def _edit_top(ligand_top, only_solvent_top):
only_solvent_lines = read_file.read_file(file_name=only_solvent_top)
for i in range(len(only_solvent_lines) - 1, -1, -1):
if only_solvent_lines[i].strip():
solvent_line = only_solvent_lines[i].rstrip()
break
del only_solvent_lines
ligand_lines = read_file.read_file(file_name=ligand_top)
for i in range(len(ligand_lines) - 1, -1, -1):
if ligand_lines[i].strip():
ligand_lines[i] = solvent_line + '\n' + ligand_lines[i]
break
write_on_files.write_file(lines=ligand_lines, file_name=ligand_top)
def test_wrong_input(self):
wrong_inputs = (1, 1.5)
with self.assertRaises(TypeError):
for i in wrong_inputs:
read_file.read_file(file_name = i)
def test_with_file(self):
file_name = "tests/files4tests/file_read_file.txt"
lines = read_file.read_file(file_name = file_name)
self.assertEqual(lines, ["aaa\n", "BBB\n", "111\n", "2.2.2\n"])
def add_chain_id(pdb_file, chain="A"):
"""
This is a patch because orac and primadorac remove the chain id from
pdb files and this confuses some pdb parsers (works on PDB files only)
pdb_file :: string, the pdb file to edit
chain :: string, default A, the chain id to add to the pdb_file
returns nothing
"""
chain = chain.upper().strip()
lines = read_file.read_file(file_name=pdb_file)
for i in range(len(lines)):
if lines[i][0:4] == 'ATOM' or lines[i][0:6] == 'HETATM' or lines[i][
0:3] == 'TER':
lines[i] = lines[i][:20] + "{0:>2}".format(chain) + lines[i][22:]
lines[i] = lines[i].strip('\n') + '\n'
write_on_files.write_file(lines=lines, file_name=pdb_file)
def _get_ligand_name_from_tpg(self):
"""
private
Gets the name of the ligand from the tpg file
for any given ligand
"""
Ligand = self.Protein.get_ligand_list()
if Ligand == [] or Ligand == None:
return ""
residue_strings = []
for ligand in Ligand:
lines = read_file.read_file(file_name=ligand.tpg_file)
for line in lines:
if 'RESIDUE' in line:
string = f" {line.split()[1].strip()} !! ligand name in tpg file"
residue_strings.append(string)
break
else:
raise RuntimeError(
f'Could not find the residue name in {ligand.tpg_file}')
return '\n'.join(residue_strings)
def _change_absolute_ligand_itp_include_in_relative(self, top_file):
"""
private
changes the #include of the ligand itp file in the protein top from the absolute one in the relative one
in the given topology file in order to get a meaningful #include in the self.HREM_dir directory
that will be copied on a different Computer (the HPC cluster)
"""
lines = read_file.read_file(file_name = top_file)
ligand_itp_files = []
for lgand in self.Protein.get_ligand_list():
ligand_itp_files.append(lgand.itp_file)
for i in range(len(lines)):
if lines[i].strip()[:8] == "#include":
if lines[i].split()[1].replace('"', '').strip() in ligand_itp_files or 'DUM' in lines[i]:
itp_file = lines[i].split()[1].replace('"', '').strip()
itp_file = itp_file.split("/")[-1]
lines[i] = f'#include "{itp_file}"\n'
write_on_files.write_file(lines = lines, file_name = top_file)
def _get_ligand_name_from_tpg(self, Ligand):
"""
private
Gets the name of the ligand from the tpg file
"""
lines = read_file.read_file(file_name=Ligand.tpg_file)
for line in lines:
if 'RESIDUE' in line:
string = f" {line.split()[1].strip()} !! ligand name in tpg file"
break
else:
raise RuntimeError(
f'Could not find the residue name in {Ligand.tpg_file}')
return string
def _edit_itp(self, ligand_resname, itp_file):
"""
private
primadorac itp call any lignd LIG i change it to the ligand_resname
and removes the first 9 lines of the file (they make gromacs fail)
"""
lines = read_file.read_file(file_name=itp_file)
del lines[0:9]
for i in range(len(lines)):
lines[i] = lines[i].replace('LIG', ligand_resname)
lines[i] = lines[i].replace('name-p', ligand_resname)
lines[i] = lines[i].strip('\n')
lines[i] = lines[i] + '\n'
write_on_files.write_file(lines=lines, file_name=itp_file)
return path.absolute_filepath(path=itp_file)
def _edit_top_file(self):
"""
Private
Adds the needed #include and other informations
to the protein top file in order
to include the ligand
"""
Ligand = self.Protein.get_ligand_list()
top = read_file.read_file(file_name=self.Protein.top_file)
itp_insertion_string = ''
for lgand in Ligand:
itp_insertion_string = itp_insertion_string + f'#include "{lgand.itp_file}"\n'
compound_string = f'{lgand.resname} 1 \n'
top.append(compound_string)
for i in range(len(top)):
if top[i].strip() == "":
pass
elif top[i].strip()[0] == ";":
pass
elif top[i].strip()[0:8] == "#include":
top[i] = top[i] + '\n' + itp_insertion_string + '\n'
break
elif top[i].strip().replace(" ",
"").split(";")[0] == '[moleculetype]':
top[i] = itp_insertion_string + '\n' + top[i]
break
write_on_files.write_file(lines=top, file_name=self.output_top_file)
self.Protein.top_file = self.output_top_file
return self.Protein
def execute(self):
if self.HREM_dir != os.getcwd():
os.chdir(self.HREM_dir)
fsdam_dir = "RESTART"
os.makedirs(fsdam_dir, exist_ok=True)
os.makedirs(f'{fsdam_dir}/not_used_frames', exist_ok=True)
#making some defaults
useful_info = {
"ligand_resname": 'LIG',
"top_file": "topol.top",
"ligand_itp": "LIG.itp",
f"only_solvent_gro": None,
f"only_solvent_top": None
}
lines = read_file.read_file(file_name="important_info.dat")
for line in lines:
line = line.strip()
if line:
line = line.split("=")
useful_info[line[0].strip()] = line[1].strip()
starting_configurations = self._create_restart_configs(
fsdam_dir=fsdam_dir)
starting_configurations = self._select_frames_to_use(
starting_configurations,
not_used_dir=f'{fsdam_dir}/not_used_frames')
if self.creation:
self._add_water_box(useful_info['only_solvent_gro'],
starting_configurations)
shutil.copy(useful_info["top_file"], 'ligand_solvent_topology.top')
useful_info["top_file"] = 'ligand_solvent_topology.top'
self._edit_top(ligand_top=useful_info["top_file"],
only_solvent_top=useful_info["only_solvent_top"])
output_dictionary = self._make_input_files(
ligand_resname=useful_info["ligand_resname"],
top_file=useful_info["top_file"],
starting_structures=starting_configurations,
)
#writes the TPR scripts in the FSDAM dir
self._make_TPR_files_script(
q_lines=output_dictionary['make_q_tpr'],
vdw_lines=output_dictionary['make_vdw_tpr'],
fsdamdir=fsdam_dir)
#lazily copy all the mdp, itp and top files in the new dir
for i in os.listdir(os.getcwd()):
if i[-4:] in ('.mdp', '.itp', '.top'):
shutil.copy(i, fsdam_dir)
#write a suggestion of run file
with open(f'{fsdam_dir}/RUN_Q.sh', 'w') as f:
number_of_runs = len(output_dictionary['run_q'])
f.write(f'# there are {number_of_runs} runs to do\n')
f.write('# I suggest to use one GPU per run\n\n\n')
f.write('\n'.join(output_dictionary['run_q']))
with open(f'{fsdam_dir}/RUN_VDW.sh', 'w') as f:
number_of_runs = len(output_dictionary['run_vdw'])
f.write(f'# there are {number_of_runs} runs to do\n')
f.write(
'# I suggest to use one GPU per run or even better a job array\n\n\n'
)
f.write('\n'.join(output_dictionary['run_vdw']))
def _edit_top_file(self, top_file = None):
"""PRIVATE"""
if top_file is None:
top_file = self.elaborated_top_file
#get the hot residues
hot_residues = self.orient.get_hot_residues_for_rem(Protein = self.Protein, Ligand = self.Protein.get_ligand_list(), cutoff = 4.5, residue_dist = 10.0)
hot_ids = []
for residue in hot_residues:
hot_ids.append(str(residue[1]).strip())
#get the ligand resnames
ligands_resnames = []
for lgand in self.Protein.get_ligand_list():
ligands_resnames.append(lgand.resname)
#read the topology
lines = read_file.read_file(file_name = top_file)
#heat the right atoms
#auxiliary bool variable
is_atoms = False
for i in range(len(lines)):
#if the line is empty or a comment i can go on with the for loop
if lines[i].strip() != "":
if lines[i].strip()[0] != ";":
tmp_line = lines[i].strip().split()
else:
continue
else:
continue
#check if we are in the atoms part
if lines[i].strip().replace(" ", "").split(";")[0] == "[atoms]":
is_atoms = True
continue
#end of atoms part
elif tmp_line[0][0] == "[":
is_atoms = False
continue
if is_atoms:
if len(tmp_line) >= 4:
#do not heat solvent
if tmp_line[3].strip() == "SOL":
continue
#heat the ligand
if tmp_line[3].strip() in ligands_resnames:
tmp_line[1] = tmp_line[1] + "_"
lines[i] = " ".join(tmp_line) + "\n"
#heat the near residues
elif tmp_line[2].strip() in hot_ids:
tmp_line[1] = tmp_line[1] + "_"
lines[i] = " ".join(tmp_line) + "\n"
write_on_files.write_file(lines = lines, file_name = top_file)
def merge_pdb(Protein):
"""
Will put all the given ligands after the protein and update the ligand resnums
this function is brutal and memory consuming I should do it better in the future
both the protein and the ligands should be in PDB files (no check will be done)
Protein :: HPC_Drug.structures.protein.Protein instance with a valid _ligands value
return Protein with updated Protein.pdb_file
"""
protein_file = read_file.read_file(file_name=Protein.pdb_file)
ligands = Protein.get_ligand_list()
#get the index of the line with the last ATOM HETATM or TER line
#and get the resnum of this last residue
for i in range(len(protein_file) - 1, -1, -1):
if protein_file[i][0:4] == 'ATOM' or protein_file[i][
0:6] == 'HETATM' or protein_file[i][0:3] == 'TER':
#some TER lines are non standard and don't contain the residue number
try:
residue_number = int(protein_file[i][22:26].strip())
except:
residue_number = int(protein_file[i - 1][22:26].strip())
index_protein_file = i + 1
break
#create the ligands list of strings
ligand_file = []
for j in range(len(ligands)):
residue_number = residue_number + 1
#update resnum
ligands[j].resnum = residue_number
tmp_ligand = read_file.read_file(file_name=ligands[j].pdb_file)
#update the residue numbers in the file
for k in range(len(tmp_ligand)):
tmp_ligand[k] = tmp_ligand[k][:22] + "{0:>4}".format(
residue_number) + tmp_ligand[k][26:]
ligand_file = ligand_file + tmp_ligand
#insert the ligands in the right place of the protein_file list
protein_file[index_protein_file:index_protein_file] = ligand_file
#be sure to get the right formatting
for i in range(len(protein_file)):
protein_file[i] = protein_file[i].strip('\n') + '\n'
#overwrite Protein.pdb_file
write_on_files.write_file(lines=protein_file,
file_name=f"{Protein.protein_id}_joined.pdb")
Protein.pdb_file = path.absolute_filepath(
path=f"{Protein.protein_id}_joined.pdb")
return Protein
def remove_trash_metal_ions(Protein, trash = important_lists.trash_ions):
"""This function removes unwanted metal ions
that are still inside the structure after it went through prody
selection (updates Protein.pdb_file)
This is a brutal function I will need to do a better job
Protein :: HPC_Drug.structures.protein.Protein instance
Protein.file_type must be pdb or cif otherwise TypeError will be raised
return Protein
"""
def determine_pdb(line, trash = trash):
#if it is a line containing atom coordinates
if line[0:4] == 'ATOM' or line[0:6] == 'HETATM' or line[0:3] == 'TER':
bool_output = line[17:20].strip().upper() not in trash
#if it is not will not be kept anyway
else:
bool_output = False
return bool_output
def determine_cif(line, trash = trash):
_line = line.split()
#if it is a line containing atom coordinates
if _line[0].strip() == 'ATOM' or _line[0].strip() == 'HETATM':
bool_output = _line[5].strip().upper() not in trash
#if it is not will be kept anyway
else:
bool_output = True
return bool_output
file_name = Protein.pdb_file
lines = read_file.read_file(file_name = file_name)
#keeping only the "good" lines
if Protein.file_type == 'pdb':
lines[:] = [x for x in lines if determine_pdb(x)]
lines.append("end")
elif Protein.file_type == 'cif':
lines[:] = [x for x in lines if determine_cif(x)]
else:
raise TypeError(f"Protein.file_type must be pdb or cif not {Protein.file_type}")
for i in range(len(lines)):
lines[i] = lines[i].strip() + '\n'
write_on_files.write_file(lines = lines, file_name = file_name)
Protein.pdb_file = file_name
return Protein