Improved temporary file generation and allowing auto-generation of required…

Improved temporary file generation and allowing auto-generation of required output folder structure for both ligand embedding and docking phases.

Improved temporary file generation and allowing auto-generation of required…
Improved temporary file generation and allowing auto-generation of required output folder structure for both ligand embedding and docking phases.
8985aee6 · Christian Margreitter · ac69a9e6 · 8985aee6 · 8985aee6 · 8985aee6
Commit 8985aee6 authored Sep 08, 2021 by Christian Margreitter
17 changed files
--- a/dockstream/core/AutodockVina/AutodockVina_docker.py
+++ b/dockstream/core/AutodockVina/AutodockVina_docker.py
@@ -18,6 +18,7 @@ from dockstream.utils.enums.AutodockVina_enums import AutodockVinaExecutablesEnu
 from dockstream.utils.execute_external.OpenBabel import OpenBabelExecutor
 from dockstream.utils.enums.OpenBabel_enums import OpenBabelExecutablesEnum
 from dockstream.utils.enums.RDkit_enums import RDkitLigandPreparationEnum
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.translations.molecule_translator import MoleculeTranslator
 from dockstream.utils.dockstream_exceptions import DockingRunFailed
@@ -112,7 +113,7 @@ class AutodockVina(Docker, BaseModel):
    def _write_molecule_to_pdbqt(self, path, molecule) -> bool:
        # generate temporary copy as PDB
-        _, temp_pdb = tempfile.mkstemp(suffix=".pdb")
+        temp_pdb = gen_temp_file(suffix=".pdb")
        Chem.MolToPDBFile(mol=molecule, filename=temp_pdb)
        # Note: In contrast to the target preparation,
@@ -149,8 +150,8 @@ class AutodockVina(Docker, BaseModel):
            for ligand in sublist:
                # generate temporary input files and output directory
                cur_tmp_output_dir = tempfile.mkdtemp()
-                _, cur_tmp_input_pdbqt = tempfile.mkstemp(prefix=str(start_index), suffix=".pdbqt", dir=cur_tmp_output_dir)
+                cur_tmp_input_pdbqt = gen_temp_file(prefix=str(start_index), suffix=".pdbqt", dir=cur_tmp_output_dir)
-                _, cur_tmp_output_sdf = tempfile.mkstemp(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
+                cur_tmp_output_sdf = gen_temp_file(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
                # write-out the temporary input file
                if ligand.get_molecule() is None:
@@ -250,11 +251,11 @@ class AutodockVina(Docker, BaseModel):
    def _dock_subjob(self, input_path_pdbqt, output_path_sdf):
        # at the moment, the log file is not parsed
-        _, path_log_file = tempfile.mkstemp(suffix=".log", dir=os.path.dirname(input_path_pdbqt))
+        path_log_file = gen_temp_file(suffix=".log", dir=os.path.dirname(input_path_pdbqt))
        # set up arguments list and execute
        # TODO: support "ensemble docking" - currently, only the first entry is used
-        _, tmp_pdbqt_docked = tempfile.mkstemp(suffix=".pdbqt", dir=os.path.dirname(input_path_pdbqt))
+        tmp_pdbqt_docked = gen_temp_file(suffix=".pdbqt", dir=os.path.dirname(input_path_pdbqt))
        search_space = self.parameters.search_space
        arguments = [_EE.VINA_RECEPTOR, self.parameters.receptor_pdbqt_path[0],
                     _EE.VINA_LIGAND, input_path_pdbqt,

--- a/dockstream/core/AutodockVina/AutodockVina_target_preparator.py
+++ b/dockstream/core/AutodockVina/AutodockVina_target_preparator.py
@@ -60,7 +60,7 @@ class AutodockVinaTargetPreparator(TargetPreparator):
    def _export_as_pdb2pdbqt(self, path):
        # generate temporary copy
-        _, temp_target_pdb = tempfile.mkstemp(suffix=".pdb")
+        temp_target_pdb = gen_temp_file(suffix=".pdb")
        Chem.MolToPDBFile(mol=self._target, filename=temp_target_pdb)
        # set target pH value that determines the protein's side-chain states

--- a/dockstream/core/Corina/Corina_ligand_preparator.py
+++ b/dockstream/core/Corina/Corina_ligand_preparator.py
@@ -14,6 +14,7 @@ from dockstream.core.ligand.ligand import get_next_enumeration_number_for_ligand
 from dockstream.core.ligand_preparator import LigandPreparator, _LE
 from dockstream.utils.dockstream_exceptions import LigandPreparationFailed
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.smiles import to_mol
 from dockstream.utils.execute_external.Corina import CorinaExecutor
@@ -137,8 +138,8 @@ class CorinaLigandPreparator(LigandPreparator, BaseModel):
        # 1) generate temporary folder and files
        tmp_output_dir_path = tempfile.mkdtemp()
-        fd, tmp_smiles_path = tempfile.mkstemp(suffix=".smi", dir=tmp_output_dir_path)
+        tmp_smiles_path = gen_temp_file(suffix=".smi", dir=tmp_output_dir_path)
-        fd, tmp_molecules_path = tempfile.mkstemp(suffix=".sdf", dir=tmp_output_dir_path)
+        tmp_molecules_path = gen_temp_file(suffix=".sdf", dir=tmp_output_dir_path)
        # 2) save the SMILES
        with open(tmp_smiles_path, 'w') as f:

--- a/dockstream/core/Gold/Gold_docker.py
+++ b/dockstream/core/Gold/Gold_docker.py
@@ -23,6 +23,7 @@ from dockstream.core.docker import Docker
 from dockstream.core.Gold.Gold_result_parser import GoldResultParser
 from dockstream.utils.enums.Gold_enums import GoldLigandPreparationEnum
 from dockstream.utils.enums.Gold_enums import GoldTargetKeywordEnum, GoldExecutablesEnum, GoldOutputEnum
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.translations.molecule_translator import MoleculeTranslator
 from dockstream.utils.dockstream_exceptions import DockingRunFailed
@@ -114,15 +115,14 @@ class Gold(Docker):
        tmpdir = tempfile.mkdtemp()
        if self._target_dict[_TK.CAVITY_METHOD] == _TK.CAVITY_METHOD_REFERENCE:
            # write ligand to temporary file (ending copied over in settings)
-            _, tmp_ref_ligand_path = tempfile.mkstemp(suffix=self._target_dict[_TK.REFERENCE_LIGAND_FILENAME],
+            tmp_ref_ligand_path = gen_temp_file(suffix=self._target_dict[_TK.REFERENCE_LIGAND_FILENAME], dir=tmpdir)
-                                                      dir=tmpdir)
            with open(tmp_ref_ligand_path, 'w') as file:
                for line in self._target_dict[_TK.REFERENCE_LIGAND]:
                    file.write(line)
                self._logger.log(f"Wrote temporary ligand file {tmp_ref_ligand_path} with {len(self._target_dict[_TK.REFERENCE_LIGAND])} lines.", _LE.DEBUG)
            # write target PDB to temporary file
-            _, tmp_target_path = tempfile.mkstemp(suffix=".pdb", dir=tmpdir)
+            tmp_target_path = gen_temp_file(suffix=".pdb", dir=tmpdir)
            with open(tmp_target_path, 'w') as file:
                for line in self._target_dict[_TK.TARGET_PDB]:
                    file.write(line)
@@ -172,7 +172,7 @@ class Gold(Docker):
        for start_index, sublist in zip(start_indices, sublists):
            # generate temporary input files and output directory
            cur_tmp_output_dir = tempfile.mkdtemp()
-            _, cur_tmp_sdf = tempfile.mkstemp(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
+            cur_tmp_sdf = gen_temp_file(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
            # write-out the temporary input file
            writer = Chem.SDWriter(cur_tmp_sdf)
@@ -191,7 +191,7 @@ class Gold(Docker):
                continue
            # add the path to which "_dock_subjob()" will write the result SDF
-            _, output_sdf_path = tempfile.mkstemp(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
+            output_sdf_path = gen_temp_file(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
            tmp_output_dirs.append(cur_tmp_output_dir)
            tmp_output_sdf_paths.append(output_sdf_path)
            tmp_input_sdf_paths.append(cur_tmp_sdf)

--- a/dockstream/core/OpenEyeHybrid/Omega_ligand_preparator.py
+++ b/dockstream/core/OpenEyeHybrid/Omega_ligand_preparator.py
@@ -14,6 +14,7 @@ from copy import deepcopy
 from dockstream.core.ligand.ligand import Ligand, get_enumerations_for_ligand
 from dockstream.core.ligand_preparator import LigandPreparator, _LE
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.parallelization.general_utils import split_into_sublists, get_progress_bar_string
 from dockstream.loggers.blank_logger import BlankLogger
@@ -158,8 +159,8 @@ class OmegaLigandPreparator(LigandPreparator, BaseModel):
            # generate temporary input files and output directory
            cur_tmp_output_dir = tempfile.mkdtemp()
            tmp_output_dirs.append(cur_tmp_output_dir)
-            _, cur_tmp_sdf = tempfile.mkstemp(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
+            cur_tmp_sdf = gen_temp_file(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
-            _, cur_tmp_smi = tempfile.mkstemp(prefix=str(start_index), suffix=".smi", dir=cur_tmp_output_dir)
+            cur_tmp_smi = gen_temp_file(prefix=str(start_index), suffix=".smi", dir=cur_tmp_output_dir)
            tmp_input_smi_paths.append(cur_tmp_smi)
            # write smiles to temporary file as "OMEGA" backend
@@ -168,7 +169,7 @@ class OmegaLigandPreparator(LigandPreparator, BaseModel):
                    f.write(lig.get_smile() + " " + lig.get_identifier() + "\n")
            # add the path to which "_dock_subjob()" will write the result SDF
-            _, output_sdf_path = tempfile.mkstemp(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
+            output_sdf_path = gen_temp_file(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
            tmp_output_sdf_paths.append(output_sdf_path)
        return tmp_output_dirs, tmp_input_smi_paths, tmp_output_sdf_paths

--- a/dockstream/core/OpenEyeHybrid/OpenEyeHybrid_docker.py
+++ b/dockstream/core/OpenEyeHybrid/OpenEyeHybrid_docker.py
@@ -18,6 +18,7 @@ from dockstream.core.docker import Docker
 from dockstream.core.OpenEyeHybrid.OpenEyeHybrid_result_parser import OpenEyeHybridResultParser
 from dockstream.utils.enums.OE_Hybrid_enums import OpenEyeHybridLigandPreparationEnum
 from dockstream.utils.enums.OE_Hybrid_enums import OpenEyeHybridExecutablesEnum, OpenEyeHybridOutputKeywordsEnum
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.translations.molecule_translator import MoleculeTranslator
 from dockstream.utils.dockstream_exceptions import DockingRunFailed
@@ -101,7 +102,7 @@ class OpenEyeHybrid(Docker):
        for start_index, sublist in zip(start_indices, sublists):
            # generate temporary input files and output directory
            cur_tmp_output_dir = tempfile.mkdtemp()
-            _, cur_tmp_sdf = tempfile.mkstemp(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
+            cur_tmp_sdf = gen_temp_file(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
            # write-out the temporary input file
            one_written = False
@@ -123,7 +124,7 @@ class OpenEyeHybrid(Docker):
            tmp_input_sdf_paths.append(cur_tmp_sdf)
            # add the path to which "_dock_subjob()" will write the result SDF
-            _, output_sdf_path = tempfile.mkstemp(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
+            output_sdf_path = gen_temp_file(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
            tmp_output_sdf_paths.append(output_sdf_path)
        return tmp_output_dirs, tmp_input_sdf_paths, tmp_output_sdf_paths

--- a/dockstream/core/Schrodinger/Glide_docker.py
+++ b/dockstream/core/Schrodinger/Glide_docker.py
@@ -21,6 +21,7 @@ from dockstream.utils.enums.ligand_preparation_enum import LigandPreparationEnum
 from dockstream.utils.enums.Schrodinger_enums import SchrodingerExecutablesEnum, \
                                                 SchrodingerDockingConfigurationEnum, \
                                                 SchrodingerOutputEnum
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.translations.molecule_translator import MoleculeTranslator
 from dockstream.utils.files_paths import any_in_file
@@ -336,7 +337,7 @@ class Glide(Docker, BaseModel):
        # arrange the elements and blocks
        if path is None:
-            _, path = tempfile.mkstemp(suffix=".in")
+            path = gen_temp_file(suffix=".in")
        with open(path, mode='w') as f:
            self._logger.log(f"Writing GLIDE input file {path}:\n", _LE.DEBUG)
            for line in element_lines:
@@ -362,8 +363,8 @@ class Glide(Docker, BaseModel):
        for start_index, sublist in zip(start_indices, sublists):
            # generate temporary input files and output directory
            cur_tmp_output_dir = tempfile.mkdtemp()
-            _, cur_tmp_sdf = tempfile.mkstemp(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
+            cur_tmp_sdf = gen_temp_file(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
-            _, cur_tmp_mae = tempfile.mkstemp(prefix=str(start_index), suffix=".mae", dir=cur_tmp_output_dir)
+            cur_tmp_mae = gen_temp_file(prefix=str(start_index), suffix=".mae", dir=cur_tmp_output_dir)
            # write-out the temporary input file
            writer = Chem.SDWriter(cur_tmp_sdf)
@@ -385,7 +386,7 @@ class Glide(Docker, BaseModel):
            self._translate_SDF_to_MAE(sdf_path=cur_tmp_sdf, mae_path=cur_tmp_mae)
            # add the path to which "_dock_subjob()" will write the result SDF
-            _, output_sdf_path = tempfile.mkstemp(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
+            output_sdf_path = gen_temp_file(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
            tmp_output_sdf_paths.append(output_sdf_path)
            tmp_input_mae_paths.append(cur_tmp_mae)
            tmp_output_dirs.append(cur_tmp_output_dir)
@@ -474,7 +475,7 @@ class Glide(Docker, BaseModel):
        keywords[_EE.GLIDE_LIGANDFILE] = mae_ligand_path
        # 2) write the keyword-input file for the "Glide" backend; write-out to temporary file
-        _, glide_keywords_path = tempfile.mkstemp(suffix=".in", dir=tmp_output_dir)
+        glide_keywords_path = gen_temp_file(suffix=".in", dir=tmp_output_dir)
        _ = self._write_keywords_to_file(keywords=keywords, path=glide_keywords_path)
        # 3) wait / sleep until job is completed

--- a/dockstream/core/Schrodinger/Ligprep_ligand_preparator.py
+++ b/dockstream/core/Schrodinger/Ligprep_ligand_preparator.py
@@ -17,6 +17,7 @@ from dockstream.core.Schrodinger.license_token_guard import SchrodingerLicenseTo
 from dockstream.loggers.blank_logger import BlankLogger
 from dockstream.core.ligand_preparator import LigandPreparator, _LE
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.parallelization.general_utils import split_into_sublists, get_progress_bar_string
@@ -215,8 +216,8 @@ class LigprepLigandPreparator(LigandPreparator, BaseModel):
            # generate temporary input files and output directory
            cur_tmp_output_dir = tempfile.mkdtemp()
            tmp_output_dirs.append(cur_tmp_output_dir)
-            _, cur_tmp_smi = tempfile.mkstemp(prefix=str(start_index), suffix=".smi", dir=cur_tmp_output_dir)
+            cur_tmp_smi = gen_temp_file(prefix=str(start_index), suffix=".smi", dir=cur_tmp_output_dir)
-            _, cur_tmp_filter = tempfile.mkstemp(prefix=str(start_index), suffix=".lff", dir=cur_tmp_output_dir)
+            cur_tmp_filter = gen_temp_file(prefix=str(start_index), suffix=".lff", dir=cur_tmp_output_dir)
            tmp_input_smi_paths.append(cur_tmp_smi)
            tmp_input_filter_paths.append(cur_tmp_filter)
@@ -226,7 +227,7 @@ class LigprepLigandPreparator(LigandPreparator, BaseModel):
                    f.write(lig.get_smile() + " " + lig.get_identifier() + "\n")
            # add the path to which "_dock_subjob()" will write the result SDF
-            _, output_sdf_path = tempfile.mkstemp(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
+            output_sdf_path = gen_temp_file(prefix=str(start_index), suffix="_result.sdf", dir=cur_tmp_output_dir)
            tmp_output_sdf_paths.append(output_sdf_path)
        return tmp_output_dirs, tmp_input_smi_paths, tmp_output_sdf_paths, tmp_input_filter_paths

--- a/dockstream/core/TautEnum/taut_enum_smile_preparation.py
+++ b/dockstream/core/TautEnum/taut_enum_smile_preparation.py
@@ -10,6 +10,7 @@ from dockstream.utils.execute_external.TautEnum import TautEnumExecutor
 from dockstream.utils.enums.taut_enum_enums import TautEnumEnum
 from dockstream.utils.enums.logging_enums import LoggingConfigEnum
 from dockstream.core.ligand.ligand import Ligand, get_next_enumeration_number_for_ligand
+from dockstream.utils.general_utils import gen_temp_file
 class TautEnumSmilePreparator:
@@ -40,8 +41,8 @@ class TautEnumSmilePreparator:
        # 1) generate temporary folder and files
        tmp_output_dir_path = tempfile.mkdtemp()
-        fd, tmp_input_smiles_path = tempfile.mkstemp(suffix=".smi", dir=tmp_output_dir_path)
+        tmp_input_smiles_path = gen_temp_file(suffix=".smi", dir=tmp_output_dir_path)
-        fd, tmp_output_smiles_path = tempfile.mkstemp(suffix=".smi", dir=tmp_output_dir_path)
+        tmp_output_smiles_path = gen_temp_file(suffix=".smi", dir=tmp_output_dir_path)
        # 2) save the SMILES
        original_smiles = []

--- a/dockstream/core/docker.py
+++ b/dockstream/core/docker.py
-import os
 import abc
 import time
 from copy import deepcopy
@@ -12,6 +11,7 @@ from pydantic import BaseModel, PrivateAttr
 from dockstream.loggers.docking_logger import DockingLogger
 from dockstream.loggers.blank_logger import BlankLogger
 from dockstream.utils.dockstream_exceptions import DockingRunFailed
+from dockstream.utils.files_paths import generate_folder_structure
 from dockstream.utils.parallelization.general_utils import split_into_sublists, get_progress_bar_string
 from dockstream.utils.enums.ligand_preparation_enum import LigandPreparationEnum
@@ -285,6 +285,10 @@ class Docker(BaseModel, metaclass=abc.ABCMeta):
        if not self._docking_performed:
            raise DockingRunFailed("Do the docking first.")
        selected_conformers = self._select_conformers(mode=mode, mol_type=mol_type)
+        # generate folder structure, if not available
+        generate_folder_structure(filepath=path)
        if mol_type == _LPE.TYPE_RDKIT:
            import rdkit.Chem as Chem
            writer = Chem.SDWriter(path)
@@ -377,6 +381,10 @@ class Docker(BaseModel, metaclass=abc.ABCMeta):
                    self._logger.log(f"Score output mode \"{mode}\" is unknown - write-out of scores failed.",
                                     _LE.ERROR)
                    raise DockingRunFailed()
+                # generate folder structure, if not available
+                generate_folder_structure(filepath=path)
                df_buffer.to_csv(path_or_buf=path,
                                 sep=',',
                                 na_rep='',

--- a/dockstream/core/ligand_preparator.py
+++ b/dockstream/core/ligand_preparator.py
@@ -19,6 +19,7 @@ from dockstream.utils.enums.RDkit_enums import RDkitLigandPreparationEnum
 from dockstream.utils.enums.logging_enums import LoggingConfigEnum
 from dockstream.utils.enums.transformations_enums import TransformationEnum
 from dockstream.utils.enums.stereo_enumeration_enums import StereoEnumerationEnum
+from dockstream.utils.files_paths import generate_folder_structure
 _DE = DockingConfigurationEnum()
 _LP = LigandPreparationEnum()
@@ -220,6 +221,9 @@ class LigandPreparator(BaseModel):
        format = format.upper()
        ligands_copy = [deepcopy(lig) for lig in self.ligands]
+        # generate folder structure, if not available
+        generate_folder_structure(filepath=path)
        # check and specify format of file
        # RDkit does not support the write-out of MOL2 files (apparently because of the format's inherent ambiguity)
        if format == _LP.OUTPUT_FORMAT_SDF:

--- a/dockstream/core/rDock/rDock_docker.py
+++ b/dockstream/core/rDock/rDock_docker.py
@@ -16,6 +16,7 @@ from dockstream.utils.enums.logging_enums import LoggingConfigEnum
 from dockstream.utils.execute_external.rDock import rDockExecutor
 from dockstream.utils.enums.rDock_enums import rDockExecutablesEnum, rDockDockingConfigurationEnum, rDockRbdockOutputEnum
 from dockstream.utils.enums.RDkit_enums import RDkitLigandPreparationEnum
+from dockstream.utils.general_utils import gen_temp_file
 from dockstream.utils.translations.molecule_translator import MoleculeTranslator
 from dockstream.utils.dockstream_exceptions import DockingRunFailed
@@ -88,7 +89,7 @@ class rDock(Docker):
            # generate temporary input file and output directory into which "rbdock" will deposit the poses
            cur_tmp_output_dir = tempfile.mkdtemp()
-            _, cur_tmp_sdf = tempfile.mkstemp(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
+            cur_tmp_sdf = gen_temp_file(prefix=str(start_index), suffix=".sdf", dir=cur_tmp_output_dir)
            # write-out the temporary input file
            one_written = False

--- a/dockstream/core/rDock/rDock_target_preparator.py
+++ b/dockstream/core/rDock/rDock_target_preparator.py
@@ -70,7 +70,7 @@ class rDockTargetPreparator(TargetPreparator):
        # --> bottom line: use openbabel instead of RDkit
        # generate temporary copy in Mol2 file
-        _, temp_target_pdb = tempfile.mkstemp(suffix=".pdb")
+        temp_target_pdb = gen_temp_file(suffix=".pdb")
        Chem.MolToPDBFile(mol=self._target, filename=temp_target_pdb)
        # kekulization of bonds leads to problems with PDB conversion of aromatic side-chains - ignore for now
@@ -97,7 +97,8 @@ class rDockTargetPreparator(TargetPreparator):
        line = [line for line in lines if "Total volume" in line][0]
        result[self._GK.SPECIFYCAVITY_METADATA_TOTALVOLUME] = float(line.split()[2])
-        # cavity 1, example: Cavity #1	Size=8206 points; Vol=1025.75 A^3; Min=(-5,1.5,18.5); Max=(13.5,20,31.5); Center=(4.5011,9.94851,24.5739); Extent=(18.5,18.5,13)
+        # cavity 1, example: Cavity #1	Size=8206 points; Vol=1025.75 A^3; Min=(-5,1.5,18.5); Max=(13.5,20,31.5);
+        # Center=(4.5011,9.94851,24.5739); Extent=(18.5,18.5,13)
        line = [line for line in lines if "Cavity #1" in line][0]
        parts = line.split()
        result[self._GK.SPECIFYCAVITY_METADATA_SIZEINPOINTS] = int(parts[2].split(sep='=')[1])

--- a/dockstream/utils/files_paths.py
+++ b/dockstream/utils/files_paths.py
 import os
 import time
 import json
+from pathlib import Path
 def wait_until_file_generation(path, interval_sec=1, maximum_sec=None) -> bool:
@@ -58,3 +59,8 @@ def any_in_file(path, strings):
            return False
    else:
        return False
+def generate_folder_structure(filepath: str):
+    folder_path = os.path.dirname(filepath)
+    Path(folder_path).mkdir(parents=True, exist_ok=True)
--- a/dockstream/utils/general_utils.py
+++ b/dockstream/utils/general_utils.py
+import os
+import tempfile
 from dockstream.utils.files_paths import attach_root_path
@@ -44,3 +46,10 @@ def parse_setuppy():
            if "author" in line:
                parsed_dict["author"] = line[line.find('"')+len('"'):line.rfind('"')]
    return parsed_dict
+# note that "text" is True (in contrast to the underlying "mkstemp")
+def gen_temp_file(suffix=None, prefix=None, dir=None, text=True) -> str:
+    filehandler, path = tempfile.mkstemp(suffix=suffix, prefix=prefix, dir=dir, text=text)
+    os.close(filehandler)
+    return path
--- a/target_preparator.py
+++ b/target_preparator.py
@@ -84,7 +84,7 @@ if __name__ == "__main__":
        pdb_prep = PDBPreparator(conf=config)
        # generate a temporary PDB file, that will be the input later
-        _, temp_pdb_file = tempfile.mkstemp(suffix=".pdb")
+        temp_pdb_file = gen_temp_file(suffix=".pdb")
        # apply the specified fixing and set the input PDB file
        pdb_prep.fix_pdb(input_pdb_file=input_pdb_path,

--- a/tests/test_PDBPreparation.py
+++ b/tests/test_PDBPreparation.py
@@ -7,6 +7,7 @@ from rdkit import Chem
 from dockstream.core.pdb_preparator import PDBPreparator
 from dockstream.containers.target_preparation_container import TargetPreparationContainer
 from dockstream.utils.enums.target_preparation_enum import TargetPreparationEnum
+from dockstream.utils.general_utils import gen_temp_file
 from tests.tests_paths import PATHS_1UYD
 from dockstream.utils.files_paths import attach_root_path
@@ -33,7 +34,7 @@ class Test_PDBPreparation(unittest.TestCase):
                                                                                self._TE.FIX_REMOVEHETEROGENS: True,
                                                                                self._TE.FIX_MISSINGHYDROGENS: False}}})
        input_pdb_file = attach_root_path(PATHS_1UYD.TARGET_APO_PDB)
-        _, temp_pdb_output = tempfile.mkstemp(suffix=".pdb")
+        temp_pdb_output = gen_temp_file(suffix=".pdb")
        input_mol = Chem.MolFromPDBFile(input_pdb_file, sanitize=True)
        # first processing: add water box and do not fix hydrogens that are missing
@@ -51,7 +52,7 @@ class Test_PDBPreparation(unittest.TestCase):
        os.remove(temp_pdb_output)
        # second processing: no water box and but add hydrogens
-        _, temp_pdb_output = tempfile.mkstemp(suffix=".pdb")
+        temp_pdb_output = gen_temp_file(suffix=".pdb")
        conf[self._TE.TARGETPREP][self._TE.FIX][self._TE.FIX_ADDWATERBOX] = False
        conf[self._TE.TARGETPREP][self._TE.FIX][self._TE.FIX_MISSINGHYDROGENS] = True
        prep = PDBPreparator(conf=conf)
@@ -69,7 +70,7 @@ class Test_PDBPreparation(unittest.TestCase):
                                                                                self._TE.FIX_REMOVEHETEROGENS: True,
                                                                                self._TE.FIX_MISSINGHYDROGENS: True}}})
        input_pdb_file = attach_root_path(PATHS_1UYD.LIGAND_MISSING_PARTS_PDB)
-        _, temp_pdb_output = tempfile.mkstemp(suffix=".pdb")
+        temp_pdb_output = gen_temp_file(suffix=".pdb")
        input_mol = Chem.MolFromPDBFile(input_pdb_file, sanitize=True)
        prep = PDBPreparator(conf=conf)
        self.assertEqual(input_mol.GetNumAtoms(), 1892)