def run_protein_flexibility(pf_obj):
flex_dir = os.path.join(pf_obj.get_path(), "flexibility")
make_directory(flex_dir)
nonwater_tpr = os.path.join(pf_obj.get_path(), pf_obj.get_nonwater_tpr())
nonwater_xtc = os.path.join(pf_obj.get_path(), pf_obj.get_nonwater_xtc())
traj_gro = os.path.join(flex_dir, "".join([pf_obj.get_title_output(),
".",
pf_obj.get_output_sufix_trjconv(),
".gro"]))
rmsf_xvg = os.path.join(flex_dir, "".join(["rmsf_",
pf_obj.get_title_output(),
".",
pf_obj.get_output_sufix_rmsf(),
".xvg"]))
command = "".join(["echo ",
pf_obj.get_echo_trjconv(),
" | ",
gromacs_path.value,
"./gmx trjconv",
" -s ",
nonwater_tpr,
" -f ",
nonwater_xtc,
" -o ",
traj_gro,
" -b ",
str(pf_obj.get_initial_frame_trjconv()),
" -e ",
str(pf_obj.get_final_frame_trjconv())])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
command = "".join(["echo ",
pf_obj.get_echo_rmsf(),
" | ",
gromacs_path.value,
"./gmx rmsf",
" -f ",
nonwater_xtc,
" -s ",
traj_gro,
" -o ",
rmsf_xvg,
" -b ",
str(pf_obj.get_initial_frame_rmsf()),
" -e ",
str(pf_obj.get_final_frame_rmsf()),
" -dt ",
str(time_dt.value),
" -res"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
def run_protein_flexibility(pf_obj):
flex_dir = os.path.join(pf_obj.get_path(), "flexibility")
make_directory(flex_dir)
nonwater_tpr = os.path.join(pf_obj.get_path(),
pf_obj.get_nonwater_tpr())
nonwater_xtc = os.path.join(pf_obj.get_path(),
pf_obj.get_nonwater_xtc())
traj_gro = os.path.join(
flex_dir, "".join([
pf_obj.get_title_output(), ".",
pf_obj.get_output_sufix_trjconv(), ".gro"
]))
rmsf_xvg = os.path.join(
flex_dir, "".join([
"rmsf_",
pf_obj.get_title_output(), ".",
pf_obj.get_output_sufix_rmsf(), ".xvg"
]))
command = "".join([
"echo ",
pf_obj.get_echo_trjconv(), " | ", gromacs_path.value,
"./gmx trjconv", " -s ", nonwater_tpr, " -f ", nonwater_xtc,
" -o ", traj_gro, " -b ",
str(pf_obj.get_initial_frame_trjconv()), " -e ",
str(pf_obj.get_final_frame_trjconv())
])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
command = "".join([
"echo ",
pf_obj.get_echo_rmsf(), " | ", gromacs_path.value, "./gmx rmsf",
" -f ", nonwater_xtc, " -s ", traj_gro, " -o ", rmsf_xvg, " -b ",
str(pf_obj.get_initial_frame_rmsf()), " -e ",
str(pf_obj.get_final_frame_rmsf()), " -dt ",
str(time_dt.value), " -res"
])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
sqlCtx = SQLContext(sc)
config = configparser.ConfigParser()
config.read('config.ini')
# Vina configuration for broadcast
config_vina = config.get('VINA', 'config_file')
vina_path = config.get('VINA', 'vina_program')
pdbqt_ligand_path = config.get('DEFAULT', 'pdbqt_ligand_path')
pdbqt_receptor_path = config.get('DEFAULT', 'pdbqt_receptor_path')
path_save_output = config.get('DEFAULT', 'path_save_structure')
path_save_log = config.get('DEFAULT', 'path_save_log')
path_spark_drugdesign = config.get('DRUGDESIGN', 'path_spark_drugdesign')
path_save_log = preparing_path(path_save_log)
make_directory(path_save_log)
path_save_output = preparing_path(path_save_output)
make_directory(path_save_output)
# Adding Python Source file
sc.addPyFile(os.path.join(path_spark_drugdesign, "docking_description.py"))
# Broadcast
vina_path = sc.broadcast(vina_path)
pdbqt_ligand_path = sc.broadcast(pdbqt_ligand_path)
pdbqt_receptor_path = sc.broadcast(pdbqt_receptor_path)
path_save_output = sc.broadcast(path_save_output)
path_save_log = sc.broadcast(path_save_log)
sc.addFile(config_vina)
def run_trajetory(md_obj):
ana_dir = os.path.join(md_obj.get_path(), "analysis")
make_directory(ana_dir)
# Original file names from the simulation
reference_xtc = os.path.join(md_obj.get_path(),
md_obj.get_simulation_prefix() + ".xtc")
reference_tpr = os.path.join(md_obj.get_path(),
md_obj.get_simulation_prefix() + ".tpr")
# File names after trajectory treatment.
allatom_xtc = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_fit.",
str(md_obj.get_repetion_number()),
".xtc"]))
allatom_tpr = reference_tpr
nonwater_xtc = os.path.join(ana_dir,"".join([md_obj.get_prefix_ref(),
"_non-water.",
str(md_obj.get_repetion_number()),
".xtc"]))
nonwater_tpr = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_non-water.",
str(md_obj.get_repetion_number()),
".tpr"]))
nonwater_pdb = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_non-water.",
str(md_obj.get_repetion_number()),
".pdb"]))
waterlayer_pdb = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_water-",
str(water_layer_thickness.value),
"A-layer.",
str(md_obj.get_repetion_number()),
".pdb"]))
# Trajectory treatment to remove PBC artifacts
xtc_whole = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_whole.",
str(md_obj.get_repetion_number()),
".xtc"]))
command = "".join(["echo System | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
reference_xtc,
" -s ",
reference_tpr,
" -pbc whole",
" -o ",
xtc_whole,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Extracting first frame
gro_first_frame = os.path.join(ana_dir, "".join(["0.",
str(md_obj.get_repetion_number()),
".gro"]))
command = "".join(["echo System | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
xtc_whole,
" -s ",
reference_tpr,
" -e 0.1 ",
" -o ",
gro_first_frame,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Removing jumps
xtc_nojump = os.path.join(ana_dir,
"".join([md_obj.get_prefix_ref(),
"_nojump.",
str(md_obj.get_repetion_number()),
".xtc"]))
command = "".join(["echo System | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
xtc_whole,
" -s ",
gro_first_frame,
" -pbc nojump ",
" -o ",
xtc_nojump,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Centering the protein
xtc_center_protein = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_center.",
str(md_obj.get_repetion_number()),
".xtc"]))
command = "".join(["echo C-alpha System | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
xtc_whole,
" -s ",
gro_first_frame,
" -center ",
" -o ",
xtc_center_protein,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Putting all atoms in a compact box
xtc_atoms_box = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_atom.",
str(md_obj.get_repetion_number()),
".xtc"]))
command = "".join(["echo System | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
xtc_center_protein,
" -s ",
gro_first_frame,
" -ur compact ",
" -pbc atom ",
" -o ",
xtc_atoms_box,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Fitting the protein
command = "".join(["echo C-alpha System | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
xtc_atoms_box,
" -s ",
gro_first_frame,
" -fit rot+trans ",
" -o ",
allatom_xtc,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Creating water-free trajectory
command = "".join(["echo non-water | ",
gromacs_path.value,
"./gmx convert-tpr ",
" -s ",
reference_tpr,
" -o ",
nonwater_tpr,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
command = "".join(["echo non-water | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
allatom_xtc,
" -s ",
gro_first_frame,
" -o ",
nonwater_xtc,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
command = "".join(["echo system | ",
gromacs_path.value,
"./gmx trjconv ",
" -f ",
nonwater_xtc,
" -s ",
nonwater_tpr,
" -o ",
nonwater_pdb,
" -dt ",
str(time_dt_pdb.value),
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Creating water_layer_thickness - A water-layer pdb trajectory
t = 0
frame = 0
ndx_water_layer = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_water-layer.",
str(md_obj.get_repetion_number()),
".ndx"]))
ndx_temporary = os.path.join(ana_dir, "".join([md_obj.get_prefix_ref(),
"_temporary_",
str(md_obj.get_repetion_number()),
".ndx"]))
if os.path.isfile(waterlayer_pdb):
os.remove(waterlayer_pdb)
if os.path.isfile(ndx_water_layer):
os.remove(ndx_water_layer)
select_string = ('\'"water_layer" (same residue as ((resname SOL and within 0.'"$water_layer_thickness"' of group "Protein"))) or\
(group "Ion" and within 0.'"$water_layer_thickness"' of group "Protein") \
or (group "Protein") \'')
select_string = select_string.replace("$water_layer_thickness",
str(water_layer_thickness.value))
# Running make_ndx
command = "".join(["echo -e ",
"'chain z'\"\\n\"'q'\"\\n\" | ",
gromacs_path.value,
"gmx make_ndx ",
"-f ",
reference_tpr,
" -o ",
ndx_temporary,
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Are there ligands?
if search_for_ligand_ndx_file(ndx_temporary) is True:
select_string = (select_string
+ '\'or (same residue as ((resname SOL and within 0.'"$water_layer_thickness"' of group "Other"))) \
or (group "Ion" and within 0.'"$water_layer_thickness"' of group "Other") \
or (group "Other")\'')
select_string = select_string.replace("$water_layer_thickness",
str(water_layer_thickness.value))
command = "".join([gromacs_path.value,
"gmx select -f ",
allatom_xtc,
" -s ",
allatom_tpr,
" -on ",
ndx_water_layer,
" -select ",
select_string,
" -dt ",
str(time_dt_pdb.value),
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Creating pdb files
command = "".join(["echo ",
str(frame),
" | ",
gromacs_path.value,
"./gmx trjconv ",
"-f ",
allatom_xtc,
" -s ",
allatom_tpr,
" -n ",
ndx_water_layer,
" -o ",
"frame_",
str(frame),
".pdb ",
"-b ",
str(t),
" -e ",
str(t),
" >/dev/null 2>/dev/null"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
command = "".join(["echo MODEL ", str(frame), " >> ", waterlayer_pdb])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
command = "".join(["grep ATOM ",
"frame_",
str(frame),
".pdb ",
">> ",
waterlayer_pdb])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
command = "".join(["echo ENDML", ">> ", waterlayer_pdb])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Removing temporary files
command = "".join(["rm frame_", str(frame), ".pdb"])
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
frame = frame + 1
t = t + int(time_dt_pdb.value)
if os.path.isfile(xtc_whole):
os.remove(xtc_whole)
if os.path.isfile(xtc_nojump):
os.remove(xtc_nojump)
if os.path.isfile(xtc_center_protein):
os.remove(xtc_center_protein)
if os.path.isfile(xtc_atoms_box):
os.remove(xtc_atoms_box)
if os.path.isfile(ndx_water_layer):
os.remove(ndx_water_layer)
if os.path.isfile(gro_first_frame):
os.remove(gro_first_frame)
command = "rm \#* 2>/dev/null"
proc = Popen(command, shell=True, stdout=PIPE)
proc.communicate()
# Basic Analysis
basic_an_data = (gromacs_path.value,
nonwater_xtc,
nonwater_tpr,
md_obj.get_simulation_prefix(),
ana_dir,
time_dt.value)
run_basic_analysis(basic_an_data)
key_name = json_name.split('.')[0]
json_data = create_jsondata_from_docking_output_file(log_file)
json_final_data = {key_name: json_data}
json_name = os.path.join(path_to_save, json_name )
create_json_file(json_name, json_final_data)
if __name__ == '__main__':
config = configparser.ConfigParser()
config.read('config.ini')
sc = SparkContext()
sqlCtx = SQLContext(sc)
log_file = config.get('DEFAULT', 'path_save_log')
path_to_save = config.get('DEFAULT', 'json_log')
path_spark_drugdesign = config.get('DRUGDESIGN', 'path_spark_drugdesign')
sc.addPyFile(os.path.join(path_spark_drugdesign,"vina_utils.py"))
sc.addPyFile(os.path.join(path_spark_drugdesign,"json_utils.py"))
sc.addPyFile(os.path.join(path_spark_drugdesign,"os_utils.py"))
#Broadcast
log_file_b = sc.broadcast(log_file)
path_to_save_b = sc.broadcast(path_to_save)
make_directory(path_to_save)
all_log_files = get_files_log(log_file)
log_filesRDD = sc.parallelize(all_log_files)
log_filesRDD.foreach(log_to_json)
sqlCtx = SQLContext(sc)
config = configparser.ConfigParser()
config.read('config.ini')
# Vina configuration for broadcast
config_vina = config.get('VINA', 'config_file')
vina_path = config.get('VINA', 'vina_program')
pdbqt_ligand_path = config.get('DEFAULT', 'pdbqt_ligand_path')
pdbqt_receptor_path = config.get('DEFAULT', 'pdbqt_receptor_path')
path_save_output = config.get('DEFAULT', 'path_save_structure')
path_save_log = config.get('DEFAULT', 'path_save_log')
path_spark_drugdesign = config.get('DRUGDESIGN', 'path_spark_drugdesign')
path_save_log = preparing_path(path_save_log)
make_directory(path_save_log)
path_save_output = preparing_path(path_save_output)
make_directory(path_save_output)
# Adding Python Source file
sc.addPyFile(os.path.join(path_spark_drugdesign, "docking_description.py"))
# Broadcast
vina_path = sc.broadcast(vina_path)
pdbqt_ligand_path = sc.broadcast(pdbqt_ligand_path)
pdbqt_receptor_path = sc.broadcast(pdbqt_receptor_path)
path_save_output = sc.broadcast(path_save_output)
path_save_log = sc.broadcast(path_save_log)
sc.addFile(config_vina)
if __name__ == '__main__':
sc = SparkContext()
sqlCtx = SQLContext(sc)
config = configparser.ConfigParser()
config.read('config.ini')
pythonsh = config.get('VINA', 'pythonsh')
script_receptor4 = config.get('VINA', 'script_receptor4')
pdb_path = config.get('DEFAULT', 'pdb_path')
pdbqt_receptor_path = config.get('DEFAULT', 'pdbqt_receptor_path')
path_spark_drugdesign = config.get('DRUGDESIGN', 'path_spark_drugdesign')
make_directory(pdbqt_receptor_path)
# Adding Python Source file
sc.addPyFile(os.path.join(path_spark_drugdesign, "vina_utils.py"))
sc.addPyFile(os.path.join(path_spark_drugdesign, "json_utils.py"))
sc.addPyFile(os.path.join(path_spark_drugdesign, "os_utils.py"))
# Broadcast
pythonsh = sc.broadcast(pythonsh)
script_receptor4 = sc.broadcast(script_receptor4)
pdbqt_receptor_path = sc.broadcast(pdbqt_receptor_path)
def run_prepare_receptor_spark(receptor):
receptor_pdbqt = os.path.join(pdbqt_receptor_path.value,
get_name_model_pdb(receptor))