def launch_from_gui(self):
# If there is no internet connection, everything is blocked.
if not check_network_connection('https://www.google.it/'):
self.pymod.main_window.show_error_message(
"Connection Error",
"No internet connection. Can not update PyMod databases.")
return None
# Associates the correct installer to each component.
self.components_dict_installer = {}
self.components_list = []
# Builds Qthread objects.
_all_components_list = all_components_list[:]
if self.pymod.use_blast_v5_databases:
_all_components_list.pop(0) # Remove BLAST v4 databases.
else:
_all_components_list.pop(1) # Remove BLAST v5 databases.
for comp, installer_class in zip(
_all_components_list,
[BLAST_db_installer, Pfam_db_installer, BLAST_db_installer]):
component_qthread = installer_class(component=comp)
self.components_dict_installer[comp.name] = component_qthread
if comp.name == 'hmmscan_databases':
comp.installer.hmmpress_exe_filepath = os.path.join(
self.pymod.hmmer_tool["exe_dir_path"].get_value(),
get_exe_file_name("hmmpress"))
self.components_list.append(comp)
comp.reinitialize()
# Showing GUI.
self.window = InstallerUpdaterWindow(parent=self.pymod.main_window,
installer_protocol=self)
def get_similariry_search_exe_path(self, return_dict=False):
"""
Get the absolute filepaths of the executables files needed to perform the
similarity search.
"""
if self.protocol_name != "hmmsearch": # Returns only one executable filepath.
exe_filepath_list = [os.path.join(self.tool["exe_dir_path"].get_value(),
get_exe_file_name(self.exe_filename))]
exe_filepath_dict = {self.exe_filename: exe_filepath_list[0]}
else: # Returns two executables filepath.
exe_filepath_list = []
exe_filepath_dict = {}
for exe_filename in self.all_exe_filenames:
exe_filepath = os.path.join(self.tool["exe_dir_path"].get_value(),
get_exe_file_name(exe_filename))
exe_filepath_list.append(exe_filepath)
exe_filepath_dict[exe_filename] = exe_filepath
if return_dict:
return exe_filepath_dict
else:
return exe_filepath_list
def search_domains(self, query_element, database, evaluecutoff):
self._initializes_domains_search(query_element)
self.evaluecutoff = evaluecutoff
exe_filepath = os.path.join(
self.pymod.hmmer_tool["exe_dir_path"].get_value(),
get_exe_file_name("hmmscan"))
out_filepath = os.path.join(
self.output_directory,
"hmmscan_out_" + self.query_element_name + ".txt")
cline = [
exe_filepath, "-o", out_filepath, "-E",
str(evaluecutoff),
database.replace('.h3m', ''), self.query_filepath
]
# self.pymod.new_execute_subprocess(cline)
self.pymod.execute_subprocess(cline, new_shell=False)
return out_filepath
def check_psipred_parameters(
self): # predict_secondary_structure(self, elements=None):
"""
Checks that the files needed to run PSIPRED exists on users' machines.
"""
# First checks for PSIPRED installation.
if not self.pymod.psipred.tool_dir_exists(
parameter_name="exe_dir_path"):
self.pymod.psipred.tool_dir_not_found()
return False
psipred_exe_dirpath = self.pymod.psipred["exe_dir_path"].get_value()
for _exe_filename in ("chkparse", "psipred", "psipass2"):
exe_filename = pmos.get_exe_file_name(_exe_filename)
if not os.path.isfile(
os.path.join(psipred_exe_dirpath, exe_filename)):
message = (
"A PSIPRED executable file ('%s') does not exists in"
" the PSIPRED executables directory specified in the PyMod"
" Options Window ('%s'). If you want to use PSIPRED, please"
" specify in the Options Window a PSIPRED executables directory"
" where a '%s' file is found." %
(exe_filename, psipred_exe_dirpath, exe_filename))
self.pymod.main_window.show_error_message(
self.error_title, message)
return False
# Then checks for PSIPRED datafiles.
if not self.pymod.psipred.tool_dir_exists(
parameter_name="data_dir_path"):
message = (
"PSIPRED 'data' directory not found! Please specify an existent"
" directory in the PSIPRED Options Window of PyMod.")
self.pymod.main_window.show_error_message(self.error_title,
message)
return False
psipred_data_dirpath = self.pymod.psipred["data_dir_path"].get_value()
if len(os.listdir(psipred_data_dirpath)) == 0:
message = (
"PSIPRED 'data' directory is empty! Please specify a data"
" directory actually containing PSIPRED data file in the"
" in the Options Window of PyMod.")
self.pymod.main_window.show_error_message(self.error_title,
message)
return False
# Checks for PSI-BLAST on the user's system.
if not self.pymod.blast_plus.tool_dir_exists(
parameter_name="exe_dir_path"):
self.pymod.blast_plus.tool_dir_not_found()
return False
psiblast_exe_dirpath = self.pymod.blast_plus["exe_dir_path"].get_value(
)
if not os.path.isfile(
os.path.join(psiblast_exe_dirpath,
pmos.get_exe_file_name("psiblast"))):
message = (
"The PSI-BLAST executable file ('psiblast') does not exists in"
" the BLAST+ suite executables directory specified in the PyMod"
" Options Window ('%s'). If you want to use PSIPRED, please"
" specify in the Options Window a BLAST+ suite executables directory"
" where a 'psiblast' file is found." % (psiblast_exe_dirpath))
self.pymod.main_window.show_error_message(self.error_title,
message)
return False
# And finally checks for a BLAST database.
if not self.pymod.psipred.tool_dir_exists(
parameter_name="database_dir_path"):
message = (
"A directory containing a BLAST database was not found! Please"
" specify an existent directory in the PSIPRED options in the options window of PyMod."
)
self.pymod.main_window.show_error_message(self.error_title,
message)
return False
dbpath = self.pymod.psipred["database_dir_path"].get_value()
if not self.verify_valid_blast_dbdir(dbpath, remove_temp_files=True):
message = (
"The database '%s' directory does not contain a valid set"
" of database files." % (dbpath))
self.pymod.main_window.show_error_message(self.error_title,
message)
return False
return True
def run_psipred(self, element):
"""
Actually runs PSIPRED, collects its results and map them on the sequences in PyMod main
window.
"""
print_output = False
sequence_header = element.my_header
if print_output:
print("- Beginning PSIPRED prediction for:", sequence_header)
# The name of the BLAST database file.
# If the database files are contained in a folder like this: /home/user/pymod/databases/swissprot/swissprot
dbpath = self.pymod.psipred["database_dir_path"].get_value(
) # e.g.: /home/user/pymod/databases/swissprot
if print_output:
print("- dbpath:", dbpath)
# Where the NCBI programs have been installed.
ncbidir = self.pymod.blast_plus["exe_dir_path"].get_value()
if print_output:
print("- ncbidir:", ncbidir)
# Where the PSIPRED V2 programs have been installed.
execdir = self.pymod.psipred["exe_dir_path"].get_value()
if print_output:
print("- execdir:", execdir)
# Where the PSIPRED V2 data files have been installed.
datadir = self.pymod.psipred["data_dir_path"].get_value()
if print_output:
print("- datadir", datadir)
# Write the temporary input fasta file, setting its basename.
basename = "psipred_temp"
if print_output:
print("- basename: ", basename)
self.pymod.build_sequence_file(
[element],
basename,
file_format="fasta",
remove_indels=True,
new_directory=self.pymod.psipred_dirpath)
#---------------------
# Execute PSI-BLAST. -
#---------------------
if print_output:
print("- Running PSI-BLAST with sequence", basename, "...")
try:
self.execute_psiblast(
ncbi_dir=ncbidir,
db_path=dbpath,
query=os.path.join(self.pymod.psipred_dirpath,
basename + ".fasta"),
inclusion_ethresh=0.001,
out_pssm=os.path.join(self.pymod.psipred_dirpath,
basename + ".chk"),
out=os.path.join(self.pymod.psipred_dirpath,
basename + ".blast"),
num_iterations=3,
num_alignments=0)
# psiblast_output = open("%s.blast" % os.path.join(self.pymod.psipred_dirpath, basename),"w")
# self.pymod.execute_subprocess(psiblast_command, new_stdout=psiblast_output)
# psiblast_output.close()
except:
if print_output:
print(
"- FATAL: Error whilst running psiblast - script terminated!"
)
raise Exception(
"There was an error while running PSI-BLAST, so PSIPRED cannot perform a prediction for %s."
% (sequence_header))
#--------------------
# Execute chkparse. -
#--------------------
if print_output:
print("- Predicting secondary structure...")
chkdir_command = (
pmos.build_commandline_path_string(
os.path.join(execdir, pmos.get_exe_file_name("chkparse"))) +
" " + pmos.build_commandline_path_string(
"%s.chk" % os.path.join(self.pymod.psipred_dirpath, basename)))
try:
chkdir_output = open(
"%s.mtx" % os.path.join(self.pymod.psipred_dirpath, basename),
"w")
self.pymod.execute_subprocess(chkdir_command,
new_stdout=chkdir_output)
chkdir_output.close()
except:
if print_output:
print(
"- FATAL: Error whilst running chkdir - script terminated!"
)
raise Exception(
"No homologous sequences were found by PSI-BLAST for %s, so PSIPRED cannot perform a prediction for this sequence."
% (sequence_header))
#--------------------------
# Execute PSIPRED pass 1. -
#--------------------------
psipass1_command = (
pmos.build_commandline_path_string(
os.path.join(execdir, pmos.get_exe_file_name("psipred"))) +
" " + pmos.build_commandline_path_string(
"%s.mtx" % os.path.join(self.pymod.psipred_dirpath, basename))
+ " " + pmos.build_commandline_path_string(
os.path.join(datadir, "weights.dat")) + " " +
pmos.build_commandline_path_string(
os.path.join(datadir, "weights.dat2")) + " " +
pmos.build_commandline_path_string(
os.path.join(datadir, "weights.dat3")))
try:
psipass1_output = open(
"%s.ss" % os.path.join(self.pymod.psipred_dirpath, basename),
"w")
self.pymod.execute_subprocess(psipass1_command,
new_stdout=psipass1_output)
psipass1_output.close()
except Exception as e:
if print_output:
print(
"- FATAL: Error whilst running psipred 1 - script terminated!"
)
raise Exception(
"There was an error while running PSIPRED and no prediction was made for %s."
% (sequence_header))
#--------------------------
# Execute PSIPRED pass 2. -
#--------------------------
psipass2_command = (
pmos.build_commandline_path_string(
os.path.join(execdir, pmos.get_exe_file_name("psipass2"))) +
" " + "%s 1 1.0 1.0" % pmos.build_commandline_path_string(
os.path.join(datadir, "weights_p2.dat")) + " " +
pmos.build_commandline_path_string(
"%s.ss2" % os.path.join(self.pymod.psipred_dirpath, basename))
+ " " + pmos.build_commandline_path_string(
"%s.ss" % os.path.join(self.pymod.psipred_dirpath, basename)))
try:
psipass2_output = open(
"%s.horiz" %
os.path.join(self.pymod.psipred_dirpath, basename), "w")
self.pymod.execute_subprocess(psipass2_command,
new_stdout=psipass2_output)
psipass2_output.close()
except:
if print_output:
print(
"- FATAL: Error whilst running psipass 2 - script terminated!"
)
raise Exception(
"There was an error while running PSIPRED and no prediction was made for %s."
% (sequence_header))
#--------------------------
# Clean up PSIPRED files. -
#--------------------------
if print_output:
print("- Cleaning up ...")
# Remove temporary files.
self.remove_psipred_temp_files()
# Renames the output files.
output_files_name = pmos.clean_file_name(element.my_header)
for ext in psipred_output_extensions:
os.rename(
os.path.join(self.pymod.psipred_dirpath, basename + ext),
os.path.join(self.pymod.psipred_dirpath,
output_files_name + ext))
if print_output:
print("- Final output files:" + output_files_name + ".ss2 " +
output_files_name + ".horiz")
print("- Finished.")
#----------------------------------------------
# Parses the results from .horiz output file. -
#----------------------------------------------
results_file = open(
os.path.join(self.pymod.psipred_dirpath,
output_files_name + ".horiz"), "r")
confs = "" # String for confidence scores of each residue.
preds = "" # String for the secondary structure elements prediction of each residue.
for l in results_file.readlines():
if l.startswith("Conf:"):
rl = l[6:66].replace(" ", "").replace("\n", "")
confs += rl
elif l.startswith("Pred:"):
rl = l[6:66].replace(" ", "").replace("\n", "")
preds += rl
results_file.close()
# Actually stores in the PyMod elements the results.
element.psipred_elements_list = []
for c, e, res in zip(confs, preds, element.get_polymer_residues()):
res.psipred_result = {"confidence": int(c), "sec-str-element": e}
return True
def build_psiblast_commandline(self,
ncbi_dir,
db_path,
query,
inclusion_ethresh=0.001,
num_iterations=3,
evalue=None,
max_target_seqs=None,
num_alignments=None,
out=None,
outfmt=None,
out_pssm=None,
blast_version="psiblast"):
self.check_blast_version(blast_version)
# blastdbcmd -db "\"Users\joeuser\My Documents\Downloads\mydb\"" -info
# blastdbcmd -db ' "path with spaces/mydb" ' -info
psiblast_path = pmos.build_commandline_path_string(
os.path.join(ncbi_dir, pmos.get_exe_file_name(blast_version)))
db_path = pmos.build_commandline_file_argument("db", db_path)
query = pmos.build_commandline_file_argument("query", query)
if inclusion_ethresh != None:
inclusion_ethresh = " -inclusion_ethresh %s" % (inclusion_ethresh)
else:
inclusion_ethresh = ""
if num_iterations != None:
num_iterations = " -num_iterations %s" % (num_iterations)
else:
num_iterations = ""
if evalue != None:
evalue = " -evalue %s" % (evalue)
else:
evalue = ""
if outfmt != None:
outfmt = " -outfmt %s" % (outfmt
) # 5 produces an .xml output file.
else:
outfmt = ""
if out != None:
out = pmos.build_commandline_file_argument("out", out)
else:
out = ""
if max_target_seqs != None:
max_target_seqs = " -max_target_seqs %s" % (max_target_seqs)
else:
max_target_seqs = ""
if out_pssm != None:
out_pssm = pmos.build_commandline_file_argument(
"out_pssm", out_pssm)
else:
out_pssm = ""
if num_alignments != None:
num_alignments = " -num_alignments %s" % (num_alignments)
else:
num_alignments = ""
psiblast_command = (psiblast_path + db_path + query +
inclusion_ethresh + out + outfmt + out_pssm +
num_iterations + evalue + max_target_seqs +
num_alignments)
return psiblast_command
def check_hmmscan_program(self):
if self.father_protocol.domain_search_mode == "local":
exe_dirpath = self.pymod.hmmer_tool["exe_dir_path"].get_value()
if not (exe_dirpath and os.path.isdir(exe_dirpath)):
title = "Hmmer Suite Executable Error"
message = (
"The default Hmmer suite executables directory is missing. Please set one"
" in the 'Tools -> Options' menu.")
self.pymod.main_window.show_error_message(title, message)
return False
exe_filename = get_exe_file_name("hmmscan")
exe_filepath = os.path.join(exe_dirpath, exe_filename)
if not os.path.isfile(exe_filepath):
title = "Hmmer Suite Executable Error"
message = (
"A '%s' file is missing in the Hmmer executables directory. Please specify"
" in the 'Tools -> Options' menu a Hmmer executables directory where a"
" '%s' file is found." % (exe_filename, exe_filename))
self.pymod.main_window.show_error_message(title, message)
return False
db_dirpath = self.pymod.hmmer_tool[
"hmmscan_db_dir_path"].get_value()
if db_dirpath.replace(" ", "") == "":
title = "No Database Directory"
message = (
"No databases directory is defined for HMMSCAN. Please define a database"
" directory in the PyMod options window in order to perform a HMMSCAN search."
)
self.pymod.main_window.show_error_message(title, message)
return False
if not os.path.isdir(db_dirpath):
title = "Database Directory not Found"
message = (
"The specified database directory does not exists. Please specify an"
" existing one in the Tools -> Options menu.")
self.pymod.main_window.show_error_message(title, message)
return False
self.hmmscan_db_list = [
d for d in sorted(os.listdir(db_dirpath))
if d.endswith("hmm.h3m")
]
if len(self.hmmscan_db_list) == 0:
title = "Hmmscan Database Error"
message = (
"No valid databases files were found in the hmmscan database directory."
" Please refer to the PyMod manual to know how to install them."
)
self.pymod.main_window.show_error_message(title, message)
return False
return True
elif self.father_protocol.domain_search_mode == "remote":
if not check_network_connection("https://google.com", timeout=3):
title = "Connection Error"
message = (
"An internet connection is not available, can not connect to the EBI"
" server to run HMMSCAN.")
self.pymod.main_window.show_error_message(title, message)
return False
return True