if (args.taxids is None or args.taxids == "") and (args.taxa is None
or args.taxa == ""):
print(
colorify(f'Either --taxids or --taxa parameter is required',
'red'))
sys.exit(1)
if (args.taxids is not None and
args.taxids != "") and (args.taxa is not None and args.taxa != ""):
print(colorify(f'Use either --taxids or --taxa, not both', 'red'))
sys.exit(1)
##
if "EGGNOG_DATA_DIR" in os.environ:
set_data_path(os.environ["EGGNOG_DATA_DIR"])
if args.data_dir:
set_data_path(args.data_dir)
data_path = get_data_path()
# http://eggnog5.embl.de/download/eggnog_5.0/e5.proteomes.faa
if not pexists(get_eggnog_proteins_file()):
if args.allyes or ask(
f"Download eggnog5 proteins to {data_path}? ~9GB (It is required to create new databases)"
) == 'y':
print(
colorify(
f'Downloading eggnog5 proteins file to {data_path}...',
'green'))
def parse_args(parser):
args = parser.parse_args()
if "EGGNOG_DATA_DIR" in os.environ:
set_data_path(os.environ["EGGNOG_DATA_DIR"])
if args.data_dir:
set_data_path(args.data_dir)
if args.version:
version = ""
try:
version = get_full_version_info()
except Exception:
version = get_version()
print(version)
sys.exit(0)
args.call_info = get_call_info()
if args.list_taxa:
from eggnogmapper.vars import LEVEL_DEPTH, LEVEL_DICT, LEVEL_NAMES, LEVEL_PARENTS
print("tax_name\ttax_id\tdepth\tparents\tparents_names")
for tax_name, tax_id in LEVEL_DICT.items():
depth = LEVEL_DEPTH.get(tax_id, "-")
parents = LEVEL_PARENTS.get(tax_id, "-")
parents_names = [LEVEL_NAMES.get(x, "-") for x in parents]
print(f"{tax_name}\t{tax_id}\t{depth}\t{','.join(parents)}\t{','.join(parents_names)}")
sys.exit(0)
if args.cpu == 0:
args.cpu = multiprocessing.cpu_count()
# translate
if args.itype in [ITYPE_GENOME, ITYPE_META, ITYPE_PROTS] and args.translate == True:
parser.error('"--translate" only can be used with "--itype CDS"')
# Gene prediction
if args.training_genome is not None and args.training_file is None:
parser.error('"--training_genome requires --training_file"')
if args.training_genome is None and args.training_file is not None:
if not os.path.isfile(args.training_file):
parser.error('"--training_file must point to an existing file, if no --training_genome is provided."')
# Search modes
if args.mode == SEARCH_MODE_DIAMOND:
dmnd_db = args.dmnd_db if args.dmnd_db else get_eggnog_dmnd_db()
if not pexists(dmnd_db):
print(colorify('DIAMOND database %s not present. Use download_eggnog_database.py to fetch it' % dmnd_db, 'red'))
raise EmapperException()
if args.input is not None:
if args.annotate_hits_table is not None:
print(colorify(f"--annotate_hits_table will be ignored, due to -m {SEARCH_MODE_DIAMOND}", 'blue'))
args.annotate_hits_table = None
else:
# the default -m is diamond, but we will consider -m no_search as default when
# --annotate_hits_table has been provided and -i has not been provided
if args.annotate_hits_table is not None:
print(colorify(f"Assuming -m {SEARCH_MODE_NO_SEARCH}", 'blue'))
args.mode = SEARCH_MODE_NO_SEARCH
else:
parser.error('An input fasta file is required (-i)')
# Output file required
if not args.output:
parser.error('An output project name is required (-o)')
if args.resume == True:
print(colorify("Diamond jobs cannot be resumed. --resume will be ignored.", 'blue'))
args.resume = False
elif args.mode == SEARCH_MODE_MMSEQS2:
mmseqs_db = args.mmseqs_db if args.mmseqs_db else get_eggnog_mmseqs_db()
if not pexists(mmseqs_db):
print(colorify('MMseqs2 database %s not present. Use download_eggnog_database.py to fetch it' % mmseqs_db, 'red'))
raise EmapperException()
if not args.input:
parser.error('An input fasta file is required (-i)')
# Output file required
if not args.output:
parser.error('An output project name is required (-o)')
if args.resume == True:
print(colorify("MMseqs2 jobs cannot be resumed. --resume will be ignored.", 'blue'))
args.resume = False
if args.annotate_hits_table is not None:
print(colorify(f"--annotate_hits_table will be ignored, due to -m {SEARCH_MODE_MMSEQS2}", 'blue'))
args.annotate_hits_table = None
elif args.mode == SEARCH_MODE_HMMER:
# if args.usemem == True:
# total_workers = args.num_workers * args.num_servers
# if args.cpu < total_workers:
# parser.error(f"Less cpus ({args.cpu}) than total workers ({total_workers}) were specified.")
# if args.cpu % total_workers != 0:
# parser.error(f"Number of cpus ({args.cpu}) must be a multiple of total workers ({total_workers}).")
# args.cpus_per_worker = int(args.cpu / total_workers)
# sys.stderr.write(f"CPUs per worker: {args.cpus_per_worker}\n")
# else:
# args.cpus_per_worker = args.cpu
if not args.input:
parser.error('An input file is required (-i)')
# Output file required
if not args.output:
parser.error('An output project name is required (-o)')
# Hmmer database
# NOTE: hmmer database format, name and checking if exists is done within hmmer module
if not args.db:
parser.error('HMMER mode requires a target database (-d, --database).')
if args.itype == ITYPE_CDS:
args.translate = True
if (args.itype == ITYPE_GENOME or args.itype == ITYPE_META) and args.genepred == GENEPRED_MODE_SEARCH:
parser.error('HMMER mode is not compatible with "--genepred search" option.')
if args.annotate_hits_table is not None:
print(colorify(f"--annotate_hits_table will be ignored, due to -m {SEARCH_MODE_HMMER}", 'blue'))
args.annotate_hits_table = None
if args.clean_overlaps is not None:
if args.clean_overlaps == "none":
args.clean_overlaps = None
elif args.mode == SEARCH_MODE_CACHE:
if args.cache_file is None:
parser.error('A file with annotations and md5 of queries is required (-c FILE)')
if args.decorate_gff != DECORATE_GFF_NONE:
print(colorify("WARNING: no GFF will be created for cache-based annotations. It is not implemented yet, sorry.", 'red'))
if args.no_annot == True:
parser.error(f'Cache mode (-m {SEARCH_MODE_CACHE}) should be used to annotate.')
elif args.mode == SEARCH_MODE_NO_SEARCH:
if args.no_annot == False and not args.annotate_hits_table:
parser.error(f'No search mode (-m {SEARCH_MODE_NO_SEARCH}) requires a hits table to annotate (--annotate_hits_table FILE.seed_orthologs)')
if args.md5 == True and args.input is None:
parser.error(f'--md5 requires an input FASTA file (-i FASTA).')
# if args.no_annot == True and args.report_orthologs == False:
# parser.error(f'Nothing to do if running in no search mode (-m {SEARCH_MODE_NO_SEARCH}), with --no_annot and without --report_orthologs.')
else:
parser.error(f'unrecognized search mode (-m {args.mode})')
# Search thresholds
args.dmnd_evalue = args.mmseqs_evalue = args.hmm_evalue = args.evalue
args.dmnd_score = args.mmseqs_score = args_hmm_score = args.score
args.qcov = args.query_cover
# Annotation options
if args.no_annot == False or args.report_orthologs == True:
if not pexists(get_eggnogdb_file()):
print(colorify('Annotation database data/eggnog.db not present. Use download_eggnog_database.py to fetch it', 'red'))
raise EmapperException()
args.tax_scope_mode, args.tax_scope_id = __parse_tax_scope(args.tax_scope)
if args.target_taxa is not None:
args.target_taxa = args.target_taxa.split(",")
if args.excluded_taxa is not None:
args.excluded_taxa = args.excluded_taxa.split(",")
# Sets GO evidence bases
if args.go_evidence == 'experimental':
args.go_evidence = set(["EXP","IDA","IPI","IMP","IGI","IEP"])
args.go_excluded = set(["ND", "IEA"])
elif args.go_evidence == 'non-electronic':
args.go_evidence = None
args.go_excluded = set(["ND", "IEA"])
elif args.go_evidence == 'all':
args.go_evidence = None
args.go_excluded = None
else:
raise ValueError('Invalid --go_evidence value')
# PFAM annotation options
if args.pfam_transfer in [PFAM_TRANSFER_BEST_OG, PFAM_TRANSFER_NARROWEST_OG, PFAM_TRANSFER_SEED_ORTHOLOG]:
pass
else:
raise ValueError(f'Invalid --pfam_transfer option {args.pfam_transfer}')
if args.pfam_realign == PFAM_REALIGN_NONE:
pass
elif args.pfam_realign == PFAM_REALIGN_REALIGN or args.pfam_realign == PFAM_REALIGN_DENOVO:
if not args.input:
parser.error(f'An input fasta file is required (-i) for --pfam_realign {args.pfam_realign}')
else:
raise ValueError(f'Invalid --pfam_realign option {args.pfam_realign}')
total_workers = args.num_workers * args.num_servers
if args.cpu < total_workers:
parser.error(f"Less cpus ({args.cpu}) than total workers ({total_workers}) were specified.")
if args.cpu % total_workers != 0:
parser.error(f"Number of cpus ({args.cpu}) must be a multiple of total workers ({total_workers}).")
args.cpus_per_worker = int(args.cpu / total_workers)
return args
help='simulate and print commands. Nothing is downloaded')
parser.add_argument('-q',
action="store_true",
dest='quiet',
help='quiet_mode')
parser.add_argument("--data_dir",
metavar='',
type=existing_dir,
help='Directory to use for DATA_PATH.')
args = parser.parse_args()
if args.data_dir:
set_data_path(args.data_dir)
# if args.force or not pexists(pjoin(get_data_path(), 'og2level.tsv.gz')):
# print colorify('Downloading "og2level.tsv.gz" at %s' %get_data_path(), 'green')
# download_og2level()
# if 'all' in args.dbs:
# args.dbs = EGGNOG_DATABASES
if args.force or not pexists(pjoin(get_data_path(), 'eggnog.db')):
if args.allyes or ask("Download main annotation database?") == 'y':
print colorify(
'Downloading "eggnog.db" at %s...' % get_data_path(), 'green')
download_annotations()
else:
print 'Skipping'
def parse_args(parser):
args = parser.parse_args()
if "EGGNOG_DATA_DIR" in os.environ:
set_data_path(os.environ["EGGNOG_DATA_DIR"])
if args.data_dir:
set_data_path(args.data_dir)
if args.version:
version = ""
try:
version = get_full_version_info()
except Exception:
version = get_version()
print(version)
sys.exit(0)
args.call_info = get_call_info()
if args.list_taxa:
print_taxa()
sys.exit(0)
if args.cpu == 0:
args.cpu = multiprocessing.cpu_count()
multiprocessing.set_start_method(args.mp_start_method)
if args.resume == True and args.override == True:
parser.error('Only one of --resume or --override is allowed.')
# Gene prediction
if args.training_genome is not None and args.training_file is None:
parser.error('"--training_genome requires --training_file"')
if args.training_genome is None and args.training_file is not None:
if not os.path.isfile(args.training_file):
parser.error('"--training_file must point to an existing file, if no --training_genome is provided."')
# Search modes
if args.mode == SEARCH_MODE_DIAMOND:
dmnd_db = args.dmnd_db if args.dmnd_db else get_eggnog_dmnd_db()
if not pexists(dmnd_db):
print(colorify('DIAMOND database %s not present. Use download_eggnog_database.py to fetch it' % dmnd_db, 'red'))
raise EmapperException()
if args.input is not None:
if args.annotate_hits_table is not None:
print(colorify(f"--annotate_hits_table will be ignored, due to -m {SEARCH_MODE_DIAMOND}", 'blue'))
args.annotate_hits_table = None
else:
# the default -m is diamond, but we will consider -m no_search as default when
# --annotate_hits_table has been provided and -i has not been provided
if args.annotate_hits_table is not None:
print(colorify(f"Assuming -m {SEARCH_MODE_NO_SEARCH}", 'blue'))
args.mode = SEARCH_MODE_NO_SEARCH
else:
parser.error('An input fasta file is required (-i)')
# Output file required
if not args.output:
parser.error('An output project name is required (-o)')
elif args.mode == SEARCH_MODE_MMSEQS2:
mmseqs_db = args.mmseqs_db if args.mmseqs_db else get_eggnog_mmseqs_db()
if not pexists(mmseqs_db):
print(colorify('MMseqs2 database %s not present. Use download_eggnog_database.py to fetch it' % mmseqs_db, 'red'))
raise EmapperException()
if not args.input:
parser.error('An input fasta file is required (-i)')
# Output file required
if not args.output:
parser.error('An output project name is required (-o)')
if args.annotate_hits_table is not None:
print(colorify(f"--annotate_hits_table will be ignored, due to -m {SEARCH_MODE_MMSEQS2}", 'blue'))
args.annotate_hits_table = None
elif args.mode == SEARCH_MODE_HMMER:
if not args.input:
parser.error('An input file is required (-i)')
# Output file required
if not args.output:
parser.error('An output project name is required (-o)')
# Hmmer database
# NOTE: hmmer database format, name and checking if exists is done within hmmer module
if not args.db:
parser.error('HMMER mode requires a target database (-d, --database).')
if args.itype == ITYPE_CDS:
args.translate = True
if (args.itype == ITYPE_GENOME or args.itype == ITYPE_META) and args.genepred == GENEPRED_MODE_SEARCH:
parser.error('HMMER mode is not compatible with "--genepred search" option.')
if args.annotate_hits_table is not None:
print(colorify(f"--annotate_hits_table will be ignored, due to -m {SEARCH_MODE_HMMER}", 'blue'))
args.annotate_hits_table = None
if args.clean_overlaps is not None:
if args.clean_overlaps == "none":
args.clean_overlaps = None
elif args.mode == SEARCH_MODE_CACHE:
if args.cache_file is None:
parser.error('A file with annotations and md5 of queries is required (-c FILE)')
if args.decorate_gff != DECORATE_GFF_NONE:
print(colorify("WARNING: no GFF will be created for cache-based annotations. It is not implemented yet, sorry.", 'red'))
if args.no_annot == True:
parser.error(f'Cache mode (-m {SEARCH_MODE_CACHE}) should be used to annotate.')
elif args.mode == SEARCH_MODE_NO_SEARCH:
if args.no_annot == False and not args.annotate_hits_table:
parser.error(f'No search mode (-m {SEARCH_MODE_NO_SEARCH}) requires a hits table to annotate (--annotate_hits_table FILE.seed_orthologs)')
if args.md5 == True and args.input is None:
parser.error(f'--md5 requires an input FASTA file (-i FASTA).')
else:
parser.error(f'unrecognized search mode (-m {args.mode})')
# Search thresholds
args.dmnd_evalue = args.mmseqs_evalue = args.hmm_evalue = args.evalue
args.dmnd_score = args.mmseqs_score = args_hmm_score = args.score
args.qcov = args.query_cover
# Annotation options
if args.no_annot == False or args.report_orthologs == True:
if not pexists(get_eggnogdb_file()):
print(colorify('Annotation database data/eggnog.db not present. Use download_eggnog_database.py to fetch it', 'red'))
raise EmapperException()
args.tax_scope_ids = parse_tax_scope(args.tax_scope)
if args.target_taxa is not None:
args.target_taxa = args.target_taxa.split(",")
if args.excluded_taxa is not None:
args.excluded_taxa = args.excluded_taxa.split(",")
# Sets GO evidence bases
if args.go_evidence == 'experimental':
args.go_evidence = set(["EXP","IDA","IPI","IMP","IGI","IEP"])
args.go_excluded = set(["ND", "IEA"])
elif args.go_evidence == 'non-electronic':
args.go_evidence = None
args.go_excluded = set(["ND", "IEA"])
elif args.go_evidence == 'all':
args.go_evidence = None
args.go_excluded = None
else:
raise ValueError('Invalid --go_evidence value')
# PFAM annotation options
if args.pfam_realign == PFAM_REALIGN_NONE:
pass
elif args.pfam_realign == PFAM_REALIGN_REALIGN or args.pfam_realign == PFAM_REALIGN_DENOVO:
if not args.input:
parser.error(f'An input fasta file is required (-i) for --pfam_realign {args.pfam_realign}')
else:
raise ValueError(f'Invalid --pfam_realign option {args.pfam_realign}')
total_workers = args.num_workers * args.num_servers
if args.cpu < total_workers:
parser.error(f"Less cpus ({args.cpu}) than total workers ({total_workers}) were specified.")
if args.cpu % total_workers != 0:
parser.error(f"Number of cpus ({args.cpu}) must be a multiple of total workers ({total_workers}).")
args.cpus_per_worker = int(args.cpu / total_workers)
return args