def dep_check_blast(dir_dep, os_id, dist_id, debian_dists, redhat_dists,
force):
if os_id == 'mac':
url = ('https://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/2.10.1/'
'ncbi-blast-2.10.1+-x64-macosx.tar.gz')
elif os_id == 'linux':
if dist_id in debian_dists:
url = ('https://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/'
'2.10.1/ncbi-blast-2.10.1+-x64-linux.tar.gz')
elif dist_id in redhat_dists:
url = ('https://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/'
'2.10.1/ncbi-blast-2.10.1+-x64-linux.tar.gz')
dnld_path = opj(dir_dep, 'ncbi-blast.tar.gz')
makeblastdb = None
blastn = None
tblastn = None
try:
if force is True:
raise
makeblastdb = which('makeblastdb')
blastn = which('blastn')
tblastn = which('tblastn')
run([makeblastdb, '-help'])
except Exception:
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'ncbi-blast'))
makeblastdb = opj(dir_bin, 'bin', 'makeblastdb')
blastn = opj(dir_bin, 'bin', 'blastn')
tblastn = opj(dir_bin, 'bin', 'tblastn')
run([makeblastdb, '-help'])
except Exception:
Log.wrn('BLAST+ was not found on this system, trying to download.')
download_file(url, dnld_path)
tar_ref = tarfile.open(dnld_path, 'r:gz')
tar_ref.extractall(dir_dep)
tar_ref.close()
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'ncbi-blast'))
makeblastdb = opj(dir_bin, 'bin', 'makeblastdb')
blastn = opj(dir_bin, 'bin', 'blastn')
tblastn = opj(dir_bin, 'bin', 'tblastn')
if not ope(makeblastdb) or \
not ope(blastn) or \
not ope(tblastn):
Log.err('Could not download BLAST+.')
return None, None, None
regexp = r'\sblast\s([\d\.]*)'
v = get_dep_version([makeblastdb, '-version'], regexp)
Log.msg('makeblastdb is available:', v + ' ' + makeblastdb)
v = get_dep_version([blastn, '-version'], regexp)
Log.msg('blastn is available:', v + ' ' + blastn)
v = get_dep_version([tblastn, '-version'], regexp)
Log.msg('tblastn is available:', v + ' ' + tblastn)
return makeblastdb, blastn, tblastn
def _parse_taxa(taxa, tax_group, taxonomy, config_file_path):
txids = list()
for tax in taxa:
if tax.isdigit():
txids.append(int(tax))
else:
# tax_orig = tax
txid = taxonomy.tax_id_for_name_and_group_tax_id(
name=tax, group_tax_id=tax_group)
if txid is None:
txid = taxonomy.tax_id_for_name_and_group_tax_id(
name=tax.split(' ')[0], group_tax_id=tax_group)
if txid is None:
txids.append(txid)
msg = 'NCBI taxonomy ID could not be found for:'
Log.wrn(msg, tax)
# replace_line_in_file(
# file_path=config_file_path,
# line_str=tax_orig,
# replace_str='; NCBI taxid not found: ' + tax)
else:
txids.append(int(txid))
msg = 'NCBI taxonomy ID for ' + tax + ' is:'
Log.msg(msg, str(txid))
# replace_line_in_file(
# file_path=config_file_path,
# line_str=tax_orig,
# replace_str='; ' + tax + '\n' + str(txid))
return txids
def dnld_refseqs_for_taxid(taxid,
filter_term,
taxonomy,
dir_cache_refseqs,
query='',
db='nuccore'):
ft = None
if filter_term == 'plastid':
ft = '("chloroplast"[filter] OR "plastid"[filter])'
else:
ft = '("' + filter_term + '"[filter])'
tax_terms = tuple(reversed(taxonomy.lineage_for_taxid(taxid)['names']))
for tax_term in tax_terms:
if tax_term is None:
tax_term = taxonomy.scientific_name_for_taxid(taxid)
term = '"RefSeq"[Keyword] AND "{}"[Primary Organism] AND {}'.format(
tax_term, ft)
term = query + term
accs = set(accs_eutil(search_eutil(db, term)))
if len(accs) > 0:
plural = 'sequences'
if len(accs) == 1:
plural = 'sequence'
Log.msg(
'Found {} RefSeq {} {} for'.format(len(accs), filter_term,
plural), tax_term)
# Random sample ###################################################
if len(accs) > 10:
Log.wrn('Using a random sample of ten RefSeq sequences.')
random.seed(a=len(accs), version=2)
accs = set(random.sample(accs, 10))
###################################################################
break
else:
Log.wrn(
'No RefSeq {} sequences were found for'.format(filter_term),
tax_term)
cache_path = opj(
dir_cache_refseqs,
filter_term + '__' + tax_term.replace(' ', '_') + '.fasta')
parsed_fasta_cache = {}
if ope(cache_path):
parsed_fasta_cache = read_fasta(cache_path,
seq_type=SEQ_TYPE_NT,
def_to_first_space=True)
parsed_fasta_cache = seq_records_to_dict(parsed_fasta_cache)
for acc in parsed_fasta_cache:
if acc in accs:
accs.remove(acc)
if len(accs) > 0:
parsed_fasta = dnld_ncbi_seqs(db, list(accs))
parsed_fasta = seq_records_to_dict(parsed_fasta, prepend_acc=True)
parsed_fasta.update(parsed_fasta_cache)
write_fasta(parsed_fasta, cache_path)
return cache_path
def dep_check_bowtie2(dir_dep, os_id, force):
if os_id == 'mac':
url = ('https://sourceforge.net/projects/bowtie-bio/files/bowtie2/'
'2.4.1/bowtie2-2.4.1-macos-x86_64.zip/download')
elif os_id == 'linux':
url = ('https://sourceforge.net/projects/bowtie-bio/files/bowtie2/'
'2.4.1/bowtie2-2.4.1-linux-x86_64.zip/download')
dnld_path = opj(dir_dep, 'bowtie2.zip')
try:
if force is True:
raise
bowtie2 = which('bowtie2')
bowtie2_build = which('bowtie2-build')
run([bowtie2, '-h'])
run([bowtie2_build, '-h'])
except Exception:
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'bowtie2'))
bowtie2 = opj(dir_bin, 'bowtie2')
bowtie2_build = opj(dir_bin, 'bowtie2-build')
run([bowtie2, '-h'])
run([bowtie2_build, '-h'])
except Exception:
Log.wrn('Bowtie 2 was not found on this system, trying to '
'download.')
download_file(url, dnld_path)
zip_ref = zipfile.ZipFile(dnld_path, 'r')
zip_ref.extractall(dir_dep)
zip_ref.close()
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'bowtie2'))
bowtie2 = opj(dir_bin, 'bowtie2')
bowtie2_build = opj(dir_bin, 'bowtie2-build')
bowtie2_execs = ('', '-align-l', '-align-l-debug', '-align-s',
'-align-s-debug', '-build', '-build-l',
'-build-l-debug', '-build-s', '-build-s-debug',
'-inspect', '-inspect-l', '-inspect-l-debug',
'-inspect-s', '-inspect-s-debug')
for bt2exe in bowtie2_execs:
chmod(
bowtie2 + bt2exe, stat.S_IRWXU | stat.S_IRGRP
| stat.S_IXGRP | stat.S_IROTH | stat.S_IXOTH)
if not ope(bowtie2):
Log.err('Could not download Bowtie 2.')
return None, None
regexp = r'^.*?version\s([\d\.]*)'
v = get_dep_version([bowtie2, '--version'], regexp)
Log.msg('bowtie2 is available:', v + ' ' + bowtie2)
v = get_dep_version([bowtie2_build, '--version'], regexp)
Log.msg('bowtie2-build is available:', v + ' ' + bowtie2_build)
return bowtie2, bowtie2_build
def find_api_key() -> str:
global_variables = globals()
if 'ENTREZ_KEY' in global_variables:
api_key = global_variables['ENTREZ_KEY']
elif 'ENTREZ_KEY' in os.environ:
api_key = os.environ['ENTREZ_KEY']
else:
Log.wrn('Warning:', 'ENTREZ_KEY is not defined.')
api_key = None
return api_key
def dep_check_sra_toolkit(dir_dep, os_id, dist_id, debian_dists, redhat_dists,
force):
if os_id == 'mac':
url = ('https://ftp-trace.ncbi.nlm.nih.gov/sra/sdk/2.10.8/'
'sratoolkit.2.10.8-mac64.tar.gz')
elif os_id == 'linux':
if dist_id in debian_dists:
url = ('https://ftp-trace.ncbi.nlm.nih.gov/sra/sdk/2.10.8/'
'sratoolkit.2.10.8-ubuntu64.tar.gz')
elif dist_id in redhat_dists:
url = ('https://ftp-trace.ncbi.nlm.nih.gov/sra/sdk/2.10.8/'
'sratoolkit.2.10.8-centos_linux64.tar.gz')
dnld_path = opj(dir_dep, 'sra-toolkit.tar.gz')
fasterq_dump = None
try:
if force is True:
raise
fasterq_dump = which('fasterq-dump')
dir_bin = dirname(fasterq_dump).strip('bin')
_ensure_vdb_cfg(dir_bin)
run(fasterq_dump)
except Exception:
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'sratoolkit'))
_ensure_vdb_cfg(dir_bin)
fasterq_dump = opj(dir_bin, 'bin', 'fasterq-dump')
run(fasterq_dump)
except Exception:
Log.wrn('SRA Toolkit was not found on this system, trying to '
'download.')
download_file(url, dnld_path)
tar_ref = tarfile.open(dnld_path, 'r:gz')
tar_ref.extractall(dir_dep)
tar_ref.close()
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'sratoolkit'))
fasterq_dump = opj(dir_bin, 'bin', 'fasterq-dump')
_ensure_vdb_cfg(dir_bin)
if not ope(fasterq_dump):
Log.err('Could not download SRA Toolkit.')
return None
v = get_dep_version([fasterq_dump, '--version'], r':\s([\d\.]*)')
if v == '?':
v = get_dep_version([fasterq_dump, '--version'], r'version\s([\d\.]*)')
Log.msg('fasterq-dump is available:', v + ' ' + fasterq_dump)
return fasterq_dump
def dep_check_kakapolib(force=False, quiet=False):
kkpl = KAKAPOLIB
if not ope(kkpl):
if quiet is False:
Log.wrn('Compiling kakapolib.')
run(['make', 'install'], cwd=DIR_C_SRC)
if ope(kkpl):
if quiet is False:
Log.msg('kakapolib is available:', kkpl)
else:
Log.err('Compilation of kakapolib failed.')
return None
return ctypes.CDLL(kkpl)
def dep_check_seqtk(dir_dep, force):
url = 'https://github.com/lh3/seqtk/archive/master.zip'
dnld_path = opj(dir_dep, 'seqtk.zip')
dir_bin = opj(dir_dep, 'seqtk-master')
fp = NamedTemporaryFile()
fp.write(str.encode('>seq' + lns + 'ATGC'))
fp.seek(0)
cmd = ['', 'seq', '-r', fp.name]
try:
if force is True:
raise
seqtk = which('seqtk')
cmd[0] = seqtk
run(cmd, do_not_raise=True)
except Exception:
try:
seqtk = opj(dir_bin, 'seqtk')
cmd[0] = seqtk
run(cmd, do_not_raise=True)
except Exception:
Log.wrn('Seqtk was not found on this system, trying to download.')
download_file(url, dnld_path)
zip_ref = zipfile.ZipFile(dnld_path, 'r')
zip_ref.extractall(dir_dep)
zip_ref.close()
try:
Log.wrn('Compiling Seqtk.')
run('make', cwd=dir_bin)
run(cmd, do_not_raise=True)
except Exception:
replace_line_in_file(opj(dir_bin, 'Makefile'), 'CC=gcc',
'CC=cc')
try:
run('make', cwd=dir_bin)
run(cmd, do_not_raise=True)
except Exception:
Log.err(
'Something went wrong while trying to compile Seqtk.')
Log.msg('Try downloading and installing it manually from: '
'https://github.com/lh3/seqtk')
fp.close()
return None
fp.close()
v = get_dep_version([seqtk], r'Version\:\s([\d\w\.\-]*)')
Log.msg('Seqtk is available:', v + ' ' + seqtk)
return seqtk
def dep_check_vsearch(dir_dep, os_id, dist_id, debian_dists, redhat_dists,
force):
if os_id == 'mac':
url = ('https://github.com/torognes/vsearch/releases/download/v2.15.0/'
'vsearch-2.15.0-macos-x86_64.tar.gz')
elif os_id == 'linux':
if dist_id in debian_dists:
url = ('https://github.com/torognes/vsearch/releases/download/'
'v2.15.0/vsearch-2.15.0-linux-x86_64.tar.gz')
elif dist_id in redhat_dists:
url = ('https://github.com/torognes/vsearch/releases/download/'
'v2.15.0/vsearch-2.15.0-linux-x86_64.tar.gz')
dnld_path = opj(dir_dep, 'vsearch.tar.gz')
try:
if force is True:
raise
vsearch = which('vsearch')
run(vsearch)
except Exception:
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'vsearch'))
vsearch = opj(dir_bin, 'bin', 'vsearch')
run(vsearch)
except Exception:
Log.wrn(
'Vsearch was not found on this system, trying to download.')
download_file(url, dnld_path)
tar_ref = tarfile.open(dnld_path, 'r:gz')
tar_ref.extractall(dir_dep)
tar_ref.close()
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'vsearch'))
vsearch = opj(dir_bin, 'bin', 'vsearch')
if not ope(vsearch):
Log.err('Could not download Vsearch.')
return None
else:
run(vsearch)
except Exception:
Log.err('Vsearch was downloaded, but does not execute.')
Log.msg('Try downloading and installing it manually from: '
'https://github.com/torognes/vsearch')
return None
v = get_dep_version([vsearch, '-version'], r'vsearch\sv([\d\.]*)')
Log.msg('Vsearch is available:', v + ' ' + vsearch)
return vsearch
def dep_check_spades(dir_dep, os_id, force):
if os_id == 'mac':
url = ('http://cab.spbu.ru/files/release3.14.1/'
'SPAdes-3.14.1-Darwin.tar.gz')
elif os_id == 'linux':
url = ('http://cab.spbu.ru/files/release3.14.1/'
'SPAdes-3.14.1-Linux.tar.gz')
dnld_path = opj(dir_dep, 'SPAdes.tar.gz')
try:
if force is True:
raise
spades = which('spades.py')
run([PY3, spades])
except Exception:
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'SPAdes'))
spades = opj(dir_bin, 'bin', 'spades.py')
run([PY3, spades])
except Exception:
Log.wrn('SPAdes was not found on this system, trying to download.')
try:
download_file(url, dnld_path)
tar_ref = tarfile.open(dnld_path, 'r:gz')
tar_ref.extractall(dir_dep)
tar_ref.close()
except Exception:
Log.err('Could not download SPAdes.')
return None
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'SPAdes'))
spades = opj(dir_bin, 'bin', 'spades.py')
# replace_line_in_file(spades,
# '#!/usr/bin/env python',
# '#!/usr/bin/env python3')
if ope(spades):
run([PY3, spades])
else:
Log.err('Could not download SPAdes.')
return None
except Exception:
Log.err('SPAdes was downloaded, but does not execute.')
return None
v = get_dep_version([PY3, spades, '--version'], r'^.*SPAdes.*v([\d\.]*)')
Log.msg('SPAdes is available:', v + ' ' + spades)
return spades
def run_spades(se_fastq_files, pe_fastq_files, dir_spades_assemblies,
spades, dir_temp, ss, threads, ram):
if len(se_fastq_files) > 0 or len(pe_fastq_files) > 0:
if spades is None:
Log.err('SPAdes is not available. Cannot continue. Exiting.')
exit(0)
for se in se_fastq_files:
dir_results = opj(dir_spades_assemblies, se + '__' + ss)
fq_path = se_fastq_files[se]['vsearch_results_path' + '__' + ss]
se_fastq_files[se]['spades_assembly' + '__' + ss] = None
if ope(dir_results):
Log.msg('SPAdes assembly already exists:', se)
else:
make_dirs(dir_results)
Log.msg('Running SPAdes on:', se)
run_spades_se(spades,
out_dir=dir_results,
input_file=fq_path,
threads=threads,
memory=ram,
rna=True)
assmbl_path = opj(dir_results, 'transcripts.fasta')
if ope(assmbl_path):
count = len(read_fasta(assmbl_path, SEQ_TYPE_NT))
tr_str = ' transcripts.'
if count == 1:
tr_str = ' transcript.'
Log.msg('SPAdes produced ' + str(count) + tr_str, False)
se_fastq_files[se]['spades_assembly' + '__' + ss] = assmbl_path
else:
Log.wrn('SPAdes produced no transcripts.', False)
for pe in pe_fastq_files:
dir_results = opj(dir_spades_assemblies, pe + '__' + ss)
fq_paths = pe_fastq_files[pe]['vsearch_results_path' + '__' + ss]
pe_fastq_files[pe]['spades_assembly' + '__' + ss] = None
if ope(dir_results):
Log.msg('SPAdes assembly already exists:', pe)
else:
make_dirs(dir_results)
Log.msg('Running SPAdes on: ' + pe)
if osstat(fq_paths[0]).st_size > 0 and \
osstat(fq_paths[1]).st_size > 0:
run_spades_pe(spades,
out_dir=dir_results,
input_files=fq_paths,
threads=threads,
memory=ram,
rna=True)
else:
_ = opj(dir_temp, 'temp.fasta')
combine_text_files(fq_paths, _)
run_spades_se(spades,
out_dir=dir_results,
input_file=_,
threads=threads,
memory=ram,
rna=True)
osremove(_)
assmbl_path = opj(dir_results, 'transcripts.fasta')
if ope(assmbl_path):
count = len(read_fasta(assmbl_path, SEQ_TYPE_NT))
tr_str = ' transcripts.'
if count == 1:
tr_str = ' transcript.'
Log.msg('SPAdes produced ' + str(count) + tr_str, False)
pe_fastq_files[pe]['spades_assembly' + '__' + ss] = assmbl_path
else:
Log.wrn('SPAdes produced no transcripts.', False)
def run_kraken_filters(order, dbs, base_name, in_files, dir_out, confidence,
kraken2, threads, dir_temp):
dbs_ordered = OrderedDict()
for dbn in order:
db_name = dbn[0]
if db_name in dbs:
dbs_ordered[db_name] = dbs[db_name]
else:
Log.wrn('Kraken2 database not found:', db_name)
# SE
if isinstance(in_files, (str, bytes)):
in_file = in_files
_, in_file_ext, _ = splitext_gz(in_file)
for i, db in enumerate(dbs_ordered):
Log.msg('Filtering SE reads using Kraken2 database:', db)
dir_out_db = opj(dir_out, db)
make_dirs(dir_out_db)
report_file = opj(dir_out_db, base_name + '.txt')
out_class_file = opj(dir_out_db, base_name + in_file_ext)
out_unclass_file = opj(
dir_temp,
base_name + '_' + db + '_kraken2_unclassified' + in_file_ext)
if stat(in_file
).st_size > 0: # Kraken2 freaks out if the file is empty.
run_kraken_se(kraken=kraken2,
db=dbs_ordered[db],
in_file=in_file,
out_class_file=out_class_file,
out_unclass_file=out_unclass_file,
report_file=report_file,
confidence=confidence,
threads=threads,
dir_temp=dir_temp)
else:
copyfile(in_file, out_class_file)
copyfile(in_file, out_unclass_file)
if i > 0:
remove(in_file)
in_file = out_unclass_file
move(in_file, opj(dir_out, base_name + in_file_ext))
# PE
elif isinstance(in_files, (list, tuple)):
assert len(in_files) > 1
_, in_file_ext, _ = splitext_gz(in_files[0])
in_file_R1 = in_files[0]
in_file_R2 = in_files[1]
if len(in_files) > 2:
in_file = in_files[2]
for i, db in enumerate(dbs_ordered):
Log.msg(
'Filtering unpaired forward reads using Kraken2 database:',
db)
dir_out_db = opj(dir_out, db)
make_dirs(dir_out_db)
report_file = opj(dir_out_db, base_name + '_unpaired_1.txt')
out_class_file = opj(dir_out_db,
base_name + '_unpaired_1' + in_file_ext)
out_unclass_file = opj(
dir_temp, base_name + '_' + db + '_kraken2_unclassified' +
in_file_ext)
if stat(in_file).st_size > 0:
run_kraken_se(kraken=kraken2,
db=dbs_ordered[db],
in_file=in_file,
out_class_file=out_class_file,
out_unclass_file=out_unclass_file,
report_file=report_file,
confidence=confidence,
threads=threads,
dir_temp=dir_temp)
else:
copyfile(in_file, out_class_file)
copyfile(in_file, out_unclass_file)
if i > 0:
remove(in_file)
in_file = out_unclass_file
move(in_file, opj(dir_out,
base_name + '_unpaired_1' + in_file_ext))
if len(in_files) == 4:
in_file = in_files[3]
for i, db in enumerate(dbs_ordered):
Log.msg(
'Filtering unpaired reverse reads using Kraken2 database:',
db)
dir_out_db = opj(dir_out, db)
make_dirs(dir_out_db)
report_file = opj(dir_out_db, base_name + '_unpaired_2.txt')
out_class_file = opj(dir_out_db,
base_name + '_unpaired_2' + in_file_ext)
out_unclass_file = opj(
dir_temp, base_name + '_' + db + '_kraken2_unclassified' +
in_file_ext)
if stat(in_file).st_size > 0:
run_kraken_se(kraken=kraken2,
db=dbs_ordered[db],
in_file=in_file,
out_class_file=out_class_file,
out_unclass_file=out_unclass_file,
report_file=report_file,
confidence=confidence,
threads=threads,
dir_temp=dir_temp)
else:
copyfile(in_file, out_class_file)
copyfile(in_file, out_unclass_file)
if i > 0:
remove(in_file)
in_file = out_unclass_file
move(in_file, opj(dir_out,
base_name + '_unpaired_2' + in_file_ext))
for i, db in enumerate(dbs_ordered):
Log.msg('Filtering paired reads using Kraken2 database:', db)
dir_out_db = opj(dir_out, db)
make_dirs(dir_out_db)
report_file = opj(dir_out_db, base_name + '_paired.txt')
out_class_file = opj(dir_out_db,
base_name + '_paired#' + in_file_ext)
out_unclass_file = opj(
dir_temp, base_name + '_' + db + '_kraken2_unclassified' +
'_paired#' + in_file_ext)
if stat(in_file_R1).st_size > 0 and stat(in_file_R2).st_size > 0:
run_kraken_pe(kraken=kraken2,
db=dbs_ordered[db],
in_file_1=in_file_R1,
in_file_2=in_file_R2,
out_class_file=out_class_file,
out_unclass_file=out_unclass_file,
report_file=report_file,
confidence=confidence,
threads=threads,
dir_temp=dir_temp)
else:
copyfile(
in_file_R1,
copyfile(in_file_R1, out_class_file.replace('#', '_1')))
copyfile(
in_file_R2,
copyfile(in_file_R2, out_class_file.replace('#', '_2')))
copyfile(
in_file_R1,
copyfile(in_file_R1, out_unclass_file.replace('#', '_1')))
copyfile(
in_file_R2,
copyfile(in_file_R2, out_unclass_file.replace('#', '_2')))
if i > 0:
remove(in_file_R1)
remove(in_file_R2)
in_file_R1 = out_unclass_file.replace('#', '_1')
in_file_R2 = out_unclass_file.replace('#', '_2')
move(in_file_R1, opj(dir_out, base_name + '_paired_1' + in_file_ext))
move(in_file_R2, opj(dir_out, base_name + '_paired_2' + in_file_ext))
def run_rcorrector(se_fastq_files, pe_fastq_files, dir_fq_cor_data, rcorrector,
threads, dir_temp, should_run):
if len(se_fastq_files) > 0 or len(pe_fastq_files) > 0:
print()
if should_run is False:
Log.wrn('Skipping Rcorrector as requested.')
else:
Log.inf('Running Rcorrector.')
if rcorrector is None:
Log.err('Rcorrector is not available. Cannot continue. Exiting.')
exit(0)
for se in se_fastq_files:
dir_fq_cor_data_sample = opj(dir_fq_cor_data, se)
fq_path = se_fastq_files[se]['path']
r_mode, w_mode, a_mode, fqopen, ext = plain_or_gzip(fq_path)
log_f = opj(dir_fq_cor_data_sample, se + '.txt')
out_f = opj(dir_fq_cor_data_sample, se + '.fastq' + ext)
se_fastq_files[se]['cor_path_fq'] = out_f
if should_run is False:
se_fastq_files[se]['cor_path_fq'] = fq_path
continue
if ope(dir_fq_cor_data_sample):
Log.msg('Corrected FASTQ file already exists:', se)
else:
make_dirs(dir_fq_cor_data_sample)
Log.msg('SE mode:', se)
run_rcorrector_se(rcorrector=rcorrector,
in_file=fq_path,
out_dir=dir_fq_cor_data_sample,
threads=threads,
dir_temp=dir_temp)
fq_base_path = opj(dir_fq_cor_data_sample, basename(fq_path))
fq_cor_path = splitext_gz(fq_base_path)[0] + '.cor.fq' + ext
filter_unc_se(in_file=fq_cor_path, out_file=out_f, log_file=log_f)
remove(fq_cor_path)
for pe in pe_fastq_files:
dir_fq_cor_data_sample = opj(dir_fq_cor_data, pe)
fq_path_1 = pe_fastq_files[pe]['path'][0]
fq_path_2 = pe_fastq_files[pe]['path'][1]
fq_path_3 = None
out_f_3 = None
r_mode, w_mode, a_mode, fqopen, ext = plain_or_gzip(fq_path_1)
log_f = opj(dir_fq_cor_data_sample, pe + '.txt')
out_f_1 = opj(dir_fq_cor_data_sample, pe + '_R1.fastq' + ext)
out_f_2 = opj(dir_fq_cor_data_sample, pe + '_R2.fastq' + ext)
pe_fastq_files[pe]['cor_path_fq'] = [out_f_1, out_f_2]
if len(pe_fastq_files[pe]['path']) == 3:
fq_path_3 = pe_fastq_files[pe]['path'][2]
out_f_3 = opj(dir_fq_cor_data_sample, pe + '_R3.fastq' + ext)
pe_fastq_files[pe]['cor_path_fq'].append(out_f_3)
if should_run is False:
pe_fastq_files[pe]['cor_path_fq'] = [fq_path_1, fq_path_2]
if fq_path_3 is not None:
pe_fastq_files[pe]['cor_path_fq'].append(fq_path_3)
continue
if ope(dir_fq_cor_data_sample):
Log.msg('Corrected FASTQ files already exist:', pe)
else:
make_dirs(dir_fq_cor_data_sample)
Log.msg('PE mode:', pe)
run_rcorrector_pe(rcorrector=rcorrector,
in_file_1=fq_path_1,
in_file_2=fq_path_2,
out_dir=dir_fq_cor_data_sample,
threads=threads,
dir_temp=dir_temp)
fq_base_path_1 = opj(dir_fq_cor_data_sample, basename(fq_path_1))
fq_cor_path_1 = splitext_gz(fq_base_path_1)[0] + '.cor.fq' + ext
fq_base_path_2 = opj(dir_fq_cor_data_sample, basename(fq_path_2))
fq_cor_path_2 = splitext_gz(fq_base_path_2)[0] + '.cor.fq' + ext
filter_unc_pe(in_file_1=fq_cor_path_1,
in_file_2=fq_cor_path_2,
out_file_1=out_f_1,
out_file_2=out_f_2,
log_file=log_f)
remove(fq_cor_path_1)
remove(fq_cor_path_2)
if fq_path_3 is not None:
Log.msg(
'SE mode (Paired-read SRA run contains unpaired reads):',
pe)
run_rcorrector_se(rcorrector=rcorrector,
in_file=fq_path_3,
out_dir=dir_fq_cor_data_sample,
threads=threads,
dir_temp=dir_temp)
fq_base_path_3 = opj(dir_fq_cor_data_sample,
basename(fq_path_3))
fq_cor_path_3 = splitext_gz(fq_base_path_3)[0] + '.cor.fq'
log_f_3 = opj(dir_fq_cor_data_sample, pe + '_unpaired.txt')
filter_unc_se(in_file=fq_cor_path_3,
out_file=out_f_3,
log_file=log_f_3)
remove(fq_cor_path_3)
def dnld_sra_fastq_files(sras, sra_runs_info, dir_fq_data, fasterq_dump,
threads, dir_temp):
if len(sras) > 0:
if fasterq_dump is None:
Log.err('fasterq-dump from SRA Toolkit is not available. ' +
'Cannot continue. Exiting.')
exit(0)
print()
Log.inf('Downloading SRA read data.')
se_fastq_files = {}
pe_fastq_files = {}
for sra in sras:
sra_run_info = sra_runs_info[sra]
sra_lib_layout = sra_run_info['LibraryLayout'].lower()
sra_lib_layout_k = sra_run_info['KakapoLibraryLayout'].lower()
sample_base_name = sra_run_info['KakapoSampleBaseName']
sra_taxid = int(sra_run_info['TaxID'])
avg_len = int(sra_run_info['avgLength'])
sra_dnld_needed = False
if sra_lib_layout == 'single' or sra_lib_layout_k == 'single':
se_file = opj(dir_fq_data, sra + '.fastq')
se_fastq_files[sample_base_name] = {'path': se_file}
se_fastq_files[sample_base_name]['src'] = 'sra'
se_fastq_files[sample_base_name]['avg_len'] = avg_len
se_fastq_files[sample_base_name]['tax_id'] = sra_taxid
if not ope(se_file):
sra_dnld_needed = True
elif sra_lib_layout == 'paired':
pe_file_1 = opj(dir_fq_data, sra + '_1.fastq')
pe_file_2 = opj(dir_fq_data, sra + '_2.fastq')
pe_file_1_renamed = opj(dir_fq_data, sra + '_R1.fastq')
pe_file_2_renamed = opj(dir_fq_data, sra + '_R2.fastq')
pe_fastq_files[sample_base_name] = {
'path': [pe_file_1_renamed, pe_file_2_renamed]
}
pe_fastq_files[sample_base_name]['src'] = 'sra'
pe_fastq_files[sample_base_name]['avg_len'] = avg_len // 2
pe_fastq_files[sample_base_name]['tax_id'] = sra_taxid
if sra_lib_layout_k == 'paired_unp':
pe_file_3 = opj(dir_fq_data, sra + '.fastq')
pe_file_3_renamed = opj(dir_fq_data, sra + '_R3.fastq')
pe_fastq_files[sample_base_name]['path'].append(
pe_file_3_renamed)
if not ope(pe_file_1_renamed) or not ope(pe_file_2_renamed):
sra_dnld_needed = True
if not sra_dnld_needed:
Log.msg('FASTQ reads are available locally:', sample_base_name)
retry_count = 0
while sra_dnld_needed:
if retry_count > 50:
Log.err('Download failed. Exiting.')
rmtree(dir_temp)
exit(1)
elif retry_count > 0:
Log.wrn('Download failed. Retrying.')
sleep(2)
retry_count += 1
Log.msg('Downloading FASTQ reads for:', sample_base_name)
cmd = [
fasterq_dump, '--threads',
str(threads * 2), '--split-3', '--bufsize', '819200',
'--outdir', dir_fq_data, '--temp', dir_temp, sra
]
run(cmd, do_not_raise=True)
if sra_lib_layout == 'single' or sra_lib_layout_k == 'single':
if not ope(se_file):
continue
elif sra_lib_layout == 'paired':
if not ope(pe_file_1) or not ope(pe_file_2):
continue
else:
move(pe_file_1, pe_file_1_renamed)
move(pe_file_2, pe_file_2_renamed)
if sra_lib_layout_k == 'paired_unp':
if not ope(pe_file_3):
continue
else:
move(pe_file_3, pe_file_3_renamed)
sra_dnld_needed = False
if sra_lib_layout == 'single' or sra_lib_layout_k == 'single':
if ope(se_file):
Log.msg('Renaming FASTQ reads in:', se_file)
rename_fq_seqs(se_file, sra, '1:N:0')
elif sra_lib_layout == 'paired':
if ope(pe_file_1_renamed):
Log.msg('Renaming FASTQ reads in:', pe_file_1_renamed)
rename_fq_seqs(pe_file_1_renamed, sra, '1:N:0')
if ope(pe_file_2_renamed):
Log.msg('Renaming FASTQ reads in:', pe_file_2_renamed)
rename_fq_seqs(pe_file_2_renamed, sra, '2:N:0')
if sra_lib_layout_k == 'paired_unp':
if ope(pe_file_3_renamed):
Log.msg('Renaming FASTQ reads in:', pe_file_3_renamed)
rename_fq_seqs(pe_file_3_renamed, sra + '_unpaired',
'1:N:0')
return se_fastq_files, pe_fastq_files, sra_runs_info
def dep_check_kraken2(dir_dep, os_id, release_name, force):
url = 'https://github.com/karolisr/kraken2/archive/master.tar.gz'
dnld_path = opj(dir_dep, 'kraken2.tar.gz')
try:
if force is True:
raise
kraken2 = which('kraken2')
kraken2_build = which('kraken2-build')
dir_bin = dirname(kraken2)
classify_bin = opj(dir_bin, 'classify')
_ = run([classify_bin], do_not_raise=True)
if not _.stderr.startswith('classify: mandatory filename'):
raise
run([kraken2, '--help'])
run([kraken2_build, '--help'])
except Exception:
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'kraken2'))
kraken2 = opj(dir_bin, 'bin', 'kraken2')
kraken2_build = opj(dir_bin, 'bin', 'kraken2-build')
classify_bin = opj(dir_bin, 'bin', 'classify')
_ = run([classify_bin], do_not_raise=True)
if not _.stderr.startswith('classify: mandatory filename'):
raise
run([kraken2, '--help'])
run([kraken2_build, '--help'])
except Exception:
Log.wrn('Kraken2 was not found on this system, trying to '
'download.')
if ope(dnld_path):
remove(dnld_path)
download_file(url, dnld_path)
tar_ref = tarfile.open(dnld_path, 'r:gz')
tar_ref.extractall(dir_dep)
tar_ref.close()
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'kraken2'))
classify_bin = opj(dir_bin, 'bin', 'classify')
kraken2 = opj(dir_bin, 'bin', 'kraken2')
kraken2_build = opj(dir_bin, 'bin', 'kraken2-build')
makefile = opj(dir_bin, 'src', 'Makefile')
replace_line_in_file(makefile, 'cp $(PROGS) $(KRAKEN2_DIR)/',
'cp $(PROGS) "$(KRAKEN2_DIR)"/')
try:
Log.wrn('Compiling Kraken2 Attempt 1')
run(['./install_kraken2.sh', 'bin'], cwd=dir_bin)
_ = run([classify_bin], do_not_raise=True)
if not _.stderr.startswith('classify: mandatory filename'):
raise
run([kraken2, '--help'])
run([kraken2_build, '--help'])
except Exception:
try:
Log.wrn('Compiling Kraken2 Attempt 2')
dir_libomp = opj(dir_dep, 'libomp')
if ope(dir_libomp):
rmtree(dir_libomp)
libomp_fp, v = brew_get('libomp', os_id, release_name,
dir_dep)
tar_ref = tarfile.open(libomp_fp, 'r:gz')
tar_ref.extractall(dir_dep)
tar_ref.close()
dir_libomp_l = opj(dir_libomp, v, 'lib')
dir_libomp_i = opj(dir_libomp, v, 'include')
if os_id == 'mac':
# Changes the shared library identification name of a
# dynamic shared library.
dylib_f = opj(dir_libomp_l, 'libomp.dylib')
chmod(
dylib_f, stat.S_IRWXU | stat.S_IRUSR | stat.S_IWUSR
| stat.S_IRGRP | stat.S_IWGRP | stat.S_IROTH
| stat.S_IWOTH)
cmd = ['install_name_tool', '-id', dylib_f, dylib_f]
run(cmd)
cxx_flags = ('CXXFLAGS = -L{} -I{} -Xpreprocessor '
'-fopenmp -lomp -Wall -std=c++11 -O3')
elif os_id == 'linux':
cxx_flags = ('CXXFLAGS = -L{} -I{} -fopenmp -lomp '
'-static -Wall -std=c++11 -O3')
cxx_flags = cxx_flags.format(dir_libomp_l, dir_libomp_i)
makefile = opj(dir_bin, 'src', 'Makefile')
replace_line_in_file(
makefile, 'CXXFLAGS = -fopenmp -Wall -std=c++11'
' -O3', cxx_flags)
run(['./install_kraken2.sh', 'bin'], cwd=dir_bin)
_ = run([classify_bin], do_not_raise=True)
if not _.stderr.startswith('classify: mandatory filename'):
raise
run([kraken2, '--help'])
run([kraken2_build, '--help'])
except Exception:
try:
Log.wrn('Compiling Kraken2 Attempt 3')
makefile = opj(dir_bin, 'src', 'Makefile')
replace_line_in_file(
makefile, cxx_flags,
'CXXFLAGS = -Wall -std=c++11 -O3')
run(['./install_kraken2.sh', 'bin'], cwd=dir_bin)
_ = run([classify_bin], do_not_raise=True)
if not _.stderr.startswith(
'classify: mandatory filename'):
raise
run([kraken2, '--help'])
run([kraken2_build, '--help'])
except Exception:
pass
if not ope(kraken2):
Log.err('Something went wrong while trying to compile '
'Kraken2.')
Log.msg('Try downloading and installing it manually from: '
'https://github.com/karolisr/kraken2')
return None, None
regexp = r'^.*?version\s([\d\.\-A-Za-z]*)'
v = get_dep_version([kraken2, '--version'], regexp)
Log.msg('kraken2 is available:', v + ' ' + kraken2)
v = get_dep_version([kraken2_build, '--version'], regexp)
Log.msg('kraken2-build is available:', v + ' ' + kraken2_build)
return kraken2, kraken2_build
def run_rcorrector(se_fastq_files, pe_fastq_files, dir_fq_cor_data, rcorrector,
threads, dir_temp, fpatt, should_run):
if len(se_fastq_files) > 0 or len(pe_fastq_files) > 0:
print()
if should_run is False:
Log.wrn('Skipping Rcorrector as requested.')
else:
Log.inf('Running Rcorrector.')
if rcorrector is None:
Log.err(
'Rcorrector is not available. Cannot continue. Exiting.')
exit(0)
for se in se_fastq_files:
dir_fq_cor_data_sample = opj(dir_fq_cor_data, se)
fq_path = se_fastq_files[se]['trim_path_fq']
r_mode, w_mode, a_mode, fqopen, ext = plain_or_gzip(fq_path)
log_f = opj(dir_fq_cor_data_sample, se + '.txt')
out_f = opj(dir_fq_cor_data_sample, se + '.fastq' + ext)
se_fastq_files[se]['cor_path_fq'] = out_f
if should_run is False:
se_fastq_files[se]['cor_path_fq'] = fq_path
continue
if ope(dir_fq_cor_data_sample):
Log.msg('Corrected FASTQ file already exists:', se)
else:
make_dirs(dir_fq_cor_data_sample)
Log.msg('SE mode:', se)
run_rcorrector_se(rcorrector=rcorrector,
in_file=fq_path,
out_dir=dir_fq_cor_data_sample,
threads=threads,
dir_temp=dir_temp)
fq_base_path = opj(dir_fq_cor_data_sample, basename(fq_path))
fq_cor_path = splitext_gz(fq_base_path)[0] + '.cor.fq' + ext
filter_unc_se(in_file=fq_cor_path, out_file=out_f, log_file=log_f)
remove(fq_cor_path)
for pe in pe_fastq_files:
dir_fq_cor_data_sample = opj(dir_fq_cor_data, pe)
fq_path_1 = pe_fastq_files[pe]['trim_path_fq'][0]
fq_path_2 = pe_fastq_files[pe]['trim_path_fq'][1]
fq_path_3 = pe_fastq_files[pe]['trim_path_fq'][2]
fq_path_4 = pe_fastq_files[pe]['trim_path_fq'][3]
r_mode, w_mode, a_mode, fqopen, ext = plain_or_gzip(fq_path_1)
log_f = opj(dir_fq_cor_data_sample, pe + '_paired.txt')
out_fs = [x.replace('@D@', dir_fq_cor_data_sample) for x in fpatt]
out_fs = [x.replace('@N@', pe) for x in out_fs]
out_fs = [x + ext for x in out_fs]
pe_fastq_files[pe]['cor_path_fq'] = out_fs
if should_run is False:
pe_fastq_files[pe]['cor_path_fq'] = [
fq_path_1, fq_path_2, fq_path_3, fq_path_4
]
continue
if ope(dir_fq_cor_data_sample):
Log.msg('Corrected FASTQ files already exist:', pe)
else:
make_dirs(dir_fq_cor_data_sample)
Log.msg('PE mode:', pe)
run_rcorrector_pe(rcorrector=rcorrector,
in_file_1=fq_path_1,
in_file_2=fq_path_2,
out_dir=dir_fq_cor_data_sample,
threads=threads,
dir_temp=dir_temp)
fq_base_path_1 = opj(dir_fq_cor_data_sample, basename(fq_path_1))
fq_cor_path_1 = splitext_gz(fq_base_path_1)[0] + '.cor.fq' + ext
fq_base_path_2 = opj(dir_fq_cor_data_sample, basename(fq_path_2))
fq_cor_path_2 = splitext_gz(fq_base_path_2)[0] + '.cor.fq' + ext
filter_unc_pe(in_file_1=fq_cor_path_1,
in_file_2=fq_cor_path_2,
out_file_1=out_fs[0],
out_file_2=out_fs[1],
log_file=log_f)
remove(fq_cor_path_1)
remove(fq_cor_path_2)
# unpaired 1
if stat(fq_path_3).st_size != 0:
run_rcorrector_se(rcorrector=rcorrector,
in_file=fq_path_3,
out_dir=dir_fq_cor_data_sample,
threads=threads,
dir_temp=dir_temp)
fq_base_path_3 = opj(dir_fq_cor_data_sample,
basename(fq_path_3))
fq_cor_path_3 = splitext_gz(
fq_base_path_3)[0] + '.cor.fq' + ext
log_f_3 = opj(dir_fq_cor_data_sample, pe + '_unpaired_1.txt')
filter_unc_se(in_file=fq_cor_path_3,
out_file=out_fs[2],
log_file=log_f_3)
remove(fq_cor_path_3)
else:
with open(out_fs[2], 'w') as f:
f.write('')
# unpaired 2
if stat(fq_path_4).st_size != 0:
run_rcorrector_se(rcorrector=rcorrector,
in_file=fq_path_4,
out_dir=dir_fq_cor_data_sample,
threads=threads,
dir_temp=dir_temp)
fq_base_path_4 = opj(dir_fq_cor_data_sample,
basename(fq_path_4))
fq_cor_path_4 = splitext_gz(
fq_base_path_4)[0] + '.cor.fq' + ext
log_f_4 = opj(dir_fq_cor_data_sample, pe + '_unpaired_2.txt')
filter_unc_se(in_file=fq_cor_path_4,
out_file=out_fs[3],
log_file=log_f_4)
remove(fq_cor_path_4)
else:
with open(out_fs[3], 'w') as f:
f.write('')
def main():
"""Run the script."""
# Prepare initial logger (before we know the log file path) --------------
prj_log_file_suffix = time_stamp() + '.log'
log_stream = StringIO()
Log.set_colors(COLORS)
Log.set_file(log_stream)
Log.set_write(True)
# Prepare configuration directory ----------------------------------------
if ope(DIR_CFG):
Log.inf('Found configuration directory:', DIR_CFG)
else:
Log.wrn('Creating configuration directory:', DIR_CFG)
make_dirs(DIR_CFG)
print()
# Check for dependencies -------------------------------------------------
Log.inf('Checking for dependencies.')
make_dirs(DIR_DEP)
make_dirs(DIR_KRK)
seqtk = deps.dep_check_seqtk(DIR_DEP, FORCE_DEPS)
trimmomatic, adapters = deps.dep_check_trimmomatic(DIR_DEP)
fasterq_dump = deps.dep_check_sra_toolkit(DIR_DEP, OS_ID, DIST_ID,
DEBIAN_DISTS, REDHAT_DISTS,
FORCE_DEPS)
makeblastdb, _, tblastn = deps.dep_check_blast(DIR_DEP, OS_ID, DIST_ID,
DEBIAN_DISTS, REDHAT_DISTS,
FORCE_DEPS)
vsearch = deps.dep_check_vsearch(DIR_DEP, OS_ID, DIST_ID, DEBIAN_DISTS,
REDHAT_DISTS, FORCE_DEPS)
spades = deps.dep_check_spades(DIR_DEP, OS_ID, FORCE_DEPS)
bowtie2, bowtie2_build = deps.dep_check_bowtie2(DIR_DEP, OS_ID, FORCE_DEPS)
rcorrector = deps.dep_check_rcorrector(DIR_DEP, FORCE_DEPS)
kraken2, kraken2_build = deps.dep_check_kraken2(DIR_DEP, OS_ID,
RELEASE_NAME, FORCE_DEPS)
print()
kraken2_dbs = deps.dnld_kraken2_dbs(DIR_KRK)
if INSTALL_DEPS is True or DNLD_KRAKEN_DBS is True:
exit(0)
print()
# Initialize NCBI taxonomy database --------------------------------------
tax = Taxonomy()
if tax.is_initialized() is False:
tax.init(data_dir_path=DIR_TAX, logger=Log)
print()
# Parse configuration file -----------------------------------------------
Log.inf('Reading configuration file:', CONFIG_FILE_PATH)
_ = config_file_parse(CONFIG_FILE_PATH, tax)
allow_no_stop_cod = _['allow_no_stop_cod']
allow_no_strt_cod = _['allow_no_strt_cod']
allow_non_aug = _['allow_non_aug']
blast_1_evalue = _['blast_1_evalue']
blast_1_max_hsps = _['blast_1_max_hsps']
blast_1_qcov_hsp_perc = _['blast_1_qcov_hsp_perc']
blast_1_best_hit_overhang = _['blast_1_best_hit_overhang']
blast_1_best_hit_score_edge = _['blast_1_best_hit_score_edge']
blast_1_max_target_seqs = _['blast_1_max_target_seqs']
blast_2_evalue = _['blast_2_evalue']
blast_2_max_hsps = _['blast_2_max_hsps']
blast_2_qcov_hsp_perc = _['blast_2_qcov_hsp_perc']
blast_2_best_hit_overhang = _['blast_2_best_hit_overhang']
blast_2_best_hit_score_edge = _['blast_2_best_hit_score_edge']
blast_2_max_target_seqs = _['blast_2_max_target_seqs']
dir_out = _['output_directory']
email = _['email']
requery_after = _['requery_after']
fq_pe = _['fq_pe']
fq_se = _['fq_se']
should_run_rcorrector = _['should_run_rcorrector']
should_run_ipr = _['should_run_ipr']
bt2_order = _['bt2_order']
kraken_confidence = _['kraken_confidence']
krkn_order = _['krkn_order']
prepend_assmbl = _['prepend_assmbl']
prj_name = _['project_name']
sras = _['sras']
tax_group = _['tax_group']
# tax_group_name = _['tax_group_name']
tax_ids_user = _['tax_ids']
user_assemblies = _['assmbl']
print()
# Parse search strategies file -------------------------------------------
if SS_FILE_PATH is not None:
Log.inf('Reading search strategies file:', SS_FILE_PATH)
sss = ss_file_parse(SS_FILE_PATH)
else:
Log.wrn('Search strategies file was not provided.\n' +
'Will process reads, assemblies and then stop.')
sss = dict()
print()
# Create output directory ------------------------------------------------
if dir_out is not None:
if ope(dir_out):
Log.inf('Found output directory:', dir_out)
else:
Log.wrn('Creating output directory:', dir_out)
make_dirs(dir_out)
print()
# Write Kakapo version information to the output directory ---------------
version_file = opj(dir_out, 'kakapo_version.txt')
if ope(version_file):
with open(version_file, 'r') as f:
version_prev = f.read().strip()
if __version__ != version_prev:
Log.wrn('The output directory contains data produced by a ' +
'different version of Kakapo: ' + version_prev +
'.\nThe currently running version is: ' + __version__ +
'.\n' +
'Delete "kakapo_version.txt" file located in the ' +
'output directory if you would like to continue.')
exit(0)
with open(version_file, 'w') as f:
f.write(__version__)
# Create subdirectories in the output directory --------------------------
_ = prepare_output_directories(dir_out, prj_name)
dir_temp = _['dir_temp']
dir_cache_pfam_acc = _['dir_cache_pfam_acc']
dir_cache_fq_minlen = _['dir_cache_fq_minlen']
dir_cache_prj = _['dir_cache_prj']
dir_cache_refseqs = _['dir_cache_refseqs']
dir_prj_logs = _['dir_prj_logs']
dir_prj_queries = _['dir_prj_queries']
dir_fq_data = _['dir_fq_data']
dir_fq_cor_data = _['dir_fq_cor_data']
dir_fq_trim_data = _['dir_fq_trim_data']
dir_fq_filter_bt2_data = _['dir_fq_filter_bt2_data']
dir_fq_filter_krkn2_data = _['dir_fq_filter_krkn2_data']
dir_fa_trim_data = _['dir_fa_trim_data']
dir_blast_fa_trim = _['dir_blast_fa_trim']
dir_prj_blast_results_fa_trim = _['dir_prj_blast_results_fa_trim']
dir_prj_vsearch_results_fa_trim = _['dir_prj_vsearch_results_fa_trim']
dir_prj_spades_assemblies = _['dir_prj_spades_assemblies']
dir_prj_blast_assmbl = _['dir_prj_blast_assmbl']
dir_prj_assmbl_blast_results = _['dir_prj_assmbl_blast_results']
dir_prj_transcripts = _['dir_prj_transcripts']
dir_prj_ips = _['dir_prj_ips']
dir_prj_transcripts_combined = _['dir_prj_transcripts_combined']
# Prepare logger ---------------------------------------------------------
prj_log_file = opj(dir_prj_logs, prj_name + '_' + prj_log_file_suffix)
with open(prj_log_file, 'w') as f:
f.write(SCRIPT_INFO.strip() + '\n\n' + log_stream.getvalue())
Log.set_colors(COLORS)
Log.set_file(prj_log_file)
Log.set_write(True)
log_stream.close()
# Resolve descending taxonomy nodes --------------------------------------
tax_ids = tax.all_descending_taxids_for_taxids([tax_group])
# Pfam uniprot accessions ------------------------------------------------
pfam_uniprot_acc = OrderedDict()
for ss in sss:
pfam_acc = sss[ss]['pfam_families']
pfam_uniprot_acc[ss] = pfam_uniprot_accessions(ss, pfam_acc, tax_ids,
dir_cache_pfam_acc)
# Download Pfam uniprot sequences if needed ------------------------------
aa_uniprot_files = OrderedDict()
for ss in sss:
aa_uniprot_files[ss] = opj(dir_prj_queries,
'aa_uniprot__' + ss + '.fasta')
# ToDo: add support for the requery_after parameter.
dnld_pfam_uniprot_seqs(ss, pfam_uniprot_acc[ss], aa_uniprot_files[ss],
dir_cache_prj)
# User provided entrez query ---------------------------------------------
prot_acc_user_from_query = OrderedDict()
for ss in sss:
entrez_queries = sss[ss]['entrez_search_queries']
prot_acc_user_from_query[ss] = user_entrez_search(
ss, entrez_queries, dir_cache_prj, requery_after)
# User provided protein accessions ---------------------------------------
prot_acc_user = OrderedDict()
for ss in sss:
print()
prot_acc_all = sorted(
set(sss[ss]['ncbi_accessions_aa'] + prot_acc_user_from_query[ss]))
prot_acc_user[ss] = user_protein_accessions(ss, prot_acc_all,
dir_cache_prj, tax)
# Download from NCBI if needed -------------------------------------------
aa_prot_ncbi_files = OrderedDict()
for ss in sss:
aa_prot_ncbi_files[ss] = opj(dir_prj_queries,
'aa_prot_ncbi__' + ss + '.fasta')
prot_acc_user[ss] = dnld_prot_seqs(ss, prot_acc_user[ss],
aa_prot_ncbi_files[ss],
dir_cache_prj)
# User provided protein sequences ----------------------------------------
aa_prot_user_files = OrderedDict()
for ss in sss:
user_queries = sss[ss]['fasta_files_aa']
aa_prot_user_files[ss] = opj(dir_prj_queries,
'aa_prot_user__' + ss + '.fasta')
user_aa_fasta(ss, user_queries, aa_prot_user_files[ss])
# Combine all AA queries -------------------------------------------------
print()
aa_queries_files = OrderedDict()
for ss in sss:
aa_queries_files[ss] = opj(dir_prj_queries, 'aa_all__' + ss + '.fasta')
combine_aa_fasta(ss, [
aa_uniprot_files[ss], aa_prot_ncbi_files[ss],
aa_prot_user_files[ss]
], aa_queries_files[ss])
# Filter AA queries ------------------------------------------------------
prot_acc_user_filtered = OrderedDict()
for ss in sss:
min_query_length = sss[ss]['min_query_length']
max_query_length = sss[ss]['max_query_length']
max_query_identity = sss[ss]['max_query_identity']
# Dereplicate all queries
filter_queries(ss,
aa_queries_files[ss],
min_query_length,
max_query_length,
max_query_identity,
vsearch,
prot_acc_user[ss],
overwrite=True)
# Dereplicate only NCBI queries. CDS for these will be downloaded
# later for reference.
if ope(aa_prot_ncbi_files[ss]):
prot_acc_user_filtered[ss] = filter_queries(ss,
aa_prot_ncbi_files[ss],
min_query_length,
max_query_length,
max_query_identity,
vsearch,
prot_acc_user[ss],
overwrite=False,
logging=False)
# Download SRA run metadata if needed ------------------------------------
sra_runs_info, sras_acceptable = dnld_sra_info(sras, dir_cache_prj)
# Download SRA run FASTQ files if needed ---------------------------------
x, y, z = dnld_sra_fastq_files(sras_acceptable, sra_runs_info, dir_fq_data,
fasterq_dump, THREADS, dir_temp)
se_fastq_files_sra = x
pe_fastq_files_sra = y
sra_runs_info = z
# User provided FASTQ files ----------------------------------------------
se_fastq_files_usr, pe_fastq_files_usr = user_fastq_files(fq_se, fq_pe)
# Collate FASTQ file info ------------------------------------------------
se_fastq_files = se_fastq_files_sra.copy()
se_fastq_files.update(se_fastq_files_usr)
pe_fastq_files = pe_fastq_files_sra.copy()
pe_fastq_files.update(pe_fastq_files_usr)
def gc_tt(k, d, tax):
taxid = d[k]['tax_id']
gc = tax.genetic_code_for_taxid(taxid)
d[k]['gc_id'] = gc
d[k]['gc_tt'] = TranslationTable(gc)
gc_mito = None
tt_mito = None
gc_plastid = None
tt_plastid = None
if tax.is_eukaryote(taxid) is True:
gc_mito = tax.mito_genetic_code_for_taxid(taxid)
if gc_mito != '0':
tt_mito = TranslationTable(gc_mito)
if tax.contains_plastid(taxid) is True:
gc_plastid = tax.plastid_genetic_code_for_taxid(taxid)
if gc_plastid != '0':
tt_plastid = TranslationTable(gc_plastid)
d[k]['gc_id_mito'] = gc_mito
d[k]['gc_tt_mito'] = tt_mito
d[k]['gc_id_plastid'] = gc_plastid
d[k]['gc_tt_plastid'] = tt_plastid
for se in se_fastq_files:
gc_tt(se, se_fastq_files, tax)
for pe in pe_fastq_files:
gc_tt(pe, pe_fastq_files, tax)
# Minimum acceptable read length -----------------------------------------
min_accept_read_len(se_fastq_files, pe_fastq_files, dir_temp,
dir_cache_fq_minlen, vsearch)
# Run Rcorrector ---------------------------------------------------------
run_rcorrector(se_fastq_files, pe_fastq_files, dir_fq_cor_data, rcorrector,
THREADS, dir_temp, should_run_rcorrector)
# File name patterns -----------------------------------------------------
a, b, c, d, e = file_name_patterns()
pe_trim_fq_file_patterns = a
pe_trim_fa_file_patterns = b
pe_blast_db_file_patterns = c
pe_blast_results_file_patterns = d
pe_vsearch_results_file_patterns = e
# Run Trimmomatic --------------------------------------------------------
run_trimmomatic(se_fastq_files, pe_fastq_files, dir_fq_trim_data,
trimmomatic, adapters, pe_trim_fq_file_patterns, THREADS)
# Run Bowtie 2 -----------------------------------------------------------
run_bt2_fq(se_fastq_files, pe_fastq_files, dir_fq_filter_bt2_data, bowtie2,
bowtie2_build, THREADS, dir_temp, bt2_order,
pe_trim_fq_file_patterns, tax, dir_cache_refseqs)
# Run Kraken2 ------------------------------------------------------------
run_kraken2(krkn_order, kraken2_dbs, se_fastq_files, pe_fastq_files,
dir_fq_filter_krkn2_data, kraken_confidence, kraken2, THREADS,
dir_temp, pe_trim_fq_file_patterns)
se_fastq_files = OrderedDict(se_fastq_files)
pe_fastq_files = OrderedDict(pe_fastq_files)
se_fastq_files = OrderedDict(
sorted(se_fastq_files.items(), key=lambda x: x[1]['filter_path_fq']))
pe_fastq_files = OrderedDict(
sorted(pe_fastq_files.items(), key=lambda x: x[1]['filter_path_fq']))
# Stop After Filter ------------------------------------------------------
if STOP_AFTER_FILTER is True:
Log.wrn('Stopping after Kraken2/Bowtie2 filtering step as requested.')
exit(0)
# Convert filtered FASTQ files to FASTA ----------------------------------
filtered_fq_to_fa(se_fastq_files, pe_fastq_files, dir_fa_trim_data, seqtk,
pe_trim_fa_file_patterns)
# Run makeblastdb on reads -----------------------------------------------
makeblastdb_fq(se_fastq_files, pe_fastq_files, dir_blast_fa_trim,
makeblastdb, pe_blast_db_file_patterns)
# Check if there are any query sequences.
any_queries = False
for ss in sss:
if stat(aa_queries_files[ss]).st_size == 0:
continue
else:
any_queries = True
# Run tblastn on reads ---------------------------------------------------
for ss in sss:
if stat(aa_queries_files[ss]).st_size == 0:
continue
changed_blast_1 = run_tblastn_on_reads(
se_fastq_files, pe_fastq_files, aa_queries_files[ss], tblastn,
blast_1_evalue, blast_1_max_hsps, blast_1_qcov_hsp_perc,
blast_1_best_hit_overhang, blast_1_best_hit_score_edge,
blast_1_max_target_seqs, dir_prj_blast_results_fa_trim,
pe_blast_results_file_patterns, ss, THREADS, seqtk, vsearch,
dir_cache_prj)
if changed_blast_1 is True:
if ope(dir_prj_vsearch_results_fa_trim):
rmtree(dir_prj_vsearch_results_fa_trim)
if ope(dir_prj_spades_assemblies):
rmtree(dir_prj_spades_assemblies)
if ope(dir_prj_blast_assmbl):
rmtree(dir_prj_blast_assmbl)
if ope(dir_prj_assmbl_blast_results):
rmtree(dir_prj_assmbl_blast_results)
if ope(dir_prj_transcripts):
rmtree(dir_prj_transcripts)
if ope(dir_prj_transcripts_combined):
rmtree(dir_prj_transcripts_combined)
prepare_output_directories(dir_out, prj_name)
# Run vsearch on reads ---------------------------------------------------
# should_run_vsearch = False
# for ss in sss:
# if stat(aa_queries_files[ss]).st_size == 0:
# continue
# else:
# should_run_vsearch = True
# break
# if should_run_vsearch is True:
# print()
# Log.inf('Checking if Vsearch should be run.')
for ss in sss:
if stat(aa_queries_files[ss]).st_size == 0:
continue
print()
Log.inf('Checking if Vsearch should be run:', ss)
run_vsearch_on_reads(se_fastq_files, pe_fastq_files, vsearch,
dir_prj_vsearch_results_fa_trim,
pe_vsearch_results_file_patterns, ss, seqtk)
# Run SPAdes -------------------------------------------------------------
# should_run_spades = False
# for ss in sss:
# if stat(aa_queries_files[ss]).st_size == 0:
# continue
# else:
# should_run_spades = True
# break
# if should_run_spades is True:
# print()
# Log.inf('Checking if SPAdes should be run.')
for ss in sss:
if stat(aa_queries_files[ss]).st_size == 0:
for se in se_fastq_files:
se_fastq_files[se]['spades_assembly' + '__' + ss] = None
for pe in pe_fastq_files:
pe_fastq_files[pe]['spades_assembly' + '__' + ss] = None
continue
print()
Log.inf('Checking if SPAdes should be run:', ss)
run_spades(se_fastq_files, pe_fastq_files, dir_prj_spades_assemblies,
spades, dir_temp, ss, THREADS, RAM)
# Combine SPAdes and user provided assemblies ----------------------------
assemblies = combine_assemblies(se_fastq_files, pe_fastq_files,
user_assemblies, tax, sss)
# Run makeblastdb on assemblies -----------------------------------------
makeblastdb_assemblies(assemblies, dir_prj_blast_assmbl, makeblastdb)
if any_queries is False:
Log.wrn('No query sequences were provided.')
# Run tblastn on assemblies ----------------------------------------------
for ss in sss:
if stat(aa_queries_files[ss]).st_size == 0:
continue
should_run_tblastn = False
for a in assemblies:
assmbl_src = a['src']
assmbl_name = a['name']
if assmbl_src != 'user_fasta':
if assmbl_name.endswith('__' + ss):
should_run_tblastn = True
break
else:
should_run_tblastn = True
break
if should_run_tblastn is False:
print()
Log.inf('Will not run BLAST. No transcripts exist:', ss)
continue
blast_2_evalue_ss = sss[ss]['blast_2_evalue']
blast_2_max_hsps_ss = sss[ss]['blast_2_max_hsps']
blast_2_qcov_hsp_perc_ss = sss[ss]['blast_2_qcov_hsp_perc']
blast_2_best_hit_overhang_ss = sss[ss]['blast_2_best_hit_overhang']
blast_2_best_hit_score_edge_ss = sss[ss]['blast_2_best_hit_score_edge']
blast_2_max_target_seqs_ss = sss[ss]['blast_2_max_target_seqs']
if blast_2_evalue_ss is None:
blast_2_evalue_ss = blast_2_evalue
if blast_2_max_hsps_ss is None:
blast_2_max_hsps_ss = blast_2_max_hsps
if blast_2_qcov_hsp_perc_ss is None:
blast_2_qcov_hsp_perc_ss = blast_2_qcov_hsp_perc
if blast_2_best_hit_overhang_ss is None:
blast_2_best_hit_overhang_ss = blast_2_best_hit_overhang
if blast_2_best_hit_score_edge_ss is None:
blast_2_best_hit_score_edge_ss = blast_2_best_hit_score_edge
if blast_2_max_target_seqs_ss is None:
blast_2_max_target_seqs_ss = blast_2_max_target_seqs
run_tblastn_on_assemblies(
ss, assemblies, aa_queries_files[ss], tblastn,
dir_prj_assmbl_blast_results, blast_2_evalue_ss,
blast_2_max_hsps_ss, blast_2_qcov_hsp_perc_ss,
blast_2_best_hit_overhang_ss, blast_2_best_hit_score_edge_ss,
blast_2_max_target_seqs_ss, THREADS, dir_cache_prj, dir_prj_ips)
# Prepare BLAST hits for analysis: find ORFs, translate ------------------
for ss in sss:
if stat(aa_queries_files[ss]).st_size == 0:
continue
min_target_orf_len_ss = sss[ss]['min_target_orf_length']
max_target_orf_len_ss = sss[ss]['max_target_orf_length']
organelle = sss[ss]['organelle']
blast_2_qcov_hsp_perc_ss = sss[ss]['blast_2_qcov_hsp_perc']
if blast_2_qcov_hsp_perc_ss is None:
blast_2_qcov_hsp_perc_ss = blast_2_qcov_hsp_perc
find_orfs_translate(ss, assemblies, dir_prj_transcripts, seqtk,
dir_temp, prepend_assmbl, min_target_orf_len_ss,
max_target_orf_len_ss, allow_non_aug,
allow_no_strt_cod, allow_no_stop_cod, tax,
tax_group, tax_ids_user, blast_2_qcov_hsp_perc_ss,
organelle)
# GFF3 files from kakapo results JSON files ------------------------------
# print()
for ss in sss:
if stat(aa_queries_files[ss]).st_size == 0:
continue
gff_from_json(ss, assemblies, dir_prj_ips,
dir_prj_transcripts_combined, prj_name)
# Run InterProScan 5 -----------------------------------------------------
if should_run_ipr is True:
print()
ss_names = tuple(sss.keys())
# Determine the length of printed strings, for better spacing --------
max_title_a_len = 0
max_run_id_len = 0
for a in assemblies:
for ss in ss_names:
if 'transcripts_aa_orf_fasta_file__' + ss not in a:
continue
aa_file = a['transcripts_aa_orf_fasta_file__' + ss]
if aa_file is None:
continue
assmbl_name = a['name']
run_id = ss + '_' + assmbl_name
max_run_id_len = max(len(run_id), max_run_id_len)
seqs = seq_records_to_dict(read_fasta(aa_file, SEQ_TYPE_AA))
# Filter all ORFs except the first one.
for seq_def in tuple(seqs.keys()):
seq_def_prefix = seq_def.split(' ')[0]
if seq_def_prefix.endswith('ORF001'):
max_title_a_len = max(len(seq_def_prefix),
max_title_a_len)
max_title_a_len += 2
max_run_id_len += 2
# --------------------------------------------------------------------
parallel_run_count = min(THREADS, len(ss_names))
def run_inter_pro_scan_parallel(ss):
if stat(aa_queries_files[ss]).st_size == 0:
return
run_inter_pro_scan(ss, assemblies, email, dir_prj_ips,
dir_cache_prj, parallel_run_count,
max_title_a_len, max_run_id_len)
# GFF3 files from kakapo and InterProScan 5 results JSON files
gff_from_json(ss, assemblies, dir_prj_ips,
dir_prj_transcripts_combined, prj_name)
Parallel(n_jobs=parallel_run_count, verbose=0,
require='sharedmem')(delayed(run_inter_pro_scan_parallel)(ss)
for ss in ss_names)
# Download CDS for NCBI protein queries ----------------------------------
print()
prot_cds_ncbi_files = OrderedDict()
def dnld_cds_for_ncbi_prot_acc_parallel(ss):
if stat(aa_queries_files[ss]).st_size == 0:
return
if ss not in prot_acc_user_filtered:
return
prot_cds_ncbi_files[ss] = opj(
dir_prj_transcripts_combined,
prj_name + '_ncbi_query_cds__' + ss + '.fasta')
if len(prot_acc_user_filtered[ss]) > 0:
dnld_cds_for_ncbi_prot_acc(ss, prot_acc_user_filtered[ss],
prot_cds_ncbi_files[ss], tax,
dir_cache_prj)
ss_names = tuple(sss.keys())
Parallel(n_jobs=2, verbose=0, require='sharedmem')(
delayed(dnld_cds_for_ncbi_prot_acc_parallel)(ss) for ss in ss_names)
# ------------------------------------------------------------------------
rmtree(dir_temp)
# ------------------------------------------------------------------------
rerun = input('\nRepeat ([y]/n)? ').lower().strip()
if rerun.startswith('y') or rerun == '':
print()
return False
else:
print('\nExiting...')
return True
def run_tblastn_on_assemblies(ss, assemblies, aa_queries_file, tblastn,
dir_prj_assmbl_blast_results, blast_2_evalue,
blast_2_max_hsps, blast_2_qcov_hsp_perc,
blast_2_best_hit_overhang,
blast_2_best_hit_score_edge,
blast_2_max_target_seqs, threads, dir_cache_prj,
dir_prj_ips):
if len(assemblies) > 0:
print()
Log.inf('Running BLAST on assemblies:', ss)
if tblastn is None:
Log.err('tblastn is not available. Cannot continue. Exiting.')
exit(0)
else:
Log.wrn('There are no assemblies. Nothing to do, stopping.')
exit(0)
cache_file = opj(dir_cache_prj, 'blast_2_settings_cache__' + ss)
pickled = dict()
settings = {'blast_2_evalue': blast_2_evalue,
'blast_2_max_hsps': blast_2_max_hsps,
'blast_2_qcov_hsp_perc': blast_2_qcov_hsp_perc,
'blast_2_best_hit_overhang': blast_2_best_hit_overhang,
'blast_2_best_hit_score_edge': blast_2_best_hit_score_edge,
'blast_2_max_target_seqs': blast_2_max_target_seqs,
'queries': seq_records_to_dict(
read_fasta(aa_queries_file, SEQ_TYPE_AA))}
Log.msg('evalue:', str(blast_2_evalue))
Log.msg('max_hsps:', str(blast_2_max_hsps))
Log.msg('qcov_hsp_perc:', str(blast_2_qcov_hsp_perc))
Log.msg('best_hit_overhang:', str(blast_2_best_hit_overhang))
Log.msg('best_hit_score_edge:', str(blast_2_best_hit_score_edge))
Log.msg('max_target_seqs:', str(blast_2_max_target_seqs))
print()
for a in assemblies:
assmbl_src = a['src']
assmbl_name = a['name']
if assmbl_src != 'user_fasta':
if assmbl_name.endswith('__' + ss):
assmbl_name = assmbl_name.replace('__' + ss, '')
else:
continue
assmbl_blast_db_path = a['blast_db_path']
assmbl_genetic_code = a['gc_id']
ips_json_dump_path = opj(dir_prj_ips, assmbl_name + '_ann_ips__' + ss +
'.json')
_ = opj(dir_prj_assmbl_blast_results, assmbl_name + '__' + ss + '.tsv')
if ope(_) and ope(cache_file):
with open(cache_file, 'rb') as f:
pickled = pickle.load(f)
if ope(_) and pickled == settings:
# Log.msg('The provided BLAST settings and query sequences did '
# 'not change since the previous run.')
Log.msg('BLAST results already exist:', assmbl_name)
else:
Log.msg('Running tblastn on: ' + assmbl_name, ss)
if ope(ips_json_dump_path):
osremove(ips_json_dump_path)
run_blast(exec_file=tblastn,
task='tblastn',
threads=threads,
db_path=assmbl_blast_db_path,
queries_file=aa_queries_file,
out_file=_,
evalue=blast_2_evalue,
max_hsps=blast_2_max_hsps,
qcov_hsp_perc=blast_2_qcov_hsp_perc,
best_hit_overhang=blast_2_best_hit_overhang,
best_hit_score_edge=blast_2_best_hit_score_edge,
max_target_seqs=blast_2_max_target_seqs,
db_genetic_code=assmbl_genetic_code,
out_cols=BLST_RES_COLS_2)
a['blast_hits_aa__' + ss] = parse_blast_results_file(_, BLST_RES_COLS_2)
with open(cache_file, 'wb') as f:
pickle.dump(settings, f, protocol=PICKLE_PROTOCOL)
def dep_check_rcorrector(dir_dep, force):
url = 'https://github.com/karolisr/Rcorrector/archive/master.tar.gz'
dnld_path = opj(dir_dep, 'rcorrector.tar.gz')
try:
try:
jellyfish = which('jellyfish')
run([jellyfish, '--help'])
except Exception:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'Rcorrector'))
jellyfish = opj(dir_bin, 'jellyfish', 'bin', 'jellyfish')
raise
if force is True:
raise
rcorrector = which('run_rcorrector.pl')
run([rcorrector, '-version'])
except Exception:
try:
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'Rcorrector'))
try:
rcorrector = opj(dir_bin, 'run_rcorrector.pl')
run([rcorrector, '-version'])
except Exception:
Log.wrn('Rcorrector was not found on this system, trying to '
'download.')
raise
try:
run([jellyfish, '--version'])
except Exception:
Log.wrn(
'jellyfish is required by Rcorrector, but was not found. '
'Trying to download and recompile Rcorrector and '
'jellyfish.')
raise
except Exception:
if ope(dnld_path):
remove(dnld_path)
if dir_bin != opj(dir_dep, ''):
rmtree(dir_bin)
download_file(url, dnld_path)
tar_ref = tarfile.open(dnld_path, 'r:gz')
tar_ref.extractall(dir_dep)
tar_ref.close()
dir_bin = opj(dir_dep, get_dep_dir(dir_dep, 'Rcorrector'))
try:
Log.wrn('Compiling Rcorrector.')
run('make', cwd=dir_bin)
rcorrector = opj(dir_bin, 'run_rcorrector.pl')
jellyfish = opj(dir_bin, 'jellyfish', 'bin', 'jellyfish')
chmod(
rcorrector, stat.S_IRWXU | stat.S_IRGRP | stat.S_IXGRP
| stat.S_IROTH | stat.S_IXOTH)
run([rcorrector, '-version'])
if not ope(jellyfish):
jellyfish = which('jellyfish')
run([jellyfish, '--version'])
except Exception:
Log.err('Something went wrong while trying to compile '
'Rcorrector.')
Log.msg('Try downloading and installing it manually from: '
'https://github.com/karolisr/Rcorrector')
return None
v = get_dep_version([rcorrector, '-version'], r'^Rcorrector\sv([\d\.]*)')
Log.msg('Rcorrector is available:', v + ' ' + rcorrector)
return rcorrector