def run_agadir_on_multifastas(self, path_fastafile: str, path_dst: str):
"""
To run agadir on file that has multiple fasta sequences. Iterate through the list, write the sequence to an
individual text file passing this to agadir.compute(). This newly-written individual text file is then deleted,
crucial to prevent too much memory taken up by each file when running large numbers of sequences.
:param path_fastafile: Abs path to fasta file.
:param path_dst: Abs path to output root dir.
"""
with open(path_fastafile) as f:
is_first_line = True
fasta_str = ''
mutantfastafilename = ''
mutantfastafile = ''
path_dst_mutant_filename = ''
for line in f.readlines():
if '>' in line:
if not is_first_line:
path_dst_mutant_filename = GUM.os_makedirs(
path_dst, mutantfastafilename)
path_dst_mutant_file = os.path.join(
path_dst_mutant_filename, mutantfastafile)
with open(path_dst_mutant_file, 'w') as g:
g.write(fasta_str)
agadir = Agadir(Cond.INCELL_MAML.value)
agadir.compute(path_dst_mutant_file)
GUM.linux_remove_file(path_dst_mutant_file)
fasta_str = line
is_first_line = False
mutantfastafilename = line.split('>')[-1].split('\n')[0]
mutantfastafile = mutantfastafilename + Str.FSTAEXT.value
else:
fasta_str += line
res = regex.search(filename)
regex2 = re.compile(r"RepairPDB_[0-9]+_[A-Y][A-Y][0-9]+[A-Z]")
res2 = regex2.search(filename)
fxmutname = filename[res.end():res2.end()]
return fxmutname
"""
3. Read csv files and write to a single csvfile for mysql dump.
"""
for path_output_csvfile in path_output_csvfiles:
csvfilename = os.path.basename(path_output_csvfile)
pdbname = _extract_pdbname(csvfilename)
fxmutantname = _extract_fxmutantname(csvfilename)
if using_cluster:
path_jobq_dir = GUM.os_makedirs(Paths.CONFIG_BM_JOBQ, pdbname,
fxmutantname)
Cluster.write_job_q_bash(
jobname=Paths.PREFIX_FX_RM.value + fxmutantname,
path_job_q_dir=path_jobq_dir,
python_script_with_paths=os.path.join(
Paths.SE_SRC_CLSTR_PYSCRPTS.value,
'write_1csvfile_from_csvPerMutantfiles_zeus.py' +
Str.SPCE.value + path_output_csvfile + Str.SPCE.value +
path_output_ac_or_bm_dir + Str.SPCE.value + pdbname))
Cluster.run_job_q(path_job_q_dir=path_jobq_dir)
else:
GUM.write_1_csvfile_from_csv_per_mutants(path_output_csvfile,
path_output_ac_or_bm_dir,
pdbname)
# pydevd.stoptrace()
path_output = sys.argv[2]
path_config_job = sys.argv[3]
path_output_blastp = sys.argv[4]
write_idmaps_for_mysldb = sys.argv[5] == 'True'
write_csv = sys.argv[6] == 'True'
write_xml = sys.argv[7] == 'True'
write_json = sys.argv[8] == 'True'
for path_fastafile in path_input_fastafile_list:
with open(path_fastafile) as fastafile_opened:
fastafile_name = path_fastafile.split('/')[-1].split('.')[0]
jobname = 'BLSTP_' + fastafile_name
Cluster.write_job_q_bash(jobname=jobname, path_job_q_dir=path_config_job, queue='all.q', memory_limit_GB='3',
cluster_node='hodor1.vib')
path_output_blastp_fastaname = GUM.os_makedirs(path_output_blastp, fastafile_name)
os.chdir(path_output_blastp_fastaname)
Cluster.run_job_q(path_job_q_dir=path_config_job)
Cluster.wait_for_grid_engine_job_to_complete(grid_engine_job_prefix_or_full_name=jobname)
path_raw_blstp_xml = IdProt._write_raw_blast_xml(path_output, fastafile_name,
blastp_result=NCBIWWW.qblast(
program=Biopy.BlastParam.BLST_P.value,
database=Biopy.BlastParam.SWSPRT.value,
sequence=fastafile_opened.read(),
entrez_query=Biopy.BlastParam.HOMSAP_ORG.value,
alignments=Biopy.BlastParam.MAX_ALIGN_20.value,
hitlist_size=Biopy.BlastParam.MAX_HIT_20.value))
blastp_dict = Biopy.parse_filter_blastp_xml_to_dict(path_raw_blstp_xml, fastafile_name, path_fastafile)
# blastp_dict_list.append(blastp_dict)
if write_idmaps_for_mysldb:
IdProt._write_idmaps_for_mysqldb(path_output, blastp_dict, write_csv=write_csv, write_xml=write_xml,
def _write_mutants(self,
title_titleSeq_w_mutants: dict,
write_1_fasta_only: bool,
write_fasta_per_mut: bool,
path_output_3dots: str,
write_csv=False,
write_txt=False):
"""
Writes the mutants out to fastafiles and/or csv files and/or txt files. These can be written in one file
containing all mutants or one file per mutant.
The fastafiles are written to /output_data/<fastafilename>/mutants/.
The csvfiles and txtfiles are written to /output_data/<fastafilename/sequences/.
The reason for fastafile mutants being written to an input folder is that these sequences are generated as
direct inputs for the mutation operation.
:param title_titleSeq_w_mutants: Title of wild-type associated to every mutant title:sequence.
:param write_1_fasta_only: True to write one fastafile containing all mutants, separated by \n.
:param write_fasta_per_mut: True to write one fastafile per mutant.
:param path_output_3dots: Absolute path of output_data dir (where fasta, txt, csv written), includes to
subdirs: /fastas/xxxx...yyyy/ e.g. 1001...2000
:param write_csv: True to write 1 csv file for wt & mutants. False by default.
:param write_txt: True to write 1 txt file for wt & mutants. False by default.
"""
path_1_fastafile = None
path_1_fastafile_open = None
path_fastafilepermut = None
path_seqscsv = None
path_seqscsv_open = None
path_seqstxt = None
path_seqstxt_open = None
for wt_title, title_seq in title_titleSeq_w_mutants.items():
for mut_title, mut_seq in title_seq.items():
if write_1_fasta_only and path_1_fastafile is None:
path_1_fastafile = GUM.os_makedirs(path_output_3dots,
wt_title,
Paths.DIR_MUTANTS.value)
path_1_fastafile = os.path.join(
path_1_fastafile, wt_title + '_mutants.fasta')
path_1_fastafile_open = open(path_1_fastafile, 'w')
path_1_fastafile_open.write('>' + wt_title + '\n' +
title_seq[wt_title] + '\n')
if write_fasta_per_mut and path_fastafilepermut is None:
path_fastafilepermut = GUM.os_makedirs(
path_output_3dots, Paths.DIR_FASTAS.value, wt_title)
path_fastafilepermut = os.path.join(
path_fastafilepermut, wt_title + Str.FSTAEXT.value)
path_fastafilepermut_open = open(path_fastafilepermut, 'w')
path_fastafilepermut_open.write('>' + wt_title + '\n' +
title_seq[wt_title] + '\n')
path_fastafilepermut_open.close()
if write_csv and path_seqscsv is None:
path_seqscsv = GUM.os_makedirs(
path_output_3dots, Paths.DIR_SEQS_TXT_CSV.value,
wt_title)
path_seqscsv = os.path.join(path_seqscsv,
wt_title + '_mutants.csv')
path_seqscsv_open = open(path_seqscsv, 'w')
path_seqscsv_open.write(wt_title + ':' +
title_seq[wt_title] + ',')
if write_txt and path_seqstxt is None:
path_seqstxt = GUM.os_makedirs(
path_output_3dots, Paths.DIR_SEQS_TXT_CSV.value,
wt_title)
path_seqstxt = os.path.join(path_seqstxt,
wt_title + '_mutants.txt')
path_seqstxt_open = open(path_seqstxt, 'w')
path_seqstxt_open.write(wt_title + ':' +
title_seq[wt_title] + '\n')
elif mut_title is not wt_title:
if write_1_fasta_only and path_1_fastafile_open is not None:
path_1_fastafile_open.write('>' + mut_title + '\n' +
mut_seq + '\n')
if write_fasta_per_mut and path_fastafilepermut is not None:
path_fastafilepermut = os.path.join(
path_output_3dots, Paths.DIR_FASTAS.value,
wt_title, Paths.DIR_MUTANTS.value,
mut_title + Str.FSTAEXT.value)
path_fastafilepermut_open = open(
path_fastafilepermut, 'w')
path_fastafilepermut_open.write('>' + mut_title +
'\n' + mut_seq + '\n')
path_fastafilepermut_open.close()
if write_csv and path_seqscsv_open is not None:
path_seqscsv_open.write(mut_title + ':' + mut_seq +
',')
if write_txt and path_seqstxt_open is not None:
path_seqstxt_open.write(mut_title + ':' + mut_seq +
'\n')
if path_1_fastafile_open is not None:
path_1_fastafile_open.close()
if path_seqstxt_open is not None:
path_seqstxt_open.close()
if path_seqscsv_open is not None:
path_seqscsv_open.close()
:sys.argv[1] path_fastafile: Abs path to fasta file.
:sys.argv[2] path_dst: Abs path to output root dir.
"""
print('run_agadir_on_multifastas_zeus.py ###################################')
path_dst = GUM.make_path_agadir_3dots_filename_mutants_dirs(
path_dst, path_fastafile, add_filename_subdir=True)
with open(path_fastafile) as f:
is_first_line = True
fasta_str = ''
mutantfastafilename = ''
mutantfastafile = ''
for line in f.readlines():
if '>' in line:
if not is_first_line:
path_dst_mutant_filename = GUM.os_makedirs(
path_dst, mutantfastafilename)
path_dst_mutant_file = os.path.join(path_dst_mutant_filename,
mutantfastafile)
with open(path_dst_mutant_file, 'w') as g:
g.write(fasta_str)
agadir = Agadir(Cond.INCELL_MAML.value)
agadir.compute(path_dst_mutant_file)
GUM.linux_remove_file(path_dst_mutant_file)
fasta_str = line
is_first_line = False
mutantfastafilename = line.split('>')[-1].split('\n')[0]
mutantfastafile = mutantfastafilename + Str.FSTAEXT.value
else:
fasta_str += line