fix: ligandprep skip failed cases

unidock/README.md

@@ -58,7 +58,7 @@ make clang-format
 ### Using binary
 
 Please download the latest binary of Uni-Dock at the assets tab of [the Release page](https://github.com/dptech-corp/Uni-Dock/releases).
-Executable `unidock` supports vina and vinardo scoring functions, and `unidock_ad4` supports ad4 scoring function.
+Executable `unidock` to dock.
 
 After downloading, please make sure that the path to `unidock` is in your `PATH` environment variable.
 
@@ -69,7 +69,11 @@ After downloading, please make sure that the path to `unidock` is in your `PATH`
 To launch a Uni-Dock job, the most important parameters are as follows:
 
 - `--receptor`: filepath of the receptor (PDBQT)
+
+     (If you want to use `ad4` scoring function, you need to generate affinity maps first and use `--maps <mapdir/receptor_prefix>` instead of `--receptor`. The method to generate maps is in [here](https://autodock-vina.readthedocs.io/en/latest/docking_basic.html#optional-generating-affinity-maps-for-autodock-ff))
+
 - `--gpu_batch`: filepath of the ligands to dock with GPU (PDBQT), enter multiple at a time, separated by spaces (" ")
+
 - `--search_mode`: computational complexity, choose in \[*`fast`*, *`balance`*, and *`detail`*].
 
 ***Advanced options***

unidock_tools/pyproject.toml

@@ -48,6 +48,6 @@ pythonVersion = "3.11"
 pythonPlatform = "Linux"
 reportMissingImports = false
 reportMissingModuleSource = false
-reportMissingTypeStubs = false
 reportAttributeAccessIssue = false
-reportIncompatibleMethodOverride = false
+reportIncompatibleMethodOverride = false
+reportCallIssue = false

unidock_tools/src/unidock_tools/application/ligandprep.py

@@ -3,37 +3,50 @@
 import os
 from functools import partial
 from multiprocessing import Pool
+import traceback
 import logging
 import argparse
 from rdkit import Chem
 
+from unidock_tools.utils import read_ligand
 from unidock_tools.modules.ligand_prep import TopologyBuilder
 
 
 def iter_ligands(ligands: List[Path], batch_size: int = 1200,
                  use_file_name: bool = False) -> Generator[List[Tuple[Chem.Mol, str]], None, None]:
     curr_mol_name_list = []
     for ligand in ligands:
-        mols = list(Chem.SDMolSupplier(str(ligand), removeHs=False))
+        mols = read_ligand(ligand)
         for i, mol in enumerate(mols):
-            if not use_file_name and mol.HasProp("_Name") and mol.GetProp("_Name").strip():
-                name = mol.GetProp("_Name").strip()
+            if mol:
+                if not use_file_name and mol.HasProp("_Name") and mol.GetProp("_Name").strip():
+                    name = mol.GetProp("_Name").strip()
+                else:
+                    name = f"{ligand.stem}_{i}" if len(mols) > 1 else ligand.stem
+                curr_mol_name_list.append((mol, name))
+                if len(curr_mol_name_list) > batch_size:
+                    yield curr_mol_name_list
+                    curr_mol_name_list = []
             else:
-                name = f"{ligand.stem}_{i}" if len(mols) > 1 else ligand.stem
-            curr_mol_name_list.append((mol, name))
-            if len(curr_mol_name_list) > batch_size:
-                yield curr_mol_name_list
-                curr_mol_name_list = []
+                logging.warning(f"read ligand file {ligand.stem} ind {i} error")
     if len(curr_mol_name_list) > 0:
         yield curr_mol_name_list
     return
 
 
-def ligprep(mol_name_tup: Tuple[Chem.Mol, str], savedir: Path):
+def ligprep(mol_name_tup: Tuple[Chem.Mol, str], savedir: Path, save_format: str = "sdf"):
     mol, name = mol_name_tup
-    tb = TopologyBuilder(mol)
-    tb.build_molecular_graph()
-    tb.write_sdf_file(os.path.join(savedir, f"{name}.sdf"))
+    try:
+        tb = TopologyBuilder(mol)
+        tb.build_molecular_graph()
+        if save_format == "sdf":
+            tb.write_sdf_file(os.path.join(savedir, f"{name}.sdf"))
+        elif save_format == "pdbqt":
+            tb.write_pdbqt_file(os.path.join(savedir, f"{name}.pdbqt"))
+        else:
+            logging.error(f"Invalid save format: {save_format}")
+    except:
+        logging.error(f"ligprep failed for {name}: {traceback.format_exc()}")
 
 
 def main(args: dict):
@@ -52,9 +65,11 @@ def main(args: dict):
                     continue
                 ligands.append(Path(line.strip()).resolve())
 
+    os.makedirs(Path(args["savedir"]).resolve(), exist_ok=True)
     for mol_name_tup_list in iter_ligands(ligands, args["batch_size"], args["use_file_name"]):
         with Pool(os.cpu_count()) as pool:
-            pool.map(partial(ligprep, savedir=args["savedir"]), mol_name_tup_list)
+            pool.map(partial(ligprep, savedir=args["savedir"], 
+                             save_format=args["save_format"]), mol_name_tup_list)
 
 
 def get_parser() -> argparse.ArgumentParser:
@@ -65,6 +80,8 @@ def get_parser() -> argparse.ArgumentParser:
                         help="A text file containing the path of ligand files in sdf format.")
     parser.add_argument("-sd", "--savedir", type=str, default="ligprep_results",
                         help="Save directory. Default: 'MultiConfDock-Result'.")
+    parser.add_argument("-sf", "--save_format", type=str, default="sdf", 
+                        help="Ligprep result files format. Choose from ['sdf', 'pdbqt']. Default: 'sdf'.")
     parser.add_argument("-bs", "--batch_size", type=int, default=1200,
                         help="Batch size for docking. Default: 1200.")
     parser.add_argument("-ufn", "--use_file_name", action="store_true",

unidock_tools/src/unidock_tools/application/unidock_pipeline.py

@@ -257,6 +257,7 @@ def main(args: dict):
         batch_size=int(args["batch_size"]),
         score_only=bool(args["score_only"]),
         local_only=bool(args["local_only"]),
+        props_list=["fragInfo", "torsionInfo", "atomInfo"],
     )
     runner.save_results(save_dir=savedir)
     end_time = time.time()

unidock_tools/src/unidock_tools/modules/ligand_prep/torsion_tree.py

@@ -1,4 +1,4 @@
-from typing import Dict, Tuple
+from typing import Dict, Tuple, Union
 import os
 import numpy as np
 import networkx as nx
@@ -163,7 +163,7 @@ def build_molecular_graph(self):
         node_idx = 0
         root_fragment = splitted_mol_list[root_fragment_idx]
         num_root_atoms = root_fragment.GetNumAtoms()
-        atom_info_list = [None] * num_root_atoms
+        atom_info_list = [dict()] * num_root_atoms
         for root_atom_idx in range(num_root_atoms):
             root_atom = root_fragment.GetAtomWithIdx(root_atom_idx)
             atom_info_dict = {}
@@ -189,7 +189,7 @@ def build_molecular_graph(self):
             else:
                 fragment = splitted_mol_list[fragment_idx]
                 num_fragment_atoms = fragment.GetNumAtoms()
-                atom_info_list = [None] * num_fragment_atoms
+                atom_info_list = [dict()] * num_fragment_atoms
                 for atom_idx in range(num_fragment_atoms):
                     atom = fragment.GetAtomWithIdx(atom_idx)
                     atom_info_dict = {}
@@ -341,7 +341,7 @@ def _deep_first_search(self, node_idx):
                 self.pdbqt_atom_line_list.append(
                     self.pdbqt_end_branch_line_format.format('ENDBRANCH', parent_atom_idx, offspring_atom_idx))
 
-    def write_pdbqt_file(self, out_file: str = ''):
+    def write_pdbqt_file(self, out_file: Union[str, bytes, os.PathLike]):
         self.pdbqt_remark_line_list = []
         self.pdbqt_atom_line_list = []
 
@@ -383,7 +383,7 @@ def write_pdbqt_file(self, out_file: str = ''):
             for pdbqt_line in self.pdbqt_line_list:
                 f.write(pdbqt_line)
 
-    def write_constraint_bpf_file(self, out_path: str = ''):
+    def write_constraint_bpf_file(self, out_path: Union[str, bytes, os.PathLike]):
         self.core_bpf_remark_line_list = []
         self.core_bpf_atom_line_list = []
         self.core_bpf_atom_line_format = '{:8.3f}\t{:8.3f}\t{:8.3f}\t{:6.2f}\t{:6.2f}\t{:3s}\t{:<2s}\n'
@@ -452,21 +452,13 @@ def get_sdf_torsion_tree_info(self) -> Tuple[str, str, str, str]:
 
         return frag_info_str, frag_all_info_str, torsion_info_str, atom_info_str
 
-    def write_sdf_file(self, out_file: str = '', do_rigid_docking: bool = False):
+    def write_sdf_file(self, out_file: Union[str, bytes, os.PathLike], do_rigid_docking: bool = False):
         frag_info_str, frag_all_info_str, torsion_info_str, atom_info_str = self.get_sdf_torsion_tree_info()
         if do_rigid_docking:
             self.mol.SetProp("fragInfo", frag_all_info_str)
         else:
             self.mol.SetProp("fragInfo", frag_info_str)
             self.mol.SetProp("torsionInfo", torsion_info_str)
         self.mol.SetProp("atomInfo", atom_info_str)
-        if out_file:
-            os.makedirs(os.path.dirname(os.path.abspath(out_file)), exist_ok=True)
-            with Chem.SDWriter(out_file) as writer:
-                writer.write(self.mol)
-
-
-def generate_topology(mol: Chem.Mol, out_file: str = ''):
-    topology_builder = TopologyBuilder(mol)
-    topology_builder.build_molecular_graph()
-    topology_builder.write_pdbqt_file(out_file=out_file)
+        with Chem.SDWriter(str(out_file)) as writer:
+            writer.write(self.mol)

unidock_tools/src/unidock_tools/modules/protein_prep/pdb2pdbqt.py

@@ -3,7 +3,6 @@
 from math import isnan, isinf
 from rdkit import Chem
 from rdkit.Chem import AllChem
-from rdkit.Chem.rdchem import BondType
 
 
 def get_pdbqt_atom_lines(mol: Chem.Mol, donors: List[int], acceptors: List[int]):

unidock_tools/src/unidock_tools/utils/mol_group.py

@@ -57,7 +57,7 @@ def update_conf_props(self, conf_props: dict):
         self.conf_props.update(conf_props)
 
     def get_rdkit_mol_conf_with_props(self, conf_idx: int, props_list: List[str] = [], 
-                                      exclude_props_list: List[str] = []):
+                                      exclude_props_list: List[str] = []) -> Chem.Mol:
         mol = copy.copy(self.mol_confs[conf_idx])
         props = copy.deepcopy(self.get_props())
         props.update({k:v[conf_idx] for k, v in self.get_conf_props().items()})
@@ -135,10 +135,9 @@ def write_sdf_by_idx(self,
                          ) -> List[Path]:
         save_dir = make_tmp_dir(str(save_dir), False, False)
 
-        mol_confs_copy = [None] * len(self.mol_group[idx])
-        for conf_idx in range(len(self.mol_group[idx])):
-            mol_conf_copy = self.mol_group[idx].get_rdkit_mol_conf_with_props(conf_idx, props_list, exclude_props_list)
-            mol_confs_copy[conf_idx] = mol_conf_copy
+        mol_confs_copy = [self.mol_group[idx].get_rdkit_mol_conf_with_props(
+            conf_idx, props_list, exclude_props_list) for conf_idx in range(
+                len(self.mol_group[idx]))]
         # save SDF files
         file_prefix = self.mol_group[idx].get_props()['file_prefix']
         sdf_file_list = []

unidock_tools/src/unidock_tools/utils/rdkit_helper.py

@@ -32,9 +32,9 @@ def set_properties(mol: Chem.Mol, properties: dict):
     """
     for key, value in properties.items():
         try:
-            if isinstance(key, int):
+            if isinstance(value, int):
                 mol.SetIntProp(key, value)
-            elif isinstance(key, float):
+            elif isinstance(value, float):
                 mol.SetDoubleProp(key, value)
             else:
                 mol.SetProp(key, str(value))