def test_counts_from1_to1(self):
dist = word_d2.Distance([self.counts[0]])
matrix = distmatrix.create(self.pep_records.id_list, dist)
exp = [
' 4',
'seq1 0 37 57 140',
'seq2 37 0 28 137',
'seq3 57 28 0 111',
'seq4 140 137 111 0'
]
self.assertEqual(matrix.format(decimal_places=0), "\n".join(exp))
def test_counts_from1_to4(self):
dist = word_d2.Distance(self.counts)
matrix = distmatrix.create(self.pep_records.id_list, dist)
exp = [
' 4',
'seq1 0 130 236 286',
'seq2 130 0 142 258',
'seq3 236 142 0 212',
'seq4 286 258 212 0'
]
self.assertEqual(matrix.format(decimal_places=0), "\n".join(exp))
def test_freqs_from1_to4(self):
dist = word_d2.Distance(self.freqs)
matrix = distmatrix.create(self.pep_records.id_list, dist)
exp = [
' 4',
'seq1 0.0000000 0.0313590 0.0573154 0.1020235',
'seq2 0.0313590 0.0000000 0.0373677 0.0907196',
'seq3 0.0573154 0.0373677 0.0000000 0.0870581',
'seq4 0.1020235 0.0907196 0.0870581 0.0000000'
]
self.assertEqual(matrix.format(), "\n".join(exp))
def test_freqs_from1_to4_d2_squareroot(self):
dist = word_d2.Distance(self.freqs)
dist.set_disttype('d2_squareroot')
matrix = distmatrix.create(self.pep_records.id_list, dist)
exp = [
" 4",
"seq1 0.0000000 0.1770847 0.2394063 0.3194113",
"seq2 0.1770847 0.0000000 0.1933073 0.3011969",
"seq3 0.2394063 0.1933073 0.0000000 0.2950560",
"seq4 0.3194113 0.3011969 0.2950560 0.0000000"
]
self.assertEqual(matrix.format(), "\n".join(exp))
def main():
parser = get_parser()
args = validate_args(parser)
seq_records = seqrecords.read_fasta(args.fasta)
patterns = []
for i in range(args.min_word_size, args.max_word_size + 1):
p = word_pattern.create(seq_records.seq_list, i)
patterns.append(p)
vecs = []
if args.char_weights is not None:
weightmodel = word_vector.WeightModel(char_weights=args.char_weights)
vecklas = {
'counts': word_vector.CountsWeight,
'freqs': word_vector.FreqsWeight
}[args.vector]
kwargs = {
'seq_lengths': seq_records.length_list,
'weightmodel': weightmodel
}
else:
vecklas = {
'counts': word_vector.Counts,
'freqs': word_vector.Freqs
}[args.vector]
kwargs = {'seq_lengths': seq_records.length_list}
for p in patterns:
v = vecklas(patterns=p, **kwargs)
vecs.append(v)
dist = word_d2.Distance(vecs)
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)
matrix.display()
elif method == "d2":
patterns = []
for i in range(1, 5 + 1):
p = word_pattern.create(seq_records.seq_list, i)
patterns.append(p)
counts = []
for p in patterns:
c = word_vector.Counts(seq_records.length_list, p)
counts.append(c)
countsweight = []
weights = seqcontent.get_weights('protein')
weightmodel = word_vector.WeightModel(weights)
for p in patterns:
c = word_vector.CountsWeight(seq_records, p, weightmodel)
countsweight.append(c)
dist = word_d2.Distance(countsweight)
matrix = distmatrix.create(seq_records.id_list, dist)
matrix.display()
elif method == "lempelziv":
distance = lempelziv.Distance(seq_records)
l = ['d', 'd_star', 'd1', 'd1_star', 'd1_star2']
for el in l:
distance.set_disttype(el)
matrix = distmatrix.create(seq_records.id_list, distance)
matrix.display()
def test_set_disttype_throws_exception(self):
dist = word_d2.Distance(self.freqs)
with self.assertRaises(Exception) as context:
dist.set_disttype('nonexistent')
self.assertIn('unknown disttype', str(context.exception))
