def main():
parser = argparse.ArgumentParser(prog='patho_typing.py',
description='In silico pathogenic typing directly from raw Illumina reads',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('--version', help='Version information', action='version',
version='{prog} v{version}'.format(prog=parser.prog, version=__version__))
parser_required = parser.add_argument_group('Required options')
parser_required.add_argument('-f', '--fastq', nargs='+', action=utils.required_length((1, 2), '--fastq'),
type=argparse.FileType('r'), metavar=('/path/to/input/file.fq.gz'),
help='Path to single OR paired-end fastq files. If two files are passed, they will be'
' assumed as being the paired fastq files', required=True)
parser_required.add_argument('-s', '--species', nargs=2, type=str, metavar=('Yersinia', 'enterocolitica'),
help='Species name', required=True)
parser_optional_general = parser.add_argument_group('General facultative options')
parser_optional_general.add_argument('-o', '--outdir', type=str, metavar='/path/to/output/directory/',
help='Path to the directory where the information will be stored',
required=False, default='.')
parser_optional_general.add_argument('-j', '--threads', type=int, metavar='N', help='Number of threads to use',
required=False, default=1)
parser_optional_general.add_argument('--trueCoverage', action='store_true',
help='Assess true coverage before continue typing')
parser_optional_general.add_argument('--noCheckPoint', action='store_true',
help='Ignore the true coverage checking point')
parser_optional_general.add_argument('--minGeneCoverage', type=int, metavar='N',
help='Minimum typing percentage of target reference gene sequence covered to'
' consider a gene to be present (value between [0, 100])', required=False)
parser_optional_general.add_argument('--minGeneIdentity', type=int, metavar='N',
help='Minimum typing percentage of identity of reference gene sequence covered'
' to consider a gene to be present (value between [0, 100]). One INDEL'
' will be considered as one difference', required=False)
parser_optional_general.add_argument('--minGeneDepth', type=int, metavar='N',
help='Minimum typing gene average coverage depth of present positions to'
' consider a gene to be present (default is 1/3 of average sample'
' coverage or 15x)', required=False)
parser_optional_general.add_argument('--doNotRemoveConsensus', action='store_true',
help='Do not remove ReMatCh consensus sequences')
parser_optional_general.add_argument('--debug', action='store_true',
help='DeBug Mode: do not remove temporary files')
args = parser.parse_args()
if args.minGeneCoverage is not None and (args.minGeneCoverage < 0 or args.minGeneCoverage > 100):
parser.error('--minGeneCoverage should be a value between [0, 100]')
if args.minGeneIdentity is not None and (args.minGeneIdentity < 0 or args.minGeneIdentity > 100):
parser.error('--minGeneIdentity should be a value between [0, 100]')
start_time = time.time()
args.outdir = os.path.abspath(args.outdir)
if not os.path.isdir(args.outdir):
os.makedirs(args.outdir)
# Start logger
logfile, time_str = utils.start_logger(args.outdir)
script_path = utils.general_information(logfile, __version__, args.outdir, time_str)
print('\n')
rematch = include_rematch_dependencies_path()
args.fastq = [fastq.name for fastq in args.fastq]
reference_file, trueCoverage_file, trueCoverage_sequences, trueCoverage_headers, trueCoverage_config, typing_file, \
typing_sequences, typing_headers, typing_rules, typing_config = \
set_reference(args.species, args.outdir, script_path, args.trueCoverage)
original_reference_file = str(reference_file)
confirm_genes_fasta_rules(typing_headers, typing_rules)
run_successfully, bam_file = mapping_reads(args.fastq, reference_file, args.threads, args.outdir, False, 1)
if run_successfully:
rematch_dir = os.path.join(args.outdir, 'rematch', '')
if not os.path.isdir(rematch_dir):
os.makedirs(rematch_dir)
if args.trueCoverage:
if trueCoverage_file is not None:
trueCoverage_dir = os.path.join(rematch_dir, 'trueCoverage', '')
if not os.path.isdir(trueCoverage_dir):
os.makedirs(trueCoverage_dir)
print('\n')
run_successfully, trueCoverage_bam = split_bam(bam_file, trueCoverage_headers, trueCoverage_dir,
args.threads)
if run_successfully:
run_successfully = indexAlignment(trueCoverage_bam)
if run_successfully:
reference_file = os.path.join(trueCoverage_dir, 'reference.fasta')
write_sequeces(reference_file, trueCoverage_sequences)
index_fasta_samtools(reference_file, None, None, True)
config = parse_config(trueCoverage_config)
runtime, run_successfully, sample_data_general, data_by_gene = \
run_rematch.run_rematch(rematch, trueCoverage_dir, reference_file, trueCoverage_bam,
args.threads, config['length_extra_seq'],
config['minimum_depth_presence'], config['minimum_depth_call'],
config['minimum_depth_frequency_dominant_allele'],
config['minimum_gene_coverage'], config['minimum_gene_identity'],
args.debug, args.doNotRemoveConsensus)
if run_successfully and sample_data_general['mean_sample_coverage'] is not None and \
sample_data_general['number_absent_genes'] is not None and \
sample_data_general['number_genes_multiple_alleles'] is not None:
if args.minGeneDepth is None:
args.minGeneDepth = sample_data_general['mean_sample_coverage'] / 3 if \
sample_data_general['mean_sample_coverage'] / 3 > 15 else \
15
exit_info = []
if sample_data_general['mean_sample_coverage'] < config['minimum_read_coverage']:
exit_info.append('Sample coverage ({mean}) lower than the minimum'
' required ({minimum})'
''.format(mean=sample_data_general['mean_sample_coverage'],
minimum=config['minimum_read_coverage']))
if sample_data_general['number_absent_genes'] > config['maximum_number_absent_genes']:
exit_info.append('Number of absent genes ({number}) higher than the'
' maximum allowed ({maximum})'
''.format(number=sample_data_general['number_absent_genes'],
maximum=config['maximum_number_absent_genes']))
if sample_data_general['number_genes_multiple_alleles'] > \
config['maximum_number_genes_multiple_alleles']:
exit_info.append('Number of genes with multiple alleles'
' ({number}) higher than the maximum'
' allowed ({maximum})'
''.format(number=sample_data_general['number_genes_multiple_alleles'],
maximum=config['maximum_number_genes_multiple_alleles']))
if len(exit_info) > 0:
print('\n' + '\n'.join(exit_info) + '\n')
e = 'TrueCoverage requirements not fulfilled'
print('\n' + e + '\n')
if not args.noCheckPoint:
clean_pathotyping_folder(args.outdir, original_reference_file, args.debug)
_ = utils.runTime(start_time)
sys.exit(e)
else:
e = 'TrueCoverage module did not run successfully'
print('\n' + e + '\n')
if not args.noCheckPoint:
clean_pathotyping_folder(args.outdir, original_reference_file, args.debug)
_ = utils.runTime(start_time)
sys.exit(e)
print('\n')
typing_dir = os.path.join(rematch_dir, 'typing', '')
if not os.path.isdir(typing_dir):
os.makedirs(typing_dir)
run_successfully, bam_file = split_bam(bam_file, typing_headers, typing_dir, args.threads)
if run_successfully:
run_successfully = indexAlignment(bam_file)
if run_successfully:
reference_file = os.path.join(typing_dir, 'reference.fasta')
write_sequeces(reference_file, typing_sequences)
index_fasta_samtools(reference_file, None, None, True)
rematch_dir = str(typing_dir)
if not run_successfully:
if args.noCheckPoint:
clean_pathotyping_folder(args.outdir, original_reference_file, args.debug)
_ = utils.runTime(start_time)
sys.exit('Something in the required TrueCoverage analysis went wrong')
else:
print('\n'
'WARNING: it was not found trueCoverage target files. trueCoverage will not run.'
'\n')
if run_successfully:
config = parse_config(typing_config)
if args.minGeneCoverage is not None:
config['minimum_gene_coverage'] = args.minGeneCoverage
if args.minGeneIdentity is not None:
config['minimum_gene_identity'] = args.minGeneIdentity
runtime, run_successfully, sample_data_general, data_by_gene = \
run_rematch.run_rematch(rematch, rematch_dir, reference_file, bam_file, args.threads,
config['length_extra_seq'], config['minimum_depth_presence'],
config['minimum_depth_call'], config['minimum_depth_frequency_dominant_allele'],
config['minimum_gene_coverage'], config['minimum_gene_identity'],
args.debug, args.doNotRemoveConsensus)
if run_successfully and data_by_gene is not None:
if args.minGeneDepth is None:
args.minGeneDepth = sample_data_general['mean_sample_coverage'] / 3 if \
sample_data_general['mean_sample_coverage'] / 3 > 15 else \
15
_, _, _ = typing.typing(data_by_gene, typing_rules, config['minimum_gene_coverage'],
config['minimum_gene_identity'], args.minGeneDepth, args.outdir)
else:
clean_pathotyping_folder(args.outdir, original_reference_file, args.debug)
_ = utils.runTime(start_time)
sys.exit('ReMatCh run for pathotyping did not run successfully')
else:
clean_pathotyping_folder(args.outdir, original_reference_file, args.debug)
_ = utils.runTime(start_time)
sys.exit('Something did not run successfully')
clean_pathotyping_folder(args.outdir, original_reference_file, args.debug)
print('\n')
_ = utils.runTime(start_time)
def python_arguments(program_name, version):
"""
Sets pythons arguments
Parameters
----------
program_name : str
String with the name of the program
version : str
String with version
Returns
-------
parser : argparse.ArgumentParser
General argparse
parser_reads : argparse.ArgumentParser.add_subparsers.add_parser
Reads subparser
For running the program using fastq files
parser_index : argparse.ArgumentParser.add_subparsers.add_parser
Index subparser
For creating Bowtie2 index
parser_assembly : argparse.ArgumentParser.add_subparsers.add_parser
Assembly subparser
For running the program using a fasta file
parser_blast : argparse.ArgumentParser.add_subparsers.add_parser
Blast subparser
For creating Blast DB
"""
parser = argparse.ArgumentParser(
prog=program_name,
description=
'Determines which reference sequence is more likely to be present in a'
' given sample',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('--version',
help='Version information',
action='version',
version='{prog} v{version}'.format(prog=parser.prog,
version=version))
subparsers = parser.add_subparsers(title='Subcommands',
description='Valid subcommands',
help='Additional help')
parser_reads = subparsers.add_parser(
'reads',
description='Run {prog} using fastq files. If running multiple'
' samples using the same reference sequences file,'
' consider use first "{prog} index"'
' subcommand.'.format(prog=parser.prog),
help='reads --help',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser_index = subparsers.add_parser(
'index',
description='Creates Bowtie2 index. This is useful when running the'
' same reference sequences file for different reads'
' dataset.',
help='index --help',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser_assembly = subparsers.add_parser(
'assembly',
description='Run {prog} using a fasta file. If running'
' multiple samples using the same DB sequence file,'
' consider use first "{prog} blast"'
' subcommand.'.format(prog=parser.prog),
help='assembly --help',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser_blast = subparsers.add_parser(
'blast',
description='Creates Blast DB. This is useful when running the same'
' DB sequence file for different assemblies.',
help='blast --help',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser_reads_required = parser_reads.add_argument_group('Required options')
parser_reads_required.add_argument(
'-f',
'--fastq',
nargs='+',
action=utils.required_length((1, 2), '--fastq'),
type=argparse.FileType('r'),
metavar='/path/to/input/file.fq.gz',
help=
'Path to single OR paired-end fastq files. If two files are passed, they'
' will be assumed as being the paired fastq files',
required=True)
parser_reads_reference = parser_reads.add_mutually_exclusive_group(
required=True)
parser_reads_reference.add_argument(
'-r',
'--reference',
nargs='+',
type=str,
metavar='/path/to/reference/sequences.file',
help=
'Path to reference sequences file. If Bowtie2 index was already produced,'
' only provide the file name that ends with ".1.bt2", but without this'
' termination (for example, for a Bowtie2 index'
' "/file/sequences.fasta.1.bt2", only provide "/file/sequences.fasta").'
' If no Bowtie2 index files are found, those will be created in --outdir.'
' If more than one file is passed, a type for each file will be'
' determined. Give the files name in the same order that the type must be'
' determined.')
parser_reads_reference.add_argument(
'--org',
nargs=2,
type=str.lower,
metavar=('escherichia', 'coli'),
help='Name of the organism with reference sequences provided together'
' with {} for typing ("seqtyping/reference_sequences/"'
' folder)'.format(parser.prog),
action=utils.arguments_choices_words(get_species_allowed(), '--org'))
parser_reads_optional_general = parser_reads.add_argument_group(
'General facultative options')
parser_reads_optional_general.add_argument(
'-o',
'--outdir',
type=str,
metavar='/path/to/output/directory/',
help='Path to the directory where the information will be stored'
' (default: ./',
required=False,
default='.')
parser_reads_optional_general.add_argument(
'-j',
'--threads',
type=int,
metavar='N',
help='Number of threads to use (default: 1)',
required=False,
default=1)
parser_reads_optional_general.add_argument('--mapRefTogether',
action='store_true',
help=argparse.SUPPRESS)
# parser_reads_optional_general.add_argument('--mapRefTogether', action='store_true',
# help='Map the reads against all references together')
parser_reads_optional_general.add_argument(
'--typeSeparator',
type=str,
metavar='_',
help='Last single character separating the general sequence header from'
' the last part containing the type (default: _)',
required=False,
default='_')
parser_reads_optional_general.add_argument(
'--extraSeq',
type=int,
metavar='N',
help='Sequence length added to both ends of target sequences (usefull to'
' improve reads mapping to the target one) that will be trimmed in'
' ReMatCh outputs (default when not using --org: 0)',
required=False,
default=0)
parser_reads_optional_general.add_argument(
'--minCovPresence',
type=int,
metavar='N',
help='Reference position minimum coverage depth to consider the position'
' to be present in the sample (default when not using --org: 5)',
required=False,
default=5)
parser_reads_optional_general.add_argument(
'--minCovCall',
type=int,
metavar='N',
help='Reference position minimum coverage depth to perform a base call'
' (default when not using --org: 10)',
required=False,
default=10)
parser_reads_optional_general.add_argument('--minFrequencyDominantAllele',
type=float,
metavar='0.6',
help=argparse.SUPPRESS,
required=False,
default=0.6)
# parser_optional_general.add_argument('--minFrequencyDominantAllele', type=float, metavar='0.6',
# help='Minimum relative frequency of the dominant allele coverage depth'
# ' (value between [0, 1]). Positions with lower values will be'
# ' considered as having multiple alleles (and will be coded as N)',
# required=False, default=0.6)
parser_reads_optional_general.add_argument(
'--minGeneCoverage',
type=int,
metavar='N',
help='Minimum percentage of target reference sequence covered to'
' consider a sequence to be present (value between [0, 100])'
' (default when not using --org: 60)',
required=False,
default=60)
parser_reads_optional_general.add_argument(
'--minDepthCoverage',
type=int,
metavar='N',
help='Minimum depth of coverage of target reference sequence to'
' consider a sequence to be present (default: 2)',
required=False,
default=2)
# parser_reads_optional_general.add_argument('--minGeneIdentity', type=int, metavar='N', help=argparse.SUPPRESS,
# required=False, default=80)
parser_reads_optional_general.add_argument(
'--minGeneIdentity',
type=int,
metavar='N',
help='Minimum percentage of identity of reference sequence covered to'
' consider a gene to be present (value between [0, 100]). One INDEL'
' will be considered as one difference (default: 80)',
required=False,
default=80)
parser_reads_optional_general.add_argument(
'--bowtieAlgo',
type=str,
metavar='"--very-sensitive-local"',
help='Bowtie2 alignment mode. It can be an end-to-end alignment'
' (unclipped alignment) or local alignment (soft clipped'
' alignment). Also, can choose between fast or sensitive'
' alignments. Please check Bowtie2 manual for extra information:'
' http://bowtie-bio.sourceforge.net/bowtie2/index.shtml .'
' This option should be provided between quotes and starting with'
' an empty space (like --bowtieAlgo " --very-fast") or using equal'
' sign (like --bowtieAlgo="--very-fast")',
required=False,
default='--very-sensitive-local')
parser_reads_optional_general.add_argument(
'--doNotRemoveConsensus',
action='store_true',
help='Do not remove ReMatCh consensus sequences')
parser_reads_optional_general.add_argument(
'--debug',
action='store_true',
help='Debug mode: do not remove temporary files')
parser_reads_optional_general.add_argument('--resume',
action='store_true',
help='Resume %(prog)s')
parser_reads_optional_general.add_argument(
'--notClean',
action='store_true',
help='Do not remove intermediate files')
parser_index_reference = parser_index.add_mutually_exclusive_group(
required=True)
parser_index_reference.add_argument(
'-r',
'--reference',
nargs='+',
type=str,
metavar='/path/to/reference/sequences.fasta',
help=
'Path to reference sequences file. If more than one file is passed, a'
' Bowtie2 index for each file will be created.')
parser_index_reference.add_argument(
'--org',
nargs=2,
type=str.lower,
metavar=('escherichia', 'coli'),
help='Name of the organism with reference sequences provided'
' ("seqtyping/reference_sequences/" folder) together'
' with seq_typing.py for typing',
action=utils.arguments_choices_words(get_species_allowed(), '--org'))
parser_index_optional_general = parser_index.add_argument_group(
'General facultative options')
parser_index_optional_general.add_argument(
'-o',
'--outdir',
type=str,
metavar='/path/to/output/directory/',
help='Path to the directory where the information will be stored'
' (default: ./)',
required=False,
default='.')
parser_index_optional_general.add_argument(
'-j',
'--threads',
type=int,
metavar='N',
required=False,
help='Number of threads to use (default: 1)',
default=1)
parser_assembly_required = parser_assembly.add_argument_group(
'Required options')
parser_assembly_required.add_argument(
'-f',
'--fasta',
nargs=1,
type=argparse.FileType('r'),
metavar='/path/to/query/assembly_file.fasta',
help='Path to fasta file containing the query sequences from which the'
' types should be assessed',
required=True)
parser_assembly_reference = parser_assembly.add_mutually_exclusive_group(
required=True)
parser_assembly_reference.add_argument(
'-b',
'--blast',
nargs='+',
type=argparse.FileType('r'),
metavar='/path/to/Blast/db.sequences.file',
help='Path to DB sequence file. If Blast DB was already produced, only'
' provide the file that do not end with ".n*" something (do not use for'
' example /blast_db.sequences.fasta.nhr). If no Blast DB is found for'
' the DB sequence file, one will be created in --outdir. If more than'
' one Blast DB file is passed, a type for each file will be determined.'
' Give the files in the same order that the type must be determined.')
parser_assembly_reference.add_argument(
'--org',
nargs=2,
type=str.lower,
metavar=('escherichia', 'coli'),
help='Name of the organism with DB sequence file provided'
' ("seqtyping/reference_sequences/" folder) together'
' with seq_typing.py for typing',
action=utils.arguments_choices_words(get_species_allowed(), '--org'))
parser_assembly_optional_reference = parser_assembly.add_argument_group(
'Required option for --blast')
parser_assembly_optional_reference.add_argument(
'-t',
'--type',
choices=['nucl', 'prot'],
type=str,
metavar='nucl',
help='Blast DB type (available options: %(choices)s)')
parser_assembly_optional_general = parser_assembly.add_argument_group(
'General facultative options')
parser_assembly_optional_general.add_argument(
'-o',
'--outdir',
type=str,
metavar='/path/to/output/directory/',
help='Path to the directory where the information will be stored'
' (default: ./)',
required=False,
default='.')
parser_assembly_optional_general.add_argument(
'-j',
'--threads',
type=int,
metavar='N',
required=False,
help='Number of threads to use (default: 1)',
default=1)
parser_assembly_optional_general.add_argument(
'--typeSeparator',
type=str,
metavar='_',
help='Last single character separating the general sequence header'
' from the last part containing the type (default: _)',
required=False,
default='_')
parser_assembly_optional_general.add_argument(
'--minGeneCoverage',
type=int,
metavar='N',
help='Minimum percentage of target reference sequence covered to'
' consider a sequence to be present (value between [0, 100])'
' (default when not using --org: 60)',
required=False,
default=60)
parser_assembly_optional_general.add_argument(
'--minGeneIdentity',
type=int,
metavar='N',
help='Minimum percentage of identity of reference sequence covered'
' to consider a gene to be present (value between [0, 100])'
' (default: 80)',
required=False,
default=80)
parser_assembly_optional_general.add_argument('--minDepthCoverage',
type=int,
metavar='N',
help=argparse.SUPPRESS,
required=False,
default=1)
parser_assembly_optional_general.add_argument(
'--debug',
action='store_true',
help='Debug mode: do not remove temporary files')
parser_assembly_optional_general.add_argument('--resume',
action='store_true',
help=argparse.SUPPRESS)
parser_blast_required = parser_blast.add_argument_group('Required options')
parser_blast_required.add_argument(
'-t',
'--type',
choices=['nucl', 'prot'],
type=str,
metavar='nucl',
help='Blast DB type (available options: %(choices)s)',
required=True)
parser_blast_reference = parser_blast.add_mutually_exclusive_group(
required=True)
parser_blast_reference.add_argument(
'-f',
'--fasta',
nargs='+',
type=argparse.FileType('r'),
metavar='/path/to/db.sequences.fasta',
help=
'Path to DB sequence file. If more than one file is passed, a Blast DB for'
' each file will be created.')
parser_blast_reference.add_argument(
'--org',
nargs=2,
type=str.lower,
metavar=('escherichia', 'coli'),
help='Name of the organism with DB sequence file provided'
' ("seqtyping/reference_sequences/" folder) together'
' with seq_typing.py for typing',
action=utils.arguments_choices_words(get_species_allowed(), '--org'))
parser_blast_optional_general = parser_blast.add_argument_group(
'General facultative options')
parser_blast_optional_general.add_argument(
'-o',
'--outdir',
type=str,
metavar='/path/to/output/directory/',
help='Path to the directory where the information will be stored'
' (default: ./)',
required=False,
default='.')
parser_reads.set_defaults(func=reads_subcommand)
parser_index.set_defaults(func=index_subcommand)
parser_assembly.set_defaults(func=assembly_subcommand)
parser_blast.set_defaults(func=blast_subcommand)
return parser, parser_reads, parser_index, parser_assembly, parser_blast