def plasmidID_call(plasmidID_bin, file1, file2, plasmid_contigs, database,
threads, outfolder):
name_split = os.path.basename(file1).split("_trim_R1.fastq")
name = name_split[0]
cmd = "%s --R1 %s --R2 %s --database %s --sample %s --no-trim --contigs %s --threads %s --group %s" % (
plasmidID_bin, file1, file2, database, name, plasmid_contigs, threads,
outfolder)
functions.system_call(cmd)
def run_doMLST(profile_folder, seq_folder, name, rscript, path, fileGiven,
threads):
print('+ Generating profile for sample...')
folder_results = os.path.join(path, name + '_alleles')
## success timestamp
filename_stamp = path + '/.success'
if os.path.isfile(filename_stamp):
stamp = functions.read_time_stamp(filename_stamp)
print(
colored(
"\tA previous command generated results on: %s [%s]" %
(stamp, name), 'yellow'))
res_file = path + '/' + name + "_MLST_results.csv"
return (res_file)
else:
if os.path.exists(folder_results):
shutil.rmtree(folder_results)
logFile = path + '_logFile.txt'
cmd_profiler = "%s %s --dir_profile %s --dir_seq %s --file %s --dir %s --name %s --threads %s --lib.loc %s 2> %s" % (
rscript, MLSTarR_script, profile_folder, seq_folder, fileGiven,
path, name, threads, get_MLSTar_package_installed(), logFile)
callCode = functions.system_call(cmd_profiler)
if callCode == 'OK':
res_file = path + '/' + name + "_MLST_results.csv"
stamp = functions.print_time_stamp(filename_stamp)
return (res_file)
else:
return ('FAIL')
def getPUBMLST(species, rscript, out_name):
"""
Using `MLSTar software`_ retrieve for the given `species` the available schemes in PubMLST_.
It generates information in file `out_name` in csv format.
See example in :file:`/devel/results/getPubMLST_example.csv`
.. include:: ../../devel/results/getPubMLST_example.csv
:literal:
:param species: Sample name or tag to identify sample
:param rscript: Path to Rscript to generate system call.
:param out_name: Output file to generate profile information.
:type species: string
:type rscript: string
:type out_name: string
:return: OK/FAIL
:warnings: Returns **FAIL** if process stopped.
.. include:: ../../links.inc
"""
MLSTarR_getpubmlst = tools.R_scripts('MLSTar_getpubmlst')
## species is a comma separated string
path_package = get_MLSTar_package_installed()
#MLSTar_getpubmlst
cmd_getPUBMLST = "%s %s --species %s --output %s --lib.loc %s 2> /dev/null" % (
rscript, MLSTarR_getpubmlst, species, out_name, path_package)
return (functions.system_call(cmd_getPUBMLST))
def get_MLSTar_package_installed(debug=False):
install_path = set_config.R_package_path_installed()
(check_install_system, check_install_path) = set_config.get_check_R_files()
R_script_exe = set_config.get_exe('Rscript')
if debug:
print('\n+ Check package: MLSTar')
## ATTENTION: optparse library missing, no installation withn conda
## first try to check if package available in system
cmd_check = R_script_exe + ' ' + check_install_system + ' -l MLSTar'
code = functions.system_call(cmd_check, message=False, returned=False)
if (code == 'OK'):
return ('system')
else:
## check if installed in path
cmd_check_path = R_script_exe + ' ' + check_install_path + ' -l MLSTar -p ' + install_path
code2 = functions.system_call(cmd_check_path,
message=False,
returned=False)
if (code2 == 'OK'):
return (install_path)
else:
(install_R, install_github_package
) = install_dependencies.get_install_R_files()
cmd_R = '%s %s -l iferres/MLSTar -p %s' % (
R_script_exe, install_github_package, install_path)
code3 = functions.system_call(cmd_R, message=False, returned=False)
if (code3):
return (install_path)
else:
print('ERROR')
exit()
def filter_and_cluster_database(database_path, threads):
original_file = database_path + '/plasmid_database.fna'
clustering_file = database_path + '/plasmids_clustered.fna'
if os.path.isfile(clustering_file):
print("+ Plasmid database was previously clustered...")
return (clustering_file)
else:
print('+ Cluster putative duplicated entries on database...')
folder = functions.create_subfolder('blast_search', database_path)
## makeblastDB
dbName = folder + '/plasmids_DB'
functions.makeblastdb(dbName, original_file)
## split plasmid seqs ids
tmp_folder = functions.create_subfolder("tmp", folder)
cmd_split = 'perl %s %s %s %s ' % (fasta_split_script, original_file,
threads, tmp_folder)
print("\n+ Splitting database file to speed up the computation...")
functions.system_call(cmd_split)
## get files in folder
split_files = os.listdir(tmp_folder)
num_threads = 1
print("\n+ Sending blastn commands: ")
# We can use a with statement to ensure threads are cleaned up promptly
with concurrent.futures.ThreadPoolExecutor(
max_workers=threads) as executor:
# Start the load operations and mark each future with its URL
commandsSent = {
executor.submit(functions.blastn,
tmp_folder + '/' + fil + '-blast-out.txt', dbName,
tmp_folder + '/' + fil, num_threads): fil
for fil in split_files
}
for cmd2 in concurrent.futures.as_completed(commandsSent):
details = commandsSent[cmd2]
try:
data = cmd2.result()
except Exception as exc:
print('***ERROR:')
print('%r generated an exception: %s' % (details, exc))
########################
## parseBlast results
########################
print("+ Concatenating all results into one file...")
outFile_merged = folder + '/blastn_output_concat.txt'
cmd_cat = 'cat ' + tmp_folder + '/*-blast-out.txt > ' + outFile_merged
functions.system_call(cmd_cat)
## thresholds
eval_thresh_float = float(1e-20)
aln_thresh_given = 90
min_length = 2000
outFile_parsed = folder + '/blastn_output_parsed.txt'
output_file = open(outFile_parsed, 'w')
print('+ Parsing BLAST results generated...\n')
my_cluster_list = []
sequences2discard = []
## get number lines
lines = functions.get_number_lines(outFile_merged)
n = 0
## get results
fh = open(outFile_merged)
for blast_record in HCGB_blast.parse(fh,
eval_thresh=eval_thresh_float,
aln_thresh=aln_thresh_given,
length_thresh=min_length):
### show progress bar
n += 10
functions.progbar(n, lines, 100)
for hit in blast_record.hits:
for hsp in hit:
output_file.write('****Alignment****\n')
output_file.write('query id: {}\n'.format(blast_record.qid))
sequences2discard.append(hsp.sid) ## add subject
output_file.write('sequence: %s\n' % hsp.sid)
output_file.write('e value: %s\n' % hsp.evalue)
output_file.write('aln_length: %s\n' % hsp.length)
output_file.write('qlen: %s [>%s]\n' % (hsp.qlen, min_length))
aln_perc = (int(hsp.length) / int(hsp.qlen)) * 100
output_file.write('aln_perc: %s [> %s]\n\n' %
(aln_perc, aln_thresh_given))
#print('query id: {}'.format(blast_record.qid))
#print('sequence: ', hsp.sid)
#print('e value:', hsp.evalue)
#print('aln_length:', hsp.length)
#print('qlen:', hsp.qlen, ' [>', min_length, ' bp]')
#print ('aln_perc:', aln_perc, ' [>', aln_thresh_given ,'%]')
## add query:
qseq = format(blast_record.qid)
if (qseq in sequences2discard):
continue ## avoid adding query if already reported as subject
else:
my_cluster_list.append(qseq)
print("\n\n")
fh.close()
output_file.close()
## get unique ids & print in file
print("+ Obtaining clustered sequences...")
my_cluster_list = set(my_cluster_list)
list_cluster_file = folder + '/plasmids_clustered_ids.txt'
functions.printList2file(list_cluster_file, my_cluster_list)
## retrieve ids
cluster_fasta_file = folder + '/plasmids_clustered.fna'
cmd = 'perl %s %s %s > %s' % (retrieve_seqs_script, list_cluster_file,
original_file, cluster_fasta_file)
functions.system_call(cmd)
final_cluster_fasta_file = database_path + '/plasmids_clustered.fna'
shutil.copy(cluster_fasta_file, final_cluster_fasta_file)
## maybe add a step to build bowtie index from database to speed up later process
return (final_cluster_fasta_file)
def download_PubMLST(profile_folder, scheme, seq_folder, rscript, species):
## Check if profile exists
file_prof = profile_folder + '/profile_scheme' + str(scheme) + '.tab'
######################
## download profile ##
######################
if os.path.exists(file_prof):
print('+ Profile file for scheme exists...')
## check if previously download profile and succeeded
filename_stamp = profile_folder + '/.success'
if os.path.isfile(filename_stamp):
stamp = functions.read_time_stamp(filename_stamp)
print(
colored(
"\tA previous command generated results on: %s [%s -- %s]"
% (stamp, species, 'profile'), 'yellow'))
#
#else
# [TODO: Check time passed and download again if >?? days passed]
else:
print('+ Profile file for scheme will be downloaded...')
print('+ Downloading profile...')
logFile = profile_folder + '_download.log'
cmd_profile = "%s %s --species %s --scheme %s --dir_profile %s --lib.loc %s 2> %s" % (
rscript, MLSTarR_download_prf, species, scheme, profile_folder,
get_MLSTar_package_installed(), logFile)
callCode = functions.system_call(cmd_profile)
if (callCode == 'OK'):
## success timestamp
filename_stamp = profile_folder + '/.success'
stamp = functions.print_time_stamp(filename_stamp)
#######################
## download sequence ##
#######################
seq_bool = 0
if os.path.exists(seq_folder):
print('+ Sequence folder exists...')
## check if previously download sequence and succeeded
filename_stamp = seq_folder + '/.success'
if os.path.isfile(filename_stamp):
stamp = functions.read_time_stamp(filename_stamp)
print(
colored(
"\tA previous command generated results on: %s [%s -- %s]"
% (stamp, species, 'sequence'), 'yellow'))
########################################################################
#else
# [TODO: Check time passed and download again if >?? days passed]
########################################################################
#files = os.listdir(seq_folder)
#count_fas = 0
#for f in files:
# if f.endswith('.fas'):
# count_fas += 1
# else:
# os.remove(seq_folder + '/' + f)
#
#if count_fas > 6:
# print ("+ Assuming sequences are previously downloaded...")
#else:
# seq_bool = 1
# os.rmdir(seq_folder)
# seq_folder_path = functions.create_folder(seq_folder)
else:
print('+ Sequence files for scheme will be downloaded...')
print('+ Downloading sequences...')
logFile = seq_folder + '_download.log'
cmd_seq = "%s %s --species %s --scheme %s --dir_seq %s --lib.loc %s 2> %s" % (
rscript, MLSTarR_download_seq, species, scheme, seq_folder,
get_MLSTar_package_installed(), logFile)
callCode = functions.system_call(cmd_seq)
if (callCode == 'OK'):
## success timestamp
filename_stamp = seq_folder + '/.success'
stamp = functions.print_time_stamp(filename_stamp)
def plot_MLST(results, profile, rscript):
path_package = get_MLSTar_package_installed()
path_folder = os.path.dirname(results)
cmd_plotter = "%s %s --output %s --folder_profile %s --file_result %s --lib.loc %s" % (
rscript, MLSTarR_plot, path_folder, profile, results, path_package)
return (functions.system_call(cmd_plotter))
def GI_module(genbank_file, name, outdir, Debug, cutoff_dinuc_bias=8, min_length=1000):
"""Identify genomic islands (GI) within the genbank file provided. They are calculated
based on gene annotation and dinucleotide bias region using the software `IslandPath-DIMOB`_.
:param genbank_file: Absolute path to annotation file in Genbank format.
:param name: Sample identifier.
:param outdir: Absolute path to output folder.
:param cutoff_dinuc_bias: Dinucleotide bias cutoff
:param min_length: Minimun length for the regions to be reported
:type name: string
:type genbank_file: string
:type outdir: string
:type cutoff_dinuc_bias: int
:type min_length: int
The Dimob.pl perl script has two mandatory argument which are the input :file:`genbank_file` and an output name.
.. code-block:: sh
Usage:
perl Dimob.pl <genome.gbk> <output_name> [cutoff_dinuc_bias] [min_length]
Default values:
cutoff_dinuc_bias = 8
min_length = 8000
Example:
perl Dimob.pl example/NC_003210.gbk NC_003210_GIs
perl Dimob.pl example/NC_003210.gbk NC_003210_GIs 6 10000
perl Dimob.pl example/NC_000913.embl NC_000913_GIs 6 10000
During the development of BacterialTyper, we generated a modification of the original `IslandPath-DIMOB`_ to analyze
contig sequence data and generated different output format for better clarificaiton and interpretaion of results.
We forked the original code into a new git repository and update the code accordingly. See details here: https://github.com/JFsanchezherrero/islandpath.
.. include:: ../../links.inc
"""
## filename stamp of the process
filename_stamp = outdir + '/.Dimob'
# check if previously done
if os.path.isfile(filename_stamp):
stamp = functions.read_time_stamp(filename_stamp)
print (colored("\tA previous command generated results on: %s [%s -- Dimob]" %(stamp, name), 'yellow'))
else:
## debug message
if (Debug):
print (colored("**DEBUG: Call Dimob for sample %s " %name + "**", 'yellow'))
print ("genbank_file", genbank_file)
print ("outdir: ", outdir)
## Call IslandPath Dimob executable perl file.
dimob_pl = set_config.get_exe("dimob", Debug)
perl_exe = set_config.get_exe("perl", Debug)
## command
outdir_sample = os.path.join(outdir, name)
log_file = outdir_sample + '.log'
perl_cmd = '%s %s %s %s %s %s > %s' %(perl_exe, dimob_pl, genbank_file, outdir_sample, cutoff_dinuc_bias, min_length, log_file)
code = functions.system_call(perl_cmd)
##
if code:
## when finished print time stamp in output + '/.Dimob'
stamp = functions.print_time_stamp(filename_stamp)
else:
return False
return (outdir)