def test_count_kmers_k_greater_than_text_length(self):
"""
Test if calling count_kmers with k > len(self.text)
raises ValueError
:return: None
"""
with pytest.raises(ValueError):
kmers.count_kmers("abc", 4)
def test_count_kmers_k_not_greater_than_0(self):
"""
Test if calling count_kmers with k <= 0
raises ValueError
:return: None
"""
with pytest.raises(ValueError):
kmers.count_kmers("abc", -1)
def test_count_kmers_empty_text(self):
"""
Test if calling count_kmers with empty text
raises ValueError
:return: None
"""
with pytest.raises(ValueError):
kmers.count_kmers("", 1)
def test_count_kmers_null_text(self):
"""
Test if calling count_kmers with null text
raises ValueError
:return: None
"""
with pytest.raises(ValueError):
kmers.count_kmers(None, 1)
def test_count_7mers():
seq = 'ACGTCGACCGCGTTA'
counts = count_kmers(7, seq)
assert counts.sum() == (len(seq)-6)*2
counts_rc = count_kmers(7, reverse_complement(seq))
assert (counts == counts_rc).all()
map_to_index = defaultdict(lambda: -1)
map_to_index.update({ 'A': 0, 'C': 1, 'G': 2, 'T': 3 })
seqi_fw = np.array([map_to_index[c] for c in seq])
map_to_index.update({ 'A': 3, 'C': 2, 'G': 1, 'T': 0 })
seqi_rc = np.array([map_to_index[c] for c in seq[::-1]])
for seqi in [seqi_fw, seqi_rc]:
for i in range(6, len(seqi)):
index = (((((seqi[i]*4+seqi[i-1])*4+seqi[i-2])*4+seqi[i-3])*4+seqi[i-4])*4+seqi[i-5])*4+seqi[i-6]
counts[index] -= 1
assert counts.sum() == 0
def test_count_kmers_valid(self):
"""
Test if calling count_kmers with valid k
returns right result
:return: None
"""
assert kmers.count_kmers("ATTTGGATT", 4) == 6
def test_non_acgt():
count_kmers(1, 'abcdefghijklmnopqrstvwuxyz')
def test_kmers_lookup():
seq = 'ACGTCGACCGCGTTA' * 100
counts = count_kmers(2, seq)
lookup = create_kmers_lookup(2, seq)
lcounts = count_kmers_lookup(lookup, 0, len(seq))
assert (counts == lcounts).all()
#!/usr/bin/env python
from pyfasta import Fasta
from tsh.bio import regions_reader
import sys
from kmers import count_kmers, all_kmers
if __name__ == '__main__':
genome = Fasta(sys.argv[1])
k = int(sys.argv[2])
kmers = 0
for region in regions_reader(sys.argv[3]):
seq = genome[region.chrom][region.start:region.stop]
seq = str(seq).upper()
for s in seq.split('N'):
kmers += count_kmers(k, s)
for kmer, count in zip(all_kmers(k), kmers):
print '%s\t%d' % (kmer, count)