class UPGMA:
def __init__(self, seqs, alseq):
self.seqs = seqs
self.alseq = alseq
self.create_mat_dist()
def create_mat_dist(self):
# create distance matrix with dim N x N sequences
self.matdist = NumMatrix(len(self.seqs), len(self.seqs))
for i in range(len(self.seqs)):
for j in range(i, len(self.seqs)):
# retrieve the two sequences to align
s1 = self.seqs[i]
s2 = self.seqs[j]
# align the sequences
self.alseq.needleman_Wunsch(s1, s2)
# recover the alignment
alin = self.alseq.recover_align()
ncd = 0
# fill the matrix
# count the number of different symbols in the alignment as the distance
for k in range(len(alin)):
col = alin.column(k)
if (col[0] != col[1]): ncd += 1
# set distance value in the matrix
self.matdist.set_value(i, j,
ncd) #set for cell i,j the value of ncd
def run(self):
# create an object of the class HierarchicalClustering
ch = HierarchicalClustering(self.matdist)
# execute the clustering algorithm
t = ch.execute_clustering()
return t
def create_mat_dist(self):
self.matdist = NumMatrix(len(self.seqs), len(self.seqs))
for i in range(len(self.seqs)):
for j in range(i, len(self.seqs)):
s1 = self.seqs[i]
s2 = self.seqs[j]
self.alseq.needleman_Wunsch(s1, s2)
alin = self.alseq.recover_align()
ncd = 0
for k in range(len(alin)):
col = alin.column(k)
if (col[0] != col[1]): ncd += 1
self.matdist.set_value(i, j, ncd)
def create_mat_dist(self):
# create distance matrix
self.matdist = NumMatrix(len(self.seqs), len(self.seqs))
for i in range(len(self.seqs)):
for j in range(i, len(self.seqs)):
# retrieve the two sequences to align
s1 = self.seqs[i]
s2 = self.seqs[j]
# align the sequences
self.alseq.needleman_Wunsch(s1, s2)
# recover the alignment
alin = self.alseq.recover_align()
ncd = 0
# fill the matrix
# count the number of different symbols in the alignment as the distance
for k in range(len(alin)):
col = alin.column(k)
if (col[0] != col[1]): ncd += 1
# set distance value in the matrix
self.matdist.set_value(i, j, ncd)
class UPGMA:
def __init__(self, seqs, alseq):
self.seqs = seqs
self.alseq = alseq
self.create_mat_dist()
def create_mat_dist(self):
self.matdist = NumMatrix(len(self.seqs), len(self.seqs))
for i in range(len(self.seqs)):
for j in range(i, len(self.seqs)):
s1 = self.seqs[i]
s2 = self.seqs[j]
self.alseq.needleman_Wunsch(s1, s2)
alin = self.alseq.recover_align()
ncd = 0
for k in range(len(alin)):
col = alin.column(k)
if (col[0] != col[1]): ncd += 1
self.matdist.set_value(i, j, ncd)
def run(self):
ch = HierarchicalClustering(self.matdist)
t = ch.execute_clustering()
return t
def test():
m=NumMatrix(5, 5)
m.set_value(0, 1, 2)
m.set_value(0, 2, 5)
m.set_value(0, 3, 7)
m.set_value(0, 4, 9)
m.set_value(1, 2, 4)
m.set_value(1, 3, 6)
m.set_value(1, 4, 7)
m.set_value(2, 3, 4)
m.set_value(2, 4, 6)
m.set_value(3, 4, 3)
hc=HierarchicalClustering(m)
arv=hc.execute_clustering()
arv.print_tree()
def test():
m = NumMatrix(5,5)