def align_clone_to_ref(clone, reference):
'''Aligns two pycogent Sequence objects with Smith-Waterman algorithm'''
if not isinstance(clone, SequenceI):
raise ValueError('clone must be a cogent.SequenceI object')
if not isinstance(reference, SequenceI):
raise ValueError('reference must be a cogent.SequenceI object')
aligned = sw_align(clone, reference)
ref_len = len(reference)
clone_matched = str(aligned[0].parseOutGaps()[1])
ref_matched = str(aligned[1].parseOutGaps()[1])
if ref_matched == '': raise AlignmentError('No alignment')
aln = CloneAlignment(aligned)
# find where we are in the reference
aln.first_ref_pos = str(reference).index(ref_matched)
aln.first_clone_pos = str(clone).index(clone_matched)
aln.last_ref_pos = aln.first_ref_pos + len(ref_matched)
aln.last_clone_pos = aln.first_clone_pos + len(clone_matched)
match_len = len(ref_matched)
aln.reference_len = ref_len
aln.is_truncated = not match_len == ref_len
aln.has_gaps = aligned[1].isGapped()
aln.has_mismatches = not aligned[1].canMatch(aligned[0])
aln.Seqs[0].Name = clone.Name
aln.Seqs[1].Name = reference.Name
return aln
def main():
kmers = ["TGTAT", "CGTAT", "TTAGT", "TCTAT", "TCTAC"]
motif_clust = MotifCluster(kmers)
#result = motif_clust.cluster_by_seq()
#print "Result: ", result
data = np.transpose(np.array(kmers))
print "DATA: "
print data
dist_func = lambda x, y: sw_align(x, y, return_score=True)
linkage_method = "average"
hclust = clustering.hierarchical_clust(np.array(kmers), dist_func,
linkage_method)
def main():
kmers = ["TGTAT", "CGTAT", "TTAGT", "TCTAT", "TCTAC"]
motif_clust = MotifCluster(kmers)
#result = motif_clust.cluster_by_seq()
#print "Result: ", result
data = np.transpose(np.array(kmers))
print "DATA: "
print data
dist_func = lambda x, y: sw_align(x, y, return_score=True)
linkage_method = "average"
hclust = clustering.hierarchical_clust(np.array(kmers),
dist_func,
linkage_method)
def cluster_by_sw(self):
"""
Cluster sequences pairwise by Smith-Waterman alignment.
Returns distance matrix.
"""
# Make pdist matrix with ij entry corresponding
# to alignment between sequence i and sequence j
score_matrix = []
for kmer_i in self.kmers:
score_row = []
for kmer_j in self.kmers:
alignment = sw_align(kmer_i, kmer_j, return_score=True)
sw_score = alignment[1]
score_row.append(sw_score)
score_matrix.append(score_row)
score_matrix = np.array(score_matrix)
return score_matrix
def cluster_by_sw(self):
"""
Cluster sequences pairwise by Smith-Waterman alignment.
Returns distance matrix.
"""
# Make pdist matrix with ij entry corresponding
# to alignment between sequence i and sequence j
score_matrix = []
for kmer_i in self.kmers:
score_row = []
for kmer_j in self.kmers:
alignment = sw_align(kmer_i, kmer_j,
return_score=True)
sw_score = alignment[1]
score_row.append(sw_score)
score_matrix.append(score_row)
score_matrix = np.array(score_matrix)
return score_matrix
def test_sw_align(self):
"""Tests for sw_align function."""
(first, second), score = sw_align('ACGU', 'CAGU', return_score=True)
self.assertEqual(first, 'GU')
self.assertEqual(second, 'GU')
self.assertEqual(score, 2)
def test_sw_align_empty(self):
"""Tests for sw_align function."""
(first, second), score = sw_align('', '', return_score=True)
self.assertEqual(first, '')
self.assertEqual(second, '')
self.assertEqual(score, 0)
def test_sw_align(self):
"""Tests for sw_align function."""
(first, second), score = sw_align("ACGU", "CAGU", return_score=True)
self.assertEqual(first, "GU")
self.assertEqual(second, "GU")
self.assertEqual(score, 2)
def test_sw_align_empty(self):
"""Tests for sw_align function."""
(first, second), score = sw_align("", "", return_score=True)
self.assertEqual(first, "")
self.assertEqual(second, "")
self.assertEqual(score, 0)
def test_sw_align(self):
"""Tests for sw_align function."""
(first,second),score = sw_align('ACGU','CAGU',return_score=True)
self.assertEqual(first,'GU')
self.assertEqual(second,'GU')
self.assertEqual(score,2)
