Merge pull request #61 from KULL-Centre/ligand-bug

software/rosetta_ddG_pipeline/README.md

@@ -28,7 +28,7 @@ To install the pipeline, get it from the GitHub page.
 An installation file is not yet available.
 
 
-The following links can be included in your bashrc, otherwise the default will be used:
+The following links can be included in your bashrc, otherwise the default (found in rosetta_path.py) will be used:
 ```bash
 export Rosetta_main_path= ‘{Newest Rosetta version}’
 export Rosetta_tools_path='{path to Rosetta tools}’
@@ -79,6 +79,8 @@ General flags
     --verbose increase verbose level
     --gapped_output output pism and pdb file with shifted residue numbering according to input
     --dump_pdb dumps all mutant pdbs
+    -nzip skip folder zipping (default off=false) 
+    --do_checks does concistency checks after prepare and relax (default true)
 ```
 
 Flags for membrane proteins:

software/rosetta_ddG_pipeline/args_pipeline.py

@@ -109,10 +109,10 @@ def parse_args2():
                         help='Dumps all mutant pdbs, default False.'
                         )
     parser.add_argument('--skip_zip', '-nzip',
-                        default=True,
-                        type=lambda s: s.lower() in ['false', 'f', 'no', '0'],
-                        dest='ZIP_FILES',
-                        help='Switch folder zipping off. default on'
+                        default=False,
+                        type=lambda s: s.lower() in ['true', 't', 'yes', '1'],
+                        dest='NO_ZIP',
+                        help='Skip folder zipping. default off'
                         )
     parser.add_argument('--do_checks',
                         default=True,
@@ -348,8 +348,12 @@ def parse_args2():
         args.MUT_MODE = 'mut_file'
     if args.DUMP_PDB != 0:
         args.DUMP_PDB = 1
-    if args.ZIP_FILES != True:
+    if args.NO_ZIP != False:
         args.ZIP_FILES = False
+    else:
+        args.ZIP_FILES = True
+    if args.LIGAND != None:
+        args.LIGAND = True
     if args.IS_MP == False:
         args.MP_MULTISTRUC_PROTOCOL == 0
     if args.IS_MP:

software/rosetta_ddG_pipeline/relax_parse_results.py

@@ -85,7 +85,7 @@ def parse_relax_results(folder, pdb_id='', sc_name='score_bn15_calibrated', logg
 
     # checking for consitency
     if do_checking:
-        input_pdb = glob.glob(os.path.join(folder.input, '*.pdb'))[0]
+        input_pdb = glob.glob(os.path.join(folder.relax_input, '*.pdb'))[0]
         output_pdb = os.path.join(folder.relax_run, f'{most_relaxed}.pdb')
         chain = 'A'
         with open(output_pdb, 'r') as fp:
@@ -246,10 +246,10 @@ def check_consitency(input_pdb, output_pdb, chain='A', is_MP=True, pdb_id='-'):
         pdb_id = sys.argv[3]
 
     if len(sys.argv) <= 9:
-        folder.update({'input': sys.argv[4], 'relax_run': sys.argv[5], 'relax_output': sys.argv[6], 
+        folder.update({'relax_input': sys.argv[4], 'relax_run': sys.argv[5], 'relax_output': sys.argv[6], 
             'ddG_input': sys.argv[7]})
     else:
-        folder.update({'input': sys.argv[4], 'relax_run': sys.argv[5], 'relax_output': sys.argv[6], 
+        folder.update({'relax_input': sys.argv[4], 'relax_run': sys.argv[5], 'relax_output': sys.argv[6], 
             'ddG_input': [sys.argv[x] for x in range(7, len(sys.argv)-2)]})
     print(sys.argv)
     print(folder)

software/rosetta_ddG_pipeline/run_pipeline.py

@@ -123,7 +123,7 @@ def predict_stability(args):
 
         # Cleaning pdb and making fasta based on pdb or uniprot-id if provided
         logger.info(f'Prepare the pdb and extract fasta file')
-        structure_instance.path_to_cleaned_pdb, struc_dic_cleaned = structure_instance.clean_up_and_isolate()
+        structure_instance.path_to_cleaned_pdb, struc_dic_cleaned = structure_instance.clean_up_and_isolate(ligand=args.LIGAND)
         structure_instance.fasta_seq_full,structure_instance.fasta_seq = pdb_to_fasta_seq(
             structure_instance.path_to_cleaned_pdb,chain_id)
         if os.path.isfile(uniprot_accesion):
@@ -275,8 +275,8 @@ def predict_stability(args):
                     folder,  partition=partition, output_gaps=args.GAPS_OUTPUT, zip_files=args.ZIP_FILES, sha_tag=SHA_TAG, is_MP=args.IS_MP, scale_factor=args.SCALE_FACTOR)
                 else:
                     path_to_parse_ddg_sbatch = mp_ddG.write_parse_rosetta_ddg_mp_pyrosetta_sbatch(
-                    folder, chain_id=args.CHAIN, sys_name=name, output_name='ddG.out', add_output_name='ddG_additional.out', partition=partition, 
-                    output_gaps=args.GAPS_OUTPUT, mp_multistruc=args.MP_MULTISTRUC_PROTOCOL, zip_files=args.ZIP_FILES, sha_tag=SHA_TAG)
+                        folder, chain_id=args.CHAIN, sys_name=name, output_name='ddG.out', add_output_name='ddG_additional.out', partition=partition, 
+                        output_gaps=args.GAPS_OUTPUT, mp_multistruc=args.MP_MULTISTRUC_PROTOCOL, zip_files=args.ZIP_FILES, sha_tag=SHA_TAG)
         else:
             # Parse sbatch relax file
             relax_input_relaxfile = check_path(create_copy(
@@ -294,7 +294,8 @@ def predict_stability(args):
             ddg_input_mutfile_dir = create_copy(
                 prepare_output_ddg_mutfile_dir, folder.ddG_input, name='mutfiles', directory=True)
             path_to_ddg_calc_sbatch = structure_instance.write_rosetta_cartesian_ddg_sbatch(
-                folder, ddg_input_mutfile_dir, ddgfile=ddg_input_ddgfile, sys_name=name,  partition=partition)
+                folder, ddg_input_mutfile_dir, ddgfile=ddg_input_ddgfile, sys_name=name,  
+                partition=partition, dump_pdb=args.DUMP_PDB)
             # Parse sbatch ddg parser
             path_to_parse_ddg_sbatch = structure_instance.write_parse_cartesian_ddg_sbatch(
                 folder,  partition=partition, output_gaps=args.GAPS_OUTPUT, zip_files=args.ZIP_FILES, sha_tag=SHA_TAG, is_MP=args.IS_MP, scale_factor=args.SCALE_FACTOR)

software/rosetta_ddG_pipeline/structure_input.py

@@ -77,12 +77,13 @@ def clean_up_and_isolate(self, name='input',ligand=None):
         #This cleans the protein but keeps the ligands          
         if ligand == True:
             self.path_to_clean_pdb = rosetta_paths.path_to_clean_keep_ligand
+            self.tmp_prep_struc = create_copy(self.prep_struc, self.folder.prepare_cleaning, name='withHETATM.pdb')
             #Runs shell script
-            shell_command = f'python2 {self.path_to_clean_pdb} {self.prep_struc} {self.chain_id}  --keepzeroocc'
+            shell_command = f'python2 {self.path_to_clean_pdb} {self.tmp_prep_struc} {self.chain_id}  --keepzeroocc'
             self.logger.info('Running clean_pdb_keep_ligand.py script')
             subprocess.call(shell_command, cwd=self.folder.prepare_cleaning, shell=True)
             self.logger.info('end of output from clean_pdb_keep_ligand.py')
-            path_to_cleaned_pdb = os.path.join(self.folder.prepare_cleaning, f'{name}.pdb{self.chain_id}.pdb')
+            path_to_cleaned_pdb = f'{self.tmp_prep_struc}{self.chain_id}.pdb'
         #Creates struc.json from cleaned pdb file           
         struc_dic_cleaned= get_structure_parameters(
             self.folder.prepare_cleaning, path_to_cleaned_pdb,printing=False)
@@ -422,17 +423,18 @@ def parse_relax_sbatch(self, folder, sys_name='', partition='sbinlab', sc_name='
 # launching parsing script 
 ''')
             if mp_multistruc == 0:
-                fp.write(f'python {path_to_parse_relax_script} {do_checking} {is_MP} {ref_pdb} {folder.input} {folder.relax_run} {folder.relax_output} {folder.ddG_input} {sc_name}')
+                fp.write(f'python {path_to_parse_relax_script} {do_checking} {is_MP} {ref_pdb} {folder.relax_input} {folder.relax_run} {folder.relax_output} {folder.ddG_input} {sc_name}')
             else:
-                fp.write(f'python {path_to_parse_relax_script} {do_checking} {is_MP} {ref_pdb} {folder.input} {folder.relax_run} {folder.relax_output}' )
+                fp.write(f'python {path_to_parse_relax_script} {do_checking} {is_MP} {ref_pdb} {folder.relax_input} {folder.relax_run} {folder.relax_output}' )
                 for ddg_subfolder in folder.ddG_input:
                     fp.write(f' {ddg_subfolder}')
                 fp.write(f' {sc_name} {mp_multistruc}')
         self.logger.info(path_to_sbatch)
         return path_to_sbatch
 
 
-    def write_rosetta_cartesian_ddg_sbatch(self, folder, input_mutfiles='', ddgfile='', sys_name='', partition='sbinlab'):
+    def write_rosetta_cartesian_ddg_sbatch(self, folder, input_mutfiles='', ddgfile='', sys_name='', 
+        partition='sbinlab', dump_pdb=0):
         """This script creates the rosetta_dddg.sbatch script"""
 
         path_to_sbatch = os.path.join(self.folder.ddG_input, 'rosetta_ddg.sbatch')
@@ -466,7 +468,8 @@ def write_rosetta_cartesian_ddg_sbatch(self, folder, input_mutfiles='', ddgfile=
 ''')
             fp.write((f'{os.path.join(rosetta_paths.path_to_rosetta, f"bin/cartesian_ddg.{rosetta_paths.Rosetta_extension}")} '
                       f'-s {structure_path} -ddg:mut_file ${{LST[$INDEX]}} '
-                      f' -out:prefix ddg-$SLURM_ARRAY_JOB_ID-$SLURM_ARRAY_TASK_ID @{path_to_ddgflags}'))
+                      f' -out:prefix ddg-$SLURM_ARRAY_JOB_ID-$SLURM_ARRAY_TASK_ID @{path_to_ddgflags}'
+                      f' -ddg::dump_pdbs {dump_pdb} '))
         self.logger.info(path_to_sbatch)
         return path_to_sbatch