def main():
parser = get_parser()
args = validate_args(parser)
seq_records = seqrecords.read_fasta(args.fasta)
if args.word_size:
p = word_pattern.create(seq_records.seq_list, args.word_size)
else:
p = word_pattern.read(args.word_pattern)
if args.reduce_alphabet:
p = p.reduce_alphabet(seqcontent.get_reduced_alphabet(args.molecule))
if args.merge_revcomp:
p = p.merge_revcomp()
freqs = word_vector.Freqs(seq_records.length_list, p)
dist = word_distance.Distance(freqs, args.distance)
matrix = distmatrix.create(seq_records.id_list, dist)
if args.out:
oh = open(args.out, 'w')
matrix.write_to_file(oh, args.outfmt)
oh.close()
else:
matrix.display(args.outfmt)
def validate_args(parser):
args = parser.parse_args()
if args.word_size:
if args.word_size < 3:
parser.error('Word size must be >= 3')
elif args.word_patterns:
l = []
for i in range(0, 3):
try:
p = word_pattern.read(args.word_patterns[i])
l.append(p)
except Exception:
parser.error('Invalid format for word pattern: {0}'.format(
args.word_patterns[i].name))
if len(l) == 3:
# check if follow rule
k, k1, k2 = [len(p.pat_list[0]) for p in l]
if not (k == k1 + 1 == k2 + 2):
parser.error(
'''Word pattern lengths do not follow k, k-1, k-2''')
args.word_patterns = l
else:
parser.error("Specify either: --word_size or --word_pattern.")
return args
def test_input_output_file_pattern(self):
for wordpos in [True, False]:
p1 = word_pattern.create(self.dna_records.seq_list,
word_size=1,
wordpos=wordpos)
oh = open(utils.get_test_data('pattern.txt'), 'w')
oh.write(p1.format())
oh.close()
fh = open(utils.get_test_data('pattern.txt'))
p2 = word_pattern.read(fh)
fh.close()
self.assertEqual(p1.format(), p2.format())
os.remove(utils.get_test_data('pattern.txt'))
def validate_args(parser):
args = parser.parse_args()
if args.word_size:
if args.word_size < 1:
parser.error('word size must be >= 1')
elif args.word_pattern:
p = word_pattern.read(args.word_pattern)
if not p.pos_list:
e = "{0} does not contain info on word positions.\n"
e += "Please use: create_wordpattern.py with"
e += " --word_position option."
parser.error(e.format(args.word_pattern.name))
else:
args.word_pattern = p
else:
parser.error("Specify either: --word_size or --word_pattern.")
return args
def main():
parser = get_parser()
args = validate_args(parser)
seq_records = seqrecords.read_fasta(args.fasta)
if args.word_size:
p = word_pattern.create(seq_records.seq_list, args.word_size)
else:
p = word_pattern.read(args.word_pattern)
bools = word_vector.Bools(seq_records.length_list, p)
dist = word_bool_distance.Distance(bools, args.distance)
matrix = distmatrix.create(seq_records.id_list, dist)
if args.out:
oh = open(args.out, 'w')
matrix.write_to_file(oh, args.outfmt)
oh.close()
else:
matrix.display(args.outfmt)
def main():
parser = get_parser()
args = validate_args(parser)
seq_records = seqrecords.read_fasta(args.fasta)
if args.word_size:
p = word_pattern.create(seq_records.seq_list, args.word_size)
else:
p = word_pattern.read(args.word_pattern)
veccls = {'counts': word_vector.Counts, 'freqs': word_vector.Freqs}
vecclsw = {
'counts': word_vector.CountsWeight,
'freqs': word_vector.FreqsWeight
}
if args.vector == 'counts' or args.vector == 'freqs':
if args.char_weights is None:
vec = veccls[args.vector](seq_records.length_list, p)
else:
weightmodel = word_vector.WeightModel(
char_weights=args.char_weights)
vec = vecclsw[args.vector](seq_records.length_list, p, weightmodel)
else:
if args.alphabet_size:
freqmodel = word_vector.EqualFreqs(
alphabet_size=args.alphabet_size)
else:
freqmodel = word_vector.EquilibriumFreqs(args.char_freqs)
vec = word_vector.FreqsStd(seq_records.length_list, p, freqmodel)
dist = word_distance.Distance(vec, args.distance)
matrix = distmatrix.create(seq_records.id_list, dist)
if args.out:
oh = open(args.out, 'w')
matrix.write_to_file(oh, args.outfmt)
oh.close()
else:
matrix.display(args.outfmt)
