def find_distance(sequence_dictionary):
## Takes dictionary generated by readfasta and calculates distance matrix###
popped_list = []
p_distance = 0
evol_distance = 0
distance_matrix = {}
copied_dict = sequence_dictionary.copy(
) ###Making a copy of sequence_dictionary##
for key in copied_dict:
if key not in popped_list:
seq1 = copied_dict[key]
popped_list.append(key)
for i in sequence_dictionary:
seq2 = sequence_dictionary[i]
s = alignment_util.readScoringMatrix(
"/home/kumara3/CSE620K/proj6/Blosum62.txt")
output_key_i = alignment.SmithWatermanAffine(
seq1, seq2, s, 7, 1
) ###SmithWatermanAffine for calculating the alignment between pair of sequences###
output = [it
for it in output_key_i] ## alignment stored in a list##
count_m = 0 ## count for number of matches in sequences##
count_mis = 0 ## count for number of mismatches in sequence##
X = [x for x in output[1:2]] ## sequence 1 from alignment
Y = [y for y in output[2:3]] ## sequence 2 from alignment
X_string = ''.join(X)
Y_string = ''.join(Y)
size = max(len(X_string), len(Y_string))
for k in range(0, size):
if X_string[k] == Y_string[k]:
count_m += 1
elif X_string[k] != Y_string[k]:
count_mis += 1
else:
if X_string[k] == '-' or Y_string[k] == '-':
print "IGNORE"
p_distance = count_mis / float(
count_m + count_mis) ## calculating the p distance
evol_distance = -0.75 * log(
float(1 - 4 * p_distance /
3)) ## calculating distance for evolutionary matrix
distance_matrix.setdefault(key, {}).setdefault(
i, evol_distance) ## creating and filing the distance matrix
if key == i:
distance_matrix[key][i] = 0
return distance_matrix
def test1(self):
s1 = "AATTA"
s2 = "AAA"
S = self.S3
o = 4
c = 1
correct_score = 24
s, a1, a2 = alignment.SmithWatermanAffine(s1, s2, S, o, c)
self.assertEqual(
alignment_util.scoreAlignmentAffine(a1, a2, S, o, c), s,
"Returned alignment score does not match actual alignment score")
self.assertEqual(alignment_util.scoreAlignmentAffine(a1, a2, S, o,
c), correct_score,
"Score of returned alignment is not optimal")
s, a1, a2 = alignment.SmithWatermanAffine(s2, s1, S, o, c)
self.assertEqual(
alignment_util.scoreAlignmentAffine(a1, a2, S, o, c), s,
"Returned alignment score does not match actual alignment score")
self.assertEqual(alignment_util.scoreAlignmentAffine(a1, a2, S, o,
c), correct_score,
"Score of returned alignment is not optimal")
def mcalliTest(self):
s1 = 'CCCCCCCCCAAAAAAAAAAAAAAAAAAATAAACCCCCCC'
s2 = 'CCCCCCCCCTTTTTTTTTTTTTTTTTTTTTTTCCCCCCC'
S = {x: {y: 100 * self.S2[x][y] for y in "ACGT"} for x in "ACGT"}
o = 500
c = 1
correct_score = 5462
s, a1, a2 = alignment.SmithWatermanAffine(s1, s2, S, o, c)
self.assertEqual(
alignment_util.scoreAlignmentAffine(a1, a2, S, o, c), s,
"Returned alignment score does not match actual alignment")
self.assertEqual(alignment_util.scoreAlignmentAffine(a1, a2, S, o,
c), correct_score,
"Score of returned alignment is not optimal")
t, b1, b2 = alignment.SmithWatermanAffine(s2, s1, S, o, c)
self.assertEqual(
alignment_util.scoreAlignmentAffine(b1, b2, S, o, c), t,
"Returned alignment score does not match actual alignment")
self.assertEqual(alignment_util.scoreAlignmentAffine(b1, b2, S, o,
c), correct_score,
"Score of returned alignment is not optimal")
def test7(self):
s1 = "CCCAAAAATTTAAAAACCCCCGGG"
s2 = "GGGAAAAAAAAAATTTCCCCCCCC"
S = self.S3
o = 5
c = 2
correct_score = 128
s, a1, a2 = alignment.SmithWatermanAffine(s1, s2, S, o, c)
self.assertEqual(
alignment_util.scoreAlignmentAffine(a1, a2, S, o, c), s,
"Returned alignment score does not match actual alignment score")
self.assertEqual(alignment_util.scoreAlignmentAffine(a1, a2, S, o,
c), correct_score,
"Score of returned alignment is not optimal")
def test_SWA(seq1, seq2, S, o, c):
s, a1, a2 = alignment.SmithWatermanAffine(seq1, seq2, S, o, c)
s_sol, a1_sol, a2_sol = alignment_sol.SmithWatermanAffine(
seq1, seq2, S, o, c)
score = 0
# First: test that the function has returned has an optimal alignment
if alignment_util.scoreAlignmentAffine(a1, a2, S, o, c) == s_sol:
score += 70
# Second: test that the function has returned an optimal score
if s == s_sol:
score += 20
# Third: Test that the function has returned the correct score for the alignment
if alignment_util.scoreAlignmentAffine(a1, a2, S, o, c) == s:
score += 10
return score / 100.0