')
                 alpb: ALPB solvation model (Analytical Linearized Poisson-Boltzmann).
@@ -389,8 +341,7 @@ $ cp -r xtb-dist/share      /usr/local/ """)
                 geometry.write(self.to_xyz())
             
             cmd = ['xtb', geometry_input_path.as_posix()]
-            
-            options = ['--gfn', '2', '--json']
+            options = ['-c', str(self.charge), '--sp', '--gfn', '2', '--json']
             
             if water is not None and isinstance(water, str):
                 if water == 'gbsa':
@@ -408,7 +359,6 @@ $ cp -r xtb-dist/share      /usr/local/ """)
             # '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)
-
             # if proc.returncode == 0:
             #     print("Standard Output:")
             #     print(proc.stdout)
@@ -438,6 +388,8 @@ $ cp -r xtb-dist/share      /usr/local/ """)
                 Wiberg_bond_orders = self.load_wbo(wbo_path)
 
                 return SimpleNamespace(
+                    natoms = self.natoms,
+                    charge = self.charge,
                     PE = datadict['total energy'] * hartree2kcalpermol,
                     Gsolv = Gsolv, 
                     charges = datadict['partial charges'], 
@@ -452,12 +404,29 @@ $ cp -r xtb-dist/share      /usr/local/ """)
     def optimize(self, water: str | None = None, verbose: bool = False) -> SimpleNamespace:
         """Optimize geometry.
 
-        Fortran runtime errror:
-            At line 852 of file ../src/optimizer.f90 (unit = 6, file = 'stdout')
-            Fortran runtime error: Missing comma between descriptors
-            (1x,"("f7.2"%)")
-                        ^
-            Error termination.
+        Options:
+            ```sh
+            -c, --chrg INT
+              specify molecular charge as INT, overrides .CHRG file and xcontrol option
+            -o, --opt [LEVEL]
+              call ancopt(3) to perform a geometry optimization, levels from crude, sloppy,
+              loose, normal (default), tight, verytight to extreme can be chosen
+            --gfn INT
+              specify parametrisation of GFN-xTB (default = 2)
+            --json
+              write xtbout.json file
+            ```
+
+        Notes:
+            Conda installed xtb has Fortran runtime error when optimizing geometry.
+            ```sh
+            Fortran runtime errror:
+                At line 852 of file ../src/optimizer.f90 (unit = 6, file = 'stdout')
+                Fortran runtime error: Missing comma between descriptors
+                (1x,"("f7.2"%)")
+                            ^
+                Error termination.
+            ```
 
         Args:
             water (str, optional) : water solvation model (choose 'gbsa' or 'alpb')
@@ -479,8 +448,7 @@ $ cp -r xtb-dist/share      /usr/local/ """)
                 geometry.write(self.to_xyz())
 
             cmd = ['xtb', geometry_input_path.as_posix()]
-
-            options = ['--opt', '--gfn', '2', '--json']
+            options = ['-c', str(self.charge), '-o', 'normal', '--gfn', '2', '--json']
 
             if water is not None and isinstance(water, str):
                 if water == 'gbsa':
@@ -503,6 +471,8 @@ $ cp -r xtb-dist/share      /usr/local/ """)
                 rdmol_opt = self.load_xyz(geometry_output_path)
             
                 return SimpleNamespace(
+                        natoms = self.natoms,
+                        charge = self.charge,
                         PE = datadict['total energy'] * hartree2kcalpermol,
                         charges = datadict['partial charges'],
                         wbo = Wiberg_bond_orders,

{rdworks-0.51.1 → rdworks-0.52.1}/src/rdworks.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rdworks
-Version: 0.51.1
+Version: 0.52.1
 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-0.51.1 → rdworks-0.52.1}/src/rdworks.egg-info/SOURCES.txt

@@ -7,6 +7,7 @@ src/rdworks/descriptor.py
 src/rdworks/display.py
 src/rdworks/ionized.py
 src/rdworks/matchedseries.py
+src/rdworks/microstates.py
 src/rdworks/mol.py
 src/rdworks/mollibr.py
 src/rdworks/pka.py

{rdworks-0.51.1 → rdworks-0.52.1}/tests/test_basics.py

@@ -6,15 +6,13 @@ import rdworks.autograph
 import math
 import copy
 import numpy as np
+import tempfile
 
 from rdworks import Conf, Mol, MolLibr
 from rdworks.utils import recursive_round
 
 
 datadir = Path(__file__).parent.resolve() / "data"
-workdir = Path(__file__).parent.resolve() / "outfiles"
-
-workdir.mkdir(exist_ok=True)
 
 
 # python >=3.12 raises SyntaxWarning: invalid escape sequence
@@ -190,8 +188,10 @@ def test_has_substr():
 def test_merge_csv():
     libr = MolLibr(drug_smiles, drug_names)
     libr = rdworks.merge_csv(libr, datadir / "drugs_20.csv", on='name')
-    libr.to_csv(workdir / "test_merge_csv.csv")
-    assert libr.count() == 20
+    with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+        workdir = Path(temp_dir)
+        libr.to_csv(workdir / "test_merge_csv.csv")
+        assert libr.count() == 20
 
 
 def test_read_smi():
@@ -217,42 +217,52 @@ def test_read_mae():
 
 def test_to_csv():
     libr1 = MolLibr(drug_smiles, drug_names, progress=False)
-    libr1.qed(progress=False).to_csv(workdir / "test_to_csv.csv")
-    libr2 = rdworks.read_csv(workdir / "test_to_csv.csv", smiles='smiles', name='name', progress=False)
-    assert libr1 == libr2
+    with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+        workdir = Path(temp_dir)
+        libr1.qed(progress=False).to_csv(workdir / "test_to_csv.csv")
+        libr2 = rdworks.read_csv(workdir / "test_to_csv.csv", smiles='smiles', name='name', progress=False)
+        assert libr1 == libr2
 
 
 def test_to_smi():
     libr = MolLibr(drug_smiles, drug_names, progress=False)
-    libr.to_smi(workdir / "test_to_smi.smi.gz")
-    libr.to_smi(workdir / "test_to_smi.smi")
+    with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+        workdir = Path(temp_dir)
+        libr.to_smi(workdir / "test_to_smi.smi.gz")
+        libr.to_smi(workdir / "test_to_smi.smi")
 
    
 def test_to_sdf():
     libr = MolLibr(drug_smiles, drug_names, progress=False)
-    libr.to_sdf(workdir / "test_to_sdf.sdf.gz")
-    libr.to_sdf(workdir / "test_to_sdf.sdf")
-    libr.qed().to_sdf(workdir / "test_to_sdf_with_qed.sdf") # QED and other properties should be here
-    supp = Chem.SDMolSupplier(workdir / "test_to_sdf_with_qed.sdf")
-    for m, mol in zip(supp, libr):
-        assert math.isclose(float(m.GetProp('MolWt')), 
-                            mol.props['MolWt'], rel_tol=1.0e-6, abs_tol=1.0e-3)
-        assert math.isclose(float(m.GetProp('QED')), 
-                            mol.props['QED'], rel_tol=1.0e-6, abs_tol=1.0e-3)
+    with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+        workdir = Path(temp_dir)
+        libr.to_sdf(workdir / "test_to_sdf.sdf.gz")
+        libr.to_sdf(workdir / "test_to_sdf.sdf")
+        libr.qed().to_sdf(workdir / "test_to_sdf_with_qed.sdf") # QED and other properties should be here
+        supp = Chem.SDMolSupplier(workdir / "test_to_sdf_with_qed.sdf")
+        for m, mol in zip(supp, libr):
+            assert math.isclose(float(m.GetProp('MolWt')), 
+                                mol.props['MolWt'], rel_tol=1.0e-6, abs_tol=1.0e-3)
+            assert math.isclose(float(m.GetProp('QED')), 
+                                mol.props['QED'], rel_tol=1.0e-6, abs_tol=1.0e-3)
 
 
 def test_to_png():
     libr = MolLibr(drug_smiles, drug_names, progress=False)
-    libr.to_png(workdir / "test_to_png.png")
-    libr.to_png(workdir / "test_to_png_with_index.png", atom_index=True)
+    with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+        workdir = Path(temp_dir)
+        libr.to_png(workdir / "test_to_png.png")
+        libr.to_png(workdir / "test_to_png_with_index.png", atom_index=True)
 
 
 def test_to_svg():
     libr = MolLibr(drug_smiles, drug_names, progress=False)
-    with open(workdir / "test_to_svg.svg", "w") as svg:
-        svg.write(libr.to_svg())
-    head = libr.to_svg()[:100]
-    assert head.startswith('<?xml') and ("<svg" in head)
+    with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+        workdir = Path(temp_dir)
+        with open(workdir / "test_to_svg.svg", "w") as svg:
+            svg.write(libr.to_svg())
+        head = libr.to_svg()[:100]
+        assert head.startswith('<?xml') and ("<svg" in head)
 
 
 def test_expand_rgroup():
@@ -265,9 +275,11 @@ def test_expand_rgroup():
 
 def test_scaffold_tree():
     libr = MolLibr(drug_smiles[:4], drug_names[:4])
-    for mol in libr:
-        adhoc_libr = MolLibr(rdworks.scaffold_tree(mol.rdmol)).rename(prefix=mol.name)
-        adhoc_libr.to_png(workdir / f'unittest_84_{mol.name}.png')
+    with tempfile.TemporaryDirectory() as temp_dir: # tmpdir is a string
+        workdir = Path(temp_dir)
+        for mol in libr:
+            adhoc_libr = MolLibr(rdworks.scaffold_tree(mol.rdmol)).rename(prefix=mol.name)
+            adhoc_libr.to_png(workdir / f'unittest_84_{mol.name}.png')
 
 
 def test_MatchedSeries():
@@ -465,4 +477,38 @@ def test_serialization():
     rebuilt = Mol().deserialize(serialized)
     assert rebuilt.count() == 10
     assert rebuilt.name == name
-    assert rebuilt == mol
+    assert rebuilt == mol
+
+
+def test_protonate_deprotonate():
+    _ = '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'
+    mol = Mol().deserialize(_, compressed=True)
+    
+    assert mol.confs[0].props == {'atoms': 60, 'charge': 0, 'idx': 11, 'pka_type': 'basic', 'pka': 5.066}
+    assert mol.confs[1].props == {'atoms': 60, 'charge': 0, 'idx': 16, 'pka_type': 'basic', 'pka': 5.663}
+    assert mol.confs[2].props == {'atoms': 60, 'charge': 0, 'idx': 17, 'pka_type': 'basic', 'pka': 5.298}
+    assert mol.confs[3].props == {'atoms': 60, 'charge': 0, 'idx': 4, 'pka_type': 'acidic', 'pka': 11.705}
+    
+    conf = mol.confs[0].copy()
+    conf = conf.protonate([11])
+    assert conf.positions().shape == (61, 3)
+    
+    conf = mol.confs[0].copy()
+    conf = conf.protonate([11,16])
+    assert conf.positions().shape == (62, 3)
+    
+    conf = mol.confs[0].copy()
+    conf = conf.protonate([11,16,17])
+    assert conf.positions().shape == (63, 3)
+    
+    conf = mol.confs[0].copy()
+    conf = conf.deprotonate([4])
+    assert conf.positions().shape == (59, 3)
+
+
+def test_from_molblock():
+    _ = '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'
+    mol = Mol().deserialize(_, compressed=True)
+    conf = mol.confs[0].copy()
+    mb = conf.to_molblock()
+    mol2 = Mol().from_molblock(mb)

{rdworks-0.51.1 → rdworks-0.52.1}/tests/test_qupkake.py

@@ -1,11 +1,14 @@
 import rdworks
+from pathlib import Path
+
+datadir = Path(__file__).parent.resolve() / "data"
 
 def test_readin_qupkake_output():
-    libr = rdworks.read_sdf('qupkake/output/qupkake_output.sdf', confs=True)
+    libr = rdworks.read_sdf(datadir / 'qupkake_output_SAMPL6_SM07.sdf', confs=True)
     # libr contains one molecule because of confs=True
     assert libr.count() == 1
     for m in libr:
-        assert m.count() == 4
+        assert m.count() == 3
         print(m.props)
         print()
         for c in m:

{rdworks-0.51.1 → rdworks-0.52.1}/tests/test_xtb.py

@@ -4,6 +4,7 @@ from rdworks.testdata import drugs
 
 from pathlib import Path
 
+import tempfile
 
 # In ASE, the default energy unit is eV (electron volt).
 # It will be converted to kcal/mol
@@ -24,7 +25,6 @@ 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
@@ -72,5 +72,5 @@ def test_optimize():
 
 if __name__ == '__main__':
     test_xtb_wrapper()
-    test_singlepoint()
-    test_optimize()
+    # test_singlepoint()
+    # test_optimize()