def global_alignment_v2(v,
w,
scoring_matrix=None,
sigma=None,
matches=None,
mismatches=None):
# first, build the graph string
# print("start")
adj_list, weights, sink = two_strings_to_weighted_graph(
v, w, scoring_matrix, sigma, matches, mismatches)
# print("adj_list")
cost, path = longest_path_in_dag((0, 0), sink, adj_list, weights)
# print(path)
for index, n in enumerate(path):
i = n[0]
j = n[1]
if index < len(path) - 1:
next = path[index + 1]
k = next[0]
l = next[1]
if k == (i + 1) and l == j: # if the next is horizontal
v = v[:i] + "-" + v[i:]
elif k == i and l == (j + 1): # if the next is vertical
w = w[:j] + "-" + w[j:]
# elif k == (i + 1) and l == (j + 1): # if the next is diagonal
return cost, v, w
def alignment_with_affine_gap_penalty_v2(v,
w,
scoring_matrix=None,
sigma=None,
matches=None,
mismatches=None,
epsilon=None):
adj_list, weights, sink = two_strings_to_weighted_graph(v,
w,
scoring_matrix,
sigma,
matches,
mismatches,
epsilon=epsilon)
# print("adj_list")
cost, path = longest_path_in_dag((0, 0), sink, adj_list, weights)
# print(path)
for index, n in enumerate(path):
i = n[0]
j = n[1]
if index < len(path) - 1:
next = path[index + 1]
k = next[0]
l = next[1]
if k > i and l == j: # if the next is horizontal
v = v[:i] + "-" * (k - i) + v[i:]
elif k == i and l > j: # if the next is vertical
w = w[:j] + "-" * (l - j) + w[j:]
# elif k == (i + 1) and l == (j + 1): # if the next is diagonal
return cost, v, w
def overlap_alignment_v2(v,
w,
scoring_matrix=None,
sigma=None,
matches=None,
mismatches=None):
# print("start")
adj_list, weights, sink = two_strings_to_weighted_graph(v,
w,
scoring_matrix,
sigma,
matches,
mismatches,
is_overlap=True)
# print(adj_list)
# print(weights)
cost, path = longest_path_in_dag((0, 0),
sink,
adj_list,
weights,
is_overlap=True)
print(path)
print(v, w)
sink = path[-1]
start = False
cut = 0
for index, n in enumerate(path):
i = n[0]
j = n[1]
if index < len(path) - 1:
next = path[index + 1]
k = next[0]
l = next[1]
if k == (i + 1) and l == j: # if the next is horizontal
v = v[:j] + "-" + v[j:]
start = True
elif k == i and l == (j + 1): # if the next is vertical
if not start:
cut += 1
else:
w = w[:i] + "-" + w[i:]
start = True
else:
start = True
# print(start)
v = v[cut:]
# w = w[:sink[0]]
w = w[:len(v)]
return cost, v, w
def fitting_alignment_v2(v,
w,
scoring_matrix=None,
sigma=None,
matches=None,
mismatches=None):
"""
correct score and correct right trim .. I need to trim the left side and fix the - added
"""
# print("start")
adj_list, weights, sink = two_strings_to_weighted_graph(v,
w,
scoring_matrix,
sigma,
matches,
mismatches,
is_fitted=True)
# print(adj_list)
# print(weights)
cost, path = longest_path_in_dag((0, 0),
sink,
adj_list,
weights,
is_fitting=True)
start = False
cut = 0
for index, n in enumerate(path):
i = n[0]
j = n[1]
if index < len(path) - 1:
next = path[index + 1]
k = next[0]
l = next[1]
if k == (i + 1) and l == j: # if the next is horizontal
v = v[:j] + "-" + v[j:]
start = True
elif k == i and l == (j + 1): # if the next is vertical
if not start:
# print("cut", v, v[1:])
cut += 1
else:
w = w[:i] + "-" + w[i:]
start = True
else:
start = True
v = v[cut:]
v = v[:len(w)]
return cost, v, w
def longest_common_sequence_v2(v, w):
# first, build the graph string
# print("start")
adj_list, weights, sink = two_strings_to_weighted_graph(v, w)
# print("got adj string")
path = longest_path_in_dag((0, 0), sink, adj_list, weights)[1]
# print("got path")
sequence = ""
for index, n in enumerate(path):
i = n[0]
j = n[1]
if index < len(path) - 1: # if the next node is diagonal
next = path[index + 1]
k = next[0]
l = next[1]
if k == (i + 1) and l == (j + 1) and weights[(i, j), (k, l)] == 1:
sequence += v[j]
return sequence
def local_alignment_v2(v,
w,
scoring_matrix=None,
sigma=None,
matches=None,
mismatches=None):
# print("start")
adj_list, weights, sink = two_strings_to_weighted_graph(v,
w,
scoring_matrix,
sigma,
matches,
mismatches,
is_local=True)
# print(adj_list)
# print(weights)
cost, path = longest_path_in_dag((0, 0),
sink,
adj_list,
weights,
is_local=True)
# print(path)
sink = path[-1]
v = v[:sink[1]]
w = w[:sink[0]]
start = None
for index, n in enumerate(path):
i = n[0]
j = n[1]
if index < len(path) - 1:
next = path[index + 1]
k = next[0]
l = next[1]
if weights[((i, j), (k, l))] != 0:
if k == (i + 1) and l == j: # if the next is horizontal
v = v[:i] + "-" + v[i:]
elif k == i and l == (j + 1): # if the next is vertical
w = w[:j] + "-" + w[j:]
if not start:
start = (i, j)
v = v[start[1]:]
w = w[start[0]:]
return cost, v, w
def edit_distance_v2(v, w):
# first, build the graph string
# print("start")
adj_list, weights, sink = two_strings_to_weighted_graph(v, w)
# print("got adj string")
path = longest_path_in_dag((0, 0), sink, adj_list, weights)[1]
# print("got path")
count = 0
for index, n in enumerate(path):
i = n[0]
j = n[1]
if index < len(path) - 1: # if the next node is diagonal
next = path[index + 1]
k = next[0]
l = next[1]
if k == (i + 1) and l == j: # if the next is horizontal
count += 1
elif k == i and l == (j + 1): # if the next is vertical
count += 1
elif k == (i + 1) and l == (j + 1) and weights[(i, j),
(k, l)] != 1:
count += 1
return count