def parse_args(args):
op = ArgumentParser(description='Find groups of orthologous genes.')
#op.add_argument(dest='directory')
op.add_argument('-o', '--out', '--dir', dest='out', required=True)
op.add_argument('-g', '--gbs', '--annotations', dest='annotations')
op.add_argument('-p', '--proteins', '--proteomes', dest='proteomes')
op.add_argument('--no-download', dest='download_anno', action='store_false', default=True)
op.add_argument('-s', '--species', '--species-list', dest='species_list')
op.add_argument('-i', '--ids', '--ids-list', dest='ids_list')
op.add_argument('--prot-id-field', dest='prot_id_field', default=1)
op.usage = '''Finding orthogroups for a list of annotations / proteomes / ref ids / species.
Test runs:
python scenario_1.py --ids test_input/ids.txt -o test_ids
python scenario_1.py --proteomes test_input/proteins -o test_proteomes
Usage: %s [-p <proteomes dir>] [-a <.gb files dir>] [-i <gb ids file>] [-s <strain names file>]
[-o <dir>] [--jobs 30] [--start-from <step num>]
-o Output directory.
-g Directory with .gb files for references with annotations.
-p Directory with fasta (or faa, fa) files of protein sequences. If they
are named by their reference ids (i.e. NC_005816.1.fasta), annotations
will be downloaded from NCBI.
-i File with reference ids (will be fetched from NCBI).
-s File with a list of organism names as in Genbank.
--prot-id-field
When specifying proteomes, use this fasta id field number
to retrieve protein ids (default if 1, like >NC_005816.1|NP_995567.1 ...).
''' % basename(__file__)
#-a --annotations-dir Directory with .gb files.
#-p --proteomes-dir Directory with fasta files of proteomes.
#-i --ids-list File with reference ids (will be fetched from Genbank).
#-s --species-list File with a list of organism names as in Genbank.
# For example, "Salmonella enterica subsp. enterica serovar Typhi str. P-stx-12".
#'''
#indent = ' ' * len('usage: ' + basename(__file__) + ' ')
#op.usage = basename(__file__) + ' [--annotations-dir DIR]\n' + \
# indent + '[--proteomes-dir DIR]\n' + \
# indent + '[--ids-file FILE]\n' + \
# indent + '[--species-file FILE]\n'
add_common_arguments(op)
p = op.parse_args(args)
check_common_args(p)
if not p.out:
arg_parse_error('Specify output directory with -o.')
if isfile(expanduser(p.out)):
arg_parse_error('%s is a file' % p.out)
p.out = abspath(expanduser(p.out))
if not isdir(p.out):
makedirs(p.out)
if p.species_list:
check_file(expanduser(p.species_list))
p.species_list = abspath(expanduser(p.species_list))
if p.ids_list:
check_file(expanduser(p.ids_list))
p.ids_list = abspath(expanduser(p.ids_list))
if p.proteomes:
check_dir(expanduser(p.proteomes))
p.proteomes = abspath(expanduser(p.proteomes))
if p.annotations:
check_dir(expanduser(p.annotations))
p.annotations = abspath(expanduser(p.annotations))
#if p.species_list or p.ids_list:
# if not isdir(p.out):
# mkdir(p.out)
#else:
# if not p.directory:
# arg_parse_error('Directory or file must be specified.')
# check_dir(p.directory)
return p
def parse_args(args):
op = ArgumentParser(description='Find groups of orthologous genes.')
#op.add_argument(dest='directory')
op.add_argument('-o', '--out', '--dir', dest='out', required=True)
op.add_argument('-g', '--gbs', '--annotations', dest='annotations')
op.add_argument('-p', '--proteins', '--proteomes', dest='proteomes')
op.add_argument('--no-download',
dest='download_anno',
action='store_false',
default=True)
op.add_argument('-s', '--species', '--species-list', dest='species_list')
op.add_argument('-i', '--ids', '--ids-list', dest='ids_list')
op.add_argument('--prot-id-field', dest='prot_id_field', default=1)
op.usage = '''Finding orthogroups for a list of annotations / proteomes / ref ids / species.
Test runs:
python scenario_1.py --ids test_input/ids.txt -o test_ids
python scenario_1.py --proteomes test_input/proteins -o test_proteomes
Usage: %s [-p <proteomes dir>] [-a <.gb files dir>] [-i <gb ids file>] [-s <strain names file>]
[-o <dir>] [--jobs 30] [--start-from <step num>]
-o Output directory.
-g Directory with .gb files for references with annotations.
-p Directory with fasta (or faa, fa) files of protein sequences. If they
are named by their reference ids (i.e. NC_005816.1.fasta), annotations
will be downloaded from NCBI.
-i File with reference ids (will be fetched from NCBI).
-s File with a list of organism names as in Genbank.
--prot-id-field
When specifying proteomes, use this fasta id field number
to retrieve protein ids (default if 1, like >NC_005816.1|NP_995567.1 ...).
''' % basename(__file__)
#-a --annotations-dir Directory with .gb files.
#-p --proteomes-dir Directory with fasta files of proteomes.
#-i --ids-list File with reference ids (will be fetched from Genbank).
#-s --species-list File with a list of organism names as in Genbank.
# For example, "Salmonella enterica subsp. enterica serovar Typhi str. P-stx-12".
#'''
#indent = ' ' * len('usage: ' + basename(__file__) + ' ')
#op.usage = basename(__file__) + ' [--annotations-dir DIR]\n' + \
# indent + '[--proteomes-dir DIR]\n' + \
# indent + '[--ids-file FILE]\n' + \
# indent + '[--species-file FILE]\n'
add_common_arguments(op)
p = op.parse_args(args)
check_common_args(p)
if not p.out:
arg_parse_error('Specify output directory with -o.')
if isfile(expanduser(p.out)):
arg_parse_error('%s is a file' % p.out)
p.out = abspath(expanduser(p.out))
if not isdir(p.out):
makedirs(p.out)
if p.species_list:
check_file(expanduser(p.species_list))
p.species_list = abspath(expanduser(p.species_list))
if p.ids_list:
check_file(expanduser(p.ids_list))
p.ids_list = abspath(expanduser(p.ids_list))
if p.proteomes:
check_dir(expanduser(p.proteomes))
p.proteomes = abspath(expanduser(p.proteomes))
if p.annotations:
check_dir(expanduser(p.annotations))
p.annotations = abspath(expanduser(p.annotations))
#if p.species_list or p.ids_list:
# if not isdir(p.out):
# mkdir(p.out)
#else:
# if not p.directory:
# arg_parse_error('Directory or file must be specified.')
# check_dir(p.directory)
return p
def parse_args(args):
import argparse
op = argparse.ArgumentParser(description='Find groups of orthologous genes.')
#op.add_argument(dest='directory', required=False)
op.add_argument('-s1o', dest='directory', required=True)
op.add_argument('-s2o', '-o', dest='out_dir', required=False)
op.add_argument('-a', '--assemblies', dest='assemblies')
op.add_argument('-g', '--annotations', '--gbs', dest='annotations')
op.add_argument('-p', '--proteomes', '--proteins', dest='proteomes')
op.add_argument('-i', '--ids', '--ids-list', dest='ids_list')
op.add_argument('--prot-id-field', dest='prot_id_field', default=1)
op.add_argument('--skip-blast-singletones', dest='blast_singletones',
action='store_false', default=True)
op.add_argument('--blastdb', '--blast-db', dest='blastdb')
#-o: Output directory (if not specified, the input directory will be used).
op.usage = '''Extends an orthogroup database and orthogroups files.
First argument is a path to existed Scenario 1 output.
Test runs:
python scenario_2.py -s1o test_ids -s2o test_ids_new_ids --ids test_input/new_ids.txt
python scenario_2.py -s1o test_proteomes -s2o test_prots_new_prots --proteomes test_input/new_proteins
Usage: %s -s1o <scenario_1 result dir> -s2o <new output dir> [-a <assembies dir>] [-p <proteomes dir]
[-a <.gb files dir>] [-i <gb ids file>] [-s <strain names file>]
[--jobs 30] [--start-from <step num>]
[--blast-singletones] [--blast-db <path>]
-s1o Path to existed Scenario 1 output.
-s2o Output directory (optional, if ommited, the input directory will be used).
-a --assemblies: Directory with assemblies in fasta format.
-g Directory with .gb files for references with annotations.
-p Directory with fasta (or faa, fa) files of protein sequences. If they
are named by their reference ids (i.e. NC_005816.1.fasta), annotations
will be downloaded from NCBI.
-i File with reference ids (will be fetched from NCBI).
-s File with a list of organism names as in Genbank.
--prot-id-field
When specifying proteomes, use this fasta id field number
to retrieve protein ids (default if 1, like >NC_005816.1|NP_995567.1 ...).
--blast-singletones
Search newly added proteins agains NCBI database, if they did not fit
any group with known proteins.
--blastdb
Local BLAST database path. If not set, "blastdb" value in config.txt will be used.
If it was not set either, remote NCBI will be used.
''' % basename(__file__)
#indent = ' ' * len('usage: ' + basename(__file__) + ' ')
#op.usage = basename(__file__) + ' [--existing-blast-results TSV]\n' + \
# indent + '[--existing_proteomes DIR]\n' + \
# indent + '[--assembly FASTA]\n' + \
# indent + '[--genes GFF]\n' + \
# indent + '[--proteome FASTA]\n'
add_common_arguments(op)
p = op.parse_args(args)
check_common_args(p)
if p.assemblies:
check_dir(expanduser(p.assemblies))
p.assemblies = abspath(expanduser(p.assemblies))
if p.proteomes:
check_dir(expanduser(p.proteomes))
p.proteomes = abspath(expanduser(p.proteomes))
if p.ids_list:
check_file(expanduser(p.ids_list))
p.ids_list = abspath(expanduser(p.ids_list))
if p.annotations:
check_dir(expanduser(p.annotations))
p.annotations = abspath(expanduser(p.annotations))
if p.blastdb:
p.blastdb = abspath(expanduser(p.blastdb))
if not isdir(expanduser(p.directory)):
arg_parse_error('Directory %s does not exist.' % p.directory)
p.directory = abspath(expanduser(p.directory))
if not p.out_dir:
p.out_dir = p.directory
#arg_parse_error('Specify output directory with -o.')
if isfile(expanduser(p.out_dir)):
arg_parse_error('%s is a file' % p.out_dir)
p.out_dir = abspath(expanduser(p.out_dir))
return p
def main(args):
register_ctrl_c()
p = parse_args(args)
try:
if not exists(join(p.directory, 'intermediate')):
arg_parse_error('You need to run Scenario 1 on this directory first.')
if not p.out_dir:
p.out_dir = p.directory
working_dir = p.out_dir
with open(config_file) as cf:
conf = dict(
l.strip().split('=', 1) for l
in cf.readlines() if l.strip() and l.strip()[0] != '#')
start_from, start_after = get_starting_step(p.start_from, join(working_dir, log_fname))
if (not start_from or start_from == 1) and p.out_dir != p.directory:
log_text = ''
if isdir(p.out_dir):
if not p.overwrite:
files = [f for f in listdir(p.out_dir) if f and f[0] != '.']
#log.debug(files)
if files:
print('The output directory exists. Do you want to overwrite it? ' +
'(You can run with the --overwrite option to avoid this warning.)')
try:
raw_input('Press any key to overwrite and continue, or Ctrl+C to interrupt.\n> ')
except (EOFError, KeyboardInterrupt, SystemExit, GeneratorExit):
exit(1)
if exists(join(p.out_dir, log_fname)):
with open(join(p.out_dir, log_fname)) as log_f:
log_text = log_f.read()
rmtree(p.out_dir)
makedirs(p.out_dir)
rmdir(p.out_dir)
copytree(p.directory, p.out_dir)
if isfile(join(p.out_dir, log_fname)):
remove(join(p.out_dir, log_fname))
chdir(p.out_dir)
if log_text:
with open(join(p.out_dir, log_fname), 'w') as log_f:
log_f.write(log_text)
log_fpath = set_up_logging(p.debug, p.out_dir, 'a')
log.info('python ' + basename(__file__) + ' ' + ' '.join(args))
log.info('')
check_and_install_tools(p.debug, conf.get('db_vendor', 'sqlite'), log_fpath)
log.info('Changing to %s' % working_dir)
if not isdir(working_dir):
makedirs(working_dir)
chdir(working_dir)
set_up_config(working_dir)
# Building the workflow
workflow = Workflow(working_dir, id=make_workflow_id(working_dir),
cmdline_args=['python', __file__] + args)
log.info('Workflow id is "' + workflow.id + '"')
log.info('')
if conf.get('db_vendor', 'sqlite') == 'sqlite':
suffix = ''
else:
suffix = '_' + workflow.id
workflow.extend([
step_prepare_input(p),
steps.filter_proteomes(
min_length=int(p.min_length),
max_percent_stop=int(p.max_percent_stop)),
filter_new_proteomes(
new_proteomes_dir,
min_length=int(p.min_length),
max_percent_stop=int(p.max_percent_stop)),
steps.make_blast_db(),
steps.blast(
workflow.id,
p.threads or p.jobs or 30,
on_cluster=p.threads > 0,
new_good_proteomes=new_good_proteomes,
evalue=float(p.evalue)),
steps.parse_blast_results(),
steps.clean_database(suffix),
steps.install_schema(suffix),
steps.load_blast_results(suffix),
steps.find_pairs(suffix),
steps.dump_pairs_to_files(suffix),
steps.mcl(p.debug),
steps.step_save_orthogroups(new_proteomes_dir if not p.ids_list and p.blast_singletones else None)
])
blastdb = p.blastdb or conf.get('blastdb', None)
if not p.ids_list:
workflow.extend([step_blast_singletones(p.threads, p.blast_singletones, blastdb, p.debug, p.overwrite)])
result = workflow.run(
start_after, start_from,
overwrite=True,
ask_before=p.ask_each_step)
if result == 0:
log.info('Done.')
log.info('Log is in ' + join(working_dir, log_fname))
if isfile(join(working_dir, config.orthogroups_file)):
log.info('Groups are in ' + join(working_dir, config.orthogroups_file))
if isfile(config.nice_orthogroups_file):
log.info('Groups with aligned columns are in ' +
join(working_dir, config.nice_orthogroups_file))
else:
log.info('Groups in short format are in ' + join(working_dir, config.short_orthogroups_file))
return result
except (KeyboardInterrupt, SystemExit, GeneratorExit):
return 1
except Exception as e:
log.error('')
log.exception('Unexpected error!')
raise
def main(args):
register_ctrl_c()
p = parse_args(args)
try:
if not exists(join(p.directory, 'intermediate')):
arg_parse_error(
'You need to run Scenario 1 on this directory first.')
if not p.out_dir:
p.out_dir = p.directory
working_dir = p.out_dir
with open(config_file) as cf:
conf = dict(l.strip().split('=', 1) for l in cf.readlines()
if l.strip() and l.strip()[0] != '#')
start_from, start_after = get_starting_step(
p.start_from, join(working_dir, log_fname))
if (not start_from or start_from == 1) and p.out_dir != p.directory:
log_text = ''
if isdir(p.out_dir):
if not p.overwrite:
files = [
f for f in listdir(p.out_dir) if f and f[0] != '.'
]
#log.debug(files)
if files:
print(
'The output directory exists. Do you want to overwrite it? '
+
'(You can run with the --overwrite option to avoid this warning.)'
)
try:
raw_input(
'Press any key to overwrite and continue, or Ctrl+C to interrupt.\n> '
)
except (EOFError, KeyboardInterrupt, SystemExit,
GeneratorExit):
exit(1)
if exists(join(p.out_dir, log_fname)):
with open(join(p.out_dir, log_fname)) as log_f:
log_text = log_f.read()
rmtree(p.out_dir)
makedirs(p.out_dir)
rmdir(p.out_dir)
copytree(p.directory, p.out_dir)
if isfile(join(p.out_dir, log_fname)):
remove(join(p.out_dir, log_fname))
chdir(p.out_dir)
if log_text:
with open(join(p.out_dir, log_fname), 'w') as log_f:
log_f.write(log_text)
log_fpath = set_up_logging(p.debug, p.out_dir, 'a')
log.info('python ' + basename(__file__) + ' ' + ' '.join(args))
log.info('')
check_and_install_tools(p.debug, conf.get('db_vendor', 'sqlite'),
log_fpath)
log.info('Changing to %s' % working_dir)
if not isdir(working_dir):
makedirs(working_dir)
chdir(working_dir)
set_up_config(working_dir)
# Building the workflow
workflow = Workflow(working_dir,
id=make_workflow_id(working_dir),
cmdline_args=['python', __file__] + args)
log.info('Workflow id is "' + workflow.id + '"')
log.info('')
if conf.get('db_vendor', 'sqlite') == 'sqlite':
suffix = ''
else:
suffix = '_' + workflow.id
workflow.extend([
step_prepare_input(p),
steps.filter_proteomes(min_length=int(p.min_length),
max_percent_stop=int(p.max_percent_stop)),
filter_new_proteomes(new_proteomes_dir,
min_length=int(p.min_length),
max_percent_stop=int(p.max_percent_stop)),
steps.make_blast_db(),
steps.blast(workflow.id,
p.threads or p.jobs or 30,
on_cluster=p.threads > 0,
new_good_proteomes=new_good_proteomes,
evalue=float(p.evalue)),
steps.parse_blast_results(),
steps.clean_database(suffix),
steps.install_schema(suffix),
steps.load_blast_results(suffix),
steps.find_pairs(suffix),
steps.dump_pairs_to_files(suffix),
steps.mcl(p.debug),
steps.step_save_orthogroups(new_proteomes_dir if not p.ids_list
and p.blast_singletones else None)
])
blastdb = p.blastdb or conf.get('blastdb', None)
if not p.ids_list:
workflow.extend([
step_blast_singletones(p.threads, p.blast_singletones, blastdb,
p.debug, p.overwrite)
])
result = workflow.run(start_after,
start_from,
overwrite=True,
ask_before=p.ask_each_step)
if result == 0:
log.info('Done.')
log.info('Log is in ' + join(working_dir, log_fname))
if isfile(join(working_dir, config.orthogroups_file)):
log.info('Groups are in ' +
join(working_dir, config.orthogroups_file))
if isfile(config.nice_orthogroups_file):
log.info('Groups with aligned columns are in ' +
join(working_dir, config.nice_orthogroups_file))
else:
log.info('Groups in short format are in ' +
join(working_dir, config.short_orthogroups_file))
return result
except (KeyboardInterrupt, SystemExit, GeneratorExit):
return 1
except Exception as e:
log.error('')
log.exception('Unexpected error!')
raise
def parse_args(args):
import argparse
op = argparse.ArgumentParser(
description='Find groups of orthologous genes.')
#op.add_argument(dest='directory', required=False)
op.add_argument('-s1o', dest='directory', required=True)
op.add_argument('-s2o', '-o', dest='out_dir', required=False)
op.add_argument('-a', '--assemblies', dest='assemblies')
op.add_argument('-g', '--annotations', '--gbs', dest='annotations')
op.add_argument('-p', '--proteomes', '--proteins', dest='proteomes')
op.add_argument('-i', '--ids', '--ids-list', dest='ids_list')
op.add_argument('--prot-id-field', dest='prot_id_field', default=1)
op.add_argument('--skip-blast-singletones',
dest='blast_singletones',
action='store_false',
default=True)
op.add_argument('--blastdb', '--blast-db', dest='blastdb')
#-o: Output directory (if not specified, the input directory will be used).
op.usage = '''Extends an orthogroup database and orthogroups files.
First argument is a path to existed Scenario 1 output.
Test runs:
python scenario_2.py -s1o test_ids -s2o test_ids_new_ids --ids test_input/new_ids.txt
python scenario_2.py -s1o test_proteomes -s2o test_prots_new_prots --proteomes test_input/new_proteins
Usage: %s -s1o <scenario_1 result dir> -s2o <new output dir> [-a <assembies dir>] [-p <proteomes dir]
[-a <.gb files dir>] [-i <gb ids file>] [-s <strain names file>]
[--jobs 30] [--start-from <step num>]
[--blast-singletones] [--blast-db <path>]
-s1o Path to existed Scenario 1 output.
-s2o Output directory (optional, if ommited, the input directory will be used).
-a --assemblies: Directory with assemblies in fasta format.
-g Directory with .gb files for references with annotations.
-p Directory with fasta (or faa, fa) files of protein sequences. If they
are named by their reference ids (i.e. NC_005816.1.fasta), annotations
will be downloaded from NCBI.
-i File with reference ids (will be fetched from NCBI).
-s File with a list of organism names as in Genbank.
--prot-id-field
When specifying proteomes, use this fasta id field number
to retrieve protein ids (default if 1, like >NC_005816.1|NP_995567.1 ...).
--blast-singletones
Search newly added proteins agains NCBI database, if they did not fit
any group with known proteins.
--blastdb
Local BLAST database path. If not set, "blastdb" value in config.txt will be used.
If it was not set either, remote NCBI will be used.
''' % basename(__file__)
#indent = ' ' * len('usage: ' + basename(__file__) + ' ')
#op.usage = basename(__file__) + ' [--existing-blast-results TSV]\n' + \
# indent + '[--existing_proteomes DIR]\n' + \
# indent + '[--assembly FASTA]\n' + \
# indent + '[--genes GFF]\n' + \
# indent + '[--proteome FASTA]\n'
add_common_arguments(op)
p = op.parse_args(args)
check_common_args(p)
if p.assemblies:
check_dir(expanduser(p.assemblies))
p.assemblies = abspath(expanduser(p.assemblies))
if p.proteomes:
check_dir(expanduser(p.proteomes))
p.proteomes = abspath(expanduser(p.proteomes))
if p.ids_list:
check_file(expanduser(p.ids_list))
p.ids_list = abspath(expanduser(p.ids_list))
if p.annotations:
check_dir(expanduser(p.annotations))
p.annotations = abspath(expanduser(p.annotations))
if p.blastdb:
p.blastdb = abspath(expanduser(p.blastdb))
if not isdir(expanduser(p.directory)):
arg_parse_error('Directory %s does not exist.' % p.directory)
p.directory = abspath(expanduser(p.directory))
if not p.out_dir:
p.out_dir = p.directory
#arg_parse_error('Specify output directory with -o.')
if isfile(expanduser(p.out_dir)):
arg_parse_error('%s is a file' % p.out_dir)
p.out_dir = abspath(expanduser(p.out_dir))
return p
