def setupdir(strains, genomedb):
try:
os.makedirs(outdir)
except OSError as exc:
if exc.errno == errno.EEXIST:
print "Database folder exists:", outdir
pp.createdirs(outdir, ["faa", "m8", "out", "paranoid_output", "dmnd_tmp"])
if not os.path.isdir(genomedb):
print "GenomeDB folder", genomedb, "doesn't exist...exiting."
sys.exit()
if not os.path.isdir(os.path.join(genomedb, "pep")):
print "GenomeDB folder is missing a 'pep' folder...exiting."
sys.exit()
if verbose:
print "Formatting", len(strains), "fasta files..."
for s in strains:
try:
i = open(os.path.join(genomedb, "pep", s + ".pep.fa"), "r")
except IOError as exc:
if exc.errno == 2:
print s, 'not found in database...check your strainlist.'
sys.exit()
o = open(os.path.join(outdir, "faa", s + ".faa"), "w")
for seq in SeqIO.parse(i, 'fasta'):
seq.id = s + "|" + str(seq.id)
SeqIO.write(seq, o, 'fasta')
o.close()
return
def main():
args = parse_args()
global prefix
prefix = os.path.abspath(args.prefix)
global outdir
outdir = os.path.abspath(args.outdir)
pp.createdirs(prefix, ["orthos", "ortho_align", "hmms"])
if args.strains:
strains = [
line.rstrip() for line in open(os.path.abspath(args.strains), 'r')
]
else:
strains = get_strains()
if args.orthos:
orthos = [
line.rstrip() for line in open(os.path.abspath(args.orthos), 'r')
]
else:
if args.threshold:
orthos = parse_threshold_matrix(args.threshold, strains)
else:
orthos = parse_matrix(strains)
if args.cpus:
cpus = args.cpus
else:
cpus = mp.cpu_count()
global use_MP
if args.use_MP:
use_MP = True
else:
use_MP = False
index_hmms()
extract_hmms(orthos)
get_orthos(orthos, strains)
align_orthos(orthos, cpus)
create_master_alignment(orthos, strains)
if args.clean:
pp.cleanup(os.path.join(prefix, "ortho_align"))
pp.cleanup(os.path.join(prefix, "orthos"))
pp.cleanup(os.path.join(prefix, "hmms"))
def main():
args = parse_args()
global pypath
pypath = os.path.abspath(os.path.dirname(sys.argv[0]))
global outdir
outdir = os.path.abspath(args.outdir)
genomedb = os.path.abspath(args.genomedb)
new_strains = [
line.rstrip()
for line in open(os.path.abspath(args.new_strainlist), 'r')
]
global cpus
if args.cpus:
cpus = args.cpus
else:
cpus = mp.cpu_count()
global use_MP
if args.use_MP:
use_MP = True
else:
use_MP = False
pp.createdirs(outdir, [
"prop_faa", "prop_dmnd", "prop_m8", "prop_out", "prop_paranoid_output",
"prop_homolog_faa"
])
check_strains(new_strains, genomedb)
make_diamond_databases(new_strains)
run_diamond(new_strains)
genes = get_genes(new_strains)
parse_diamond(genes)
run_inparanoid(new_strains, pypath)
group_members = parse_inparanoid(new_strains)
extract_fastas(genes, group_members)
pp.dump_matrices(outdir)
for f in [
"prop_m8", "prop_out", "prop_dmnd", "prop_paranoid_output",
"prop_faa", "prop_homolog_faa"
]:
pp.cleanup(os.path.join(outdir, f))
def main():
args = parse_args()
genomedb = os.path.abspath(args.genomedb)
strains = [x.rstrip() for x in open(os.path.abspath(args.strainlist), 'r')]
if len(set(strains)) != len(strains):
print "Duplicate entry in strainlist! Exiting..."
global outdir, pypath
outdir = os.path.abspath(args.outdir)
pypath = os.path.abspath(os.path.dirname(sys.argv[0]))
if args.mode:
if args.mode not in [
"multi_setup", "parse", "extract", "cluster", "build"
]:
print "Unknown mode!!! Exiting..."
sys.exit()
global cpus
if args.cpus:
cpus = args.cpus
else:
cpus = mp.cpu_count()
global clean
if args.clean:
clean = True
else:
clean = False
global verbose
if args.verbose:
verbose = True
else:
verbose = False
global inflate
if args.inflate:
inflate = args.inflate
else:
inflate = 2.0
global threshold
if args.threshold:
threshold = args.threshold
else:
threshold = 0
global multi
if args.multi:
multi = True
else:
multi = False
global use_MP
if args.use_MP:
use_MP = True
else:
use_MP = False
if not args.mode or args.mode == "multi_setup":
setupdir(strains, genomedb)
shutil.copy(os.path.abspath(args.strainlist),
os.path.join(outdir, "strainlist.txt"))
make_diamond_databases(strains)
run_diamond(strains)
if not args.mode or args.mode == "parse":
genes = get_genes(strains)
parse_diamond(genes, strains)
run_inparanoid(strains)
if not args.mode or args.mode == "cluster":
if clean:
pp.cleanup(os.path.join(outdir, "out"))
pp.createdirs(outdir, ["mcl"])
create_abc_file()
mcxload()
mcl_cluster()
mcxdump()
if not args.mode or args.mode == "extract":
seqdata, desc, seq_number = hash_fastas()
pp.createdirs(
outdir,
["homolog_faa", "clustered", "aligned", "hmms", "consensus_seqs"])
parse_clusters(strains, seq_number)
parse_groups(seqdata, desc)
if not args.mode or args.mode == "build":
cdhit_seqs()
align_groups()
if clean:
pp.cleanup(os.path.join(outdir, "clustered"))
build_hmms()
if clean:
pp.cleanup(os.path.join(outdir, "aligned"))
emit_consensus_seqs()
combine_seqs()
combine_homologs()
if clean:
pp.cleanup(os.path.join(outdir, "hmms"))
pp.cleanup(os.path.join(outdir, "consensus_seqs"))
pp.cleanup(os.path.join(outdir, "m8"))
pp.cleanup(os.path.join(outdir, "paranoid_output"))
pp.cleanup(os.path.join(outdir, "dmnd_tmp"))
pp.cleanup(os.path.join(outdir, "faa"))
pp.cleanup(os.path.join(outdir, "homolog_faa"))
pp.cleanup(os.path.join(outdir, "mcl"))
os.remove(os.path.join(outdir, "all_strains.dmnd"))
pp.dump_matrices(outdir)