set_lines = ["all = ["]
filenames = from_dir(seq_dir, re.compile(r'.*\..*'))
counter = 1
for filename in filenames:
print filename,
while True:
if filename.find(".gbk") > 0:
# process genbank
try:
record = load_genbank(seq_dir+filename)
except IOError:
print "failed to load Genbank file"
break
except Exception:
print "failed to handle Genbank file"
break
else:
print "...",
seq_format = 'gbk'
elif filename.find(".fas") > 0:
# process fasta (for mfas, load first record)
try:
record = load_fasta(seq_dir+filename)
except IOError:
except Exception:
print "Error retrieving record"
break
else:
if rec_id[0:2] == 'NZ': # disposition for WGS record sets
print "fetching WGS dataset",
# create a dedicated directory
seqdir = data_dir + rec_id + "/"
ensure_dir([seqdir])
# open genome record stub to get the contig count
fname = data_dir + rec_id + ".gbk"
try:
stub = load_genbank(fname)
except IOError:
print "Error loading", fname
break
base_code = stub.annotations['wgs'][0][:10] # 7 if not NZ_
ctg_num = int(stub.annotations['wgs'][-1][10:]) # 7
records = []
# fetch contig records
ctg_count = 0
while ctg_count < ctg_num:
# TODO: better formatting
ctg_count += 1
if ctg_count < 10:
filenames = from_dir(dir_in, re.compile(r'.*\.'+file_ext))
for filename in filenames:
rec_name = filename[:filename.find("."+file_ext)]
print rec_name,
genome_path = dir_in+filename
dbfile_path = "data/blast_db/"+rec_name
while True:
if not path.exists(dbfile_path+".nhr"):
if file_ext == 'gbk':
try:
print "converting,",
record = load_genbank(genome_path)
except IOError:
print "failed to load Genbank file"
break
else:
try:
genome_path = dir_in+rec_name+".fas"
write_fasta(genome_path, record)
except Exception:
print "failed to write Fasta file"
break
try:
print "making a DB,",
make_blastDB(dbfile_path, genome_path, 'nucl')
except IOError:
print "failed to make DB"
except Exception:
print "Error retrieving record"
break
else:
if rec_id[0:2] == 'NZ': # disposition for WGS record sets
print "fetching WGS dataset",
# create a dedicated directory
seqdir = data_dir+rec_id+"/"
ensure_dir([seqdir])
# open genome record stub to get the contig count
fname = data_dir+rec_id+".gbk"
try:
stub = load_genbank(fname)
except IOError:
print "Error loading", fname
break
base_code = stub.annotations['wgs'][0][:10] # 7 if not NZ_
ctg_num = int(stub.annotations['wgs'][-1][10:]) # 7
records = []
# fetch contig records
ctg_count = 0
while ctg_count < ctg_num:
# TODO: better formatting
ctg_count += 1
if ctg_count < 10:
filenames = from_dir(dir_in, re.compile(r'.*\.' + file_ext))
for filename in filenames:
rec_name = filename[:filename.find("." + file_ext)]
print rec_name,
genome_path = dir_in + filename
dbfile_path = "data/blast_db/" + rec_name
while True:
if not path.exists(dbfile_path + ".nhr"):
if file_ext == 'gbk':
try:
print "converting,",
record = load_genbank(genome_path)
except IOError:
print "failed to load Genbank file"
break
else:
try:
genome_path = dir_in + rec_name + ".fas"
write_fasta(genome_path, record)
except Exception:
print "failed to write Fasta file"
break
try:
print "making a DB,",
make_blastDB(dbfile_path, genome_path, 'nucl')
except IOError:
print "failed to make DB"
## script to combine several fasta files into a single one
import re
from sys import argv
from libs.common import from_dir, load_fasta, load_genbank, write_fasta
origin_dir = "data/" + argv[1]
destin_file = origin_dir + "/" + argv[2] + ".fas"
file_ext = argv[3]
filenames = from_dir(origin_dir, re.compile(r'.*\.' + file_ext))
records = []
for filename in filenames:
# load record
if file_ext == 'fas':
records.append(load_fasta(origin_dir + "/" + filename))
elif file_ext == 'gbk':
records.append(load_genbank(origin_dir + "/" + filename))
print filename
write_fasta(destin_file, records)
set_lines = ["all = ["]
filenames = from_dir(seq_dir, re.compile(r'.*\..*'))
counter = 1
for filename in filenames:
print filename,
while True:
if filename.find(".gbk") > 0:
# process genbank
try:
record = load_genbank(seq_dir + filename)
except IOError:
print "failed to load Genbank file"
break
except Exception:
print "failed to handle Genbank file"
break
else:
print "...",
seq_format = 'gbk'
elif filename.find(".fas") > 0:
# process fasta (for mfas, load first record)
try:
record = load_fasta(seq_dir + filename)
except IOError:
trn_file = origin_dir+"prodigal.trn"
ensure_dir([annot_gbk_dir, annot_aa_dir])
filenames = from_dir(origin_dir, re.compile(r'.*\.'+file_ext+'.*'))
for filename in filenames:
rec_name = filename[:filename.find("."+file_ext)]
print rec_name, "...",
# load data
if file_ext == 'fas':
fas_file = origin_dir+"/"+filename
gbk_file = fas2gbk(fas_file)
record = load_genbank(gbk_file)
else:
gbk_file = origin_dir+"/"+filename
fas_file = gbk2fas(gbk_file)
record = load_genbank(gbk_file)
# run prediction
annot_aa = annot_aa_dir+rec_name+"_ann.fas"
annot_gbk = annot_gbk_dir+rec_name+"_ann.gbk"
if not path.exists(trn_file):
train_prodigal(fas_file, trn_file, "-q")
if not path.exists(annot_aa):
run_prodigal(fas_file, annot_gbk, annot_aa, trn_file, "-q")
# collect orfs
record.features = []
## script to combine several fasta files into a single one
import re
from sys import argv
from libs.common import from_dir, load_fasta, load_genbank, write_fasta
origin_dir = "data/"+argv[1]
destin_file = origin_dir+"/"+argv[2]+".fas"
file_ext = argv[3]
filenames = from_dir(origin_dir, re.compile(r'.*\.'+file_ext))
records = []
for filename in filenames:
# load record
if file_ext == 'fas':
records.append(load_fasta(origin_dir+"/"+filename))
elif file_ext == 'gbk':
records.append(load_genbank(origin_dir+"/"+filename))
print filename
write_fasta(destin_file, records)
feat_tag = argv[4]
feat_name = argv[5]
main_out = data_dir + feat_name + "_seqs.fas"
records = []
ensure_dir([data_dir])
filenames = from_dir(dir_in, re.compile(r'.*\.gbk'))
for filename in filenames:
rec_name = filename[:filename.find(".gbk")]
print '.',
# load data
record = load_genbank(dir_in + "/" + filename)
# scan annotations
for feat in record.features:
if feat.type == feat_type:
try:
if feat_name in feat.qualifiers[feat_tag]:
print '\nfound', feat_name, 'in', rec_name
# extract sequence
new_rec = feat.extract(record)
new_rec.id = rec_name + '_' + feat_name
new_rec.description = "Extracted from " + new_rec.description
records.append(new_rec)
except KeyError:
pass
feat_tag = argv[4]
feat_name = argv[5]
main_out = data_dir+feat_name+"_seqs.fas"
records = []
ensure_dir([data_dir])
filenames = from_dir(dir_in, re.compile(r'.*\.gbk'))
for filename in filenames:
rec_name = filename[:filename.find(".gbk")]
print '.',
# load data
record = load_genbank(dir_in+"/"+filename)
# scan annotations
for feat in record.features:
if feat.type == feat_type:
try:
if feat_name in feat.qualifiers[feat_tag]:
print '\nfound', feat_name, 'in', rec_name
# extract sequence
new_rec = feat.extract(record)
new_rec.id = rec_name+'_'+feat_name
new_rec.description = "Extracted from "+new_rec.description
records.append(new_rec)
except KeyError:
pass
for genome in genome_list:
print genome['name'],
origin_file = origin_dir+genome['file']
while True:
if genome['input'] == 'cgbk':
print "ignoring cgbk file"
break
elif genome['input'] == 'gbk':
try:
records = [load_genbank(origin_file)]
except IOError:
print "failed to load file"
break
elif genome['input'] == 'fas':
try:
records = [load_fasta(origin_file)]
except IOError:
print "failed to load file"
break
elif genome['input'] == 'mfas':
try:
records = load_multifasta(origin_file)
except IOError:
for genome in genomes:
g_vector = []
print genome['name'],
while True:
try:
assert genome['input'] == 'gbk'
except ValueError:
print "bad format (skipping)"
break
# load genome file to extract features (to proteins in mfas file)
record = load_genbank(seq_dir + genome['file'])
select = [
feat for feat in record.features if feat.type == feat_type
]
feat_cnt = 0
# cycle through selected features
for feat in select:
feat_cnt += 1
rec = feat.extract(record)
rec.description = genome['name'] + '_' + feat_type + '_' + str(
feat_cnt)
# initialize or update blast DB
if init_DB:
for genome in genomes:
g_vector = []
print genome['name'],
while True:
try:
assert genome['input'] == 'gbk'
except ValueError:
print "bad format (skipping)"
break
# load genome file to extract features (to proteins in mfas file)
record = load_genbank(seq_dir+genome['file'])
select = [feat for feat in record.features
if feat.type == feat_type]
feat_cnt = 0
# cycle through selected features
for feat in select:
feat_cnt +=1
rec = feat.extract(record)
rec.description = genome['name']+'_'+feat_type+'_'+str(feat_cnt)
# initialize or update blast DB
if init_DB:
ref_records = [value[0] for value in symbolDB.values()]
write_fasta(db_file, ref_records)
