def question1():
human = read_protein(HUMAN_EYELESS_URL)
fly = read_protein(FRUITFLY_EYELESS_URL)
print(len(human), len(fly))
scoring = read_scoring_matrix(PAM50_URL)
local_align = compute_alignment_matrix(human, fly, scoring, False)
score, xs, ys = compute_local_alignment(human, fly, scoring, local_align)
print('Question 1')
print(score)
print(xs)
print(ys)
print()
print('Question 2')
consensus = read_protein(CONSENSUS_PAX_URL)
human_nodash = ''.join([x for x in xs if x != '-'])
fly_nodash = ''.join([x for x in ys if x != '-'])
hc_global_align = compute_alignment_matrix(human_nodash, consensus, scoring, True)
fc_global_align = compute_alignment_matrix(fly_nodash, consensus, scoring, True)
hc_agree = agreement(human_nodash, consensus, scoring, hc_global_align)
fc_agree = agreement(fly_nodash, consensus, scoring, fc_global_align)
print('Human vs Consensus agree = %s%%' % hc_agree)
print('Fly vs Consensus agree = %s%%' % fc_agree)
def question1():
human = read_protein(HUMAN_EYELESS_URL)
fly = read_protein(FRUITFLY_EYELESS_URL)
print(len(human), len(fly))
scoring = read_scoring_matrix(PAM50_URL)
local_align = compute_alignment_matrix(human, fly, scoring, False)
score, xs, ys = compute_local_alignment(human, fly, scoring, local_align)
print('Question 1')
print(score)
print(xs)
print(ys)
print()
print('Question 2')
consensus = read_protein(CONSENSUS_PAX_URL)
human_nodash = ''.join([x for x in xs if x != '-'])
fly_nodash = ''.join([x for x in ys if x != '-'])
hc_global_align = compute_alignment_matrix(human_nodash, consensus,
scoring, True)
fc_global_align = compute_alignment_matrix(fly_nodash, consensus, scoring,
True)
hc_agree = agreement(human_nodash, consensus, scoring, hc_global_align)
fc_agree = agreement(fly_nodash, consensus, scoring, fc_global_align)
print('Human vs Consensus agree = %s%%' % hc_agree)
print('Fly vs Consensus agree = %s%%' % fc_agree)
def generate_null_distribution(seq_x, seq_y, scoring_matrix, num_trials):
distr = {}
raw = []
try:
with open('distr.json') as f:
pair = loads(f.read())
return pair['distr'], pair['raw']
except Exception as e:
print('cant open file', str(e))
for _ in range(num_trials):
temp = list(seq_y)
shuffle(temp)
rand_y = ''.join(temp)
align = compute_alignment_matrix(seq_x, rand_y, scoring_matrix, False)
score, _, _ = compute_local_alignment(seq_x, rand_y, scoring_matrix,
align)
if score not in distr:
distr[score] = 0
distr[score] += 1
raw.append(score)
with open('distr.json', 'w') as f:
f.write(dumps({'distr': distr, 'raw': raw}))
return distr, raw
def generate_null_distribution(seq_x, seq_y, scoring_matrix, num_trials):
distr = {}
raw = []
try:
with open('distr.json') as f:
pair = loads(f.read())
return pair['distr'], pair['raw']
except Exception as e:
print('cant open file', str(e))
for _ in range(num_trials):
temp = list(seq_y)
shuffle(temp)
rand_y = ''.join(temp)
align = compute_alignment_matrix(seq_x, rand_y, scoring_matrix, False)
score, _, _ = compute_local_alignment(seq_x, rand_y, scoring_matrix, align)
if score not in distr:
distr[score] = 0
distr[score] += 1
raw.append(score)
with open('distr.json', 'w') as f:
f.write(dumps({'distr': distr, 'raw': raw}))
return distr, raw
def edit_dist(xs, ys):
alphabet = ascii_lowercase
scoring = build_scoring_matrix(alphabet, 2, 1, 0)
align = compute_alignment_matrix(xs, ys, scoring, True)
score, x, y = compute_global_alignment(xs, ys, scoring, align)
return len(xs) + len(ys) - score
def edit_dist(xs, ys):
alphabet = ascii_lowercase
scoring = build_scoring_matrix(alphabet, 2, 1, 0)
align = compute_alignment_matrix(xs, ys, scoring, True)
score, x, y = compute_global_alignment(xs, ys, scoring, align)
return len(xs) + len(ys) - score