def JoinCtgs(smb, f1, f2, id1, id2, submat, abet):
print 'JoinCtgs', f1, f2
bv = dynprog.FastSubMatrix(submat, abet, smb[f1][id1], smb[f2][id2])
scm, ar = dynprog.FastSW(bv, smb[f1][id1], smb[f2][id2])
t1, t2, vgap, hgap = SWBacktrace2(scm, ar, smb[f1][id1], smb[f2][id2])
GapCtgByList(smb[f1], hgap)
GapCtgByList(smb[f2], vgap)
for i in smb[f2].keys():
smb[f1][i] = smb[f2][i]
del (smb[f2])
def InitAligns(ids, seqs, submat, abet):
lines = []
N = len(seqs)
for i in range(N):
for j in range(i):
if i != j:
bv = dynprog.FastSubMatrix(submat, abet, seqs[i], seqs[j])
sc, ar = dynprog.FastSW(bv, seqs[i], seqs[j])
t1, t2, vs, hs = SWBacktrace2(sc, ar, seqs[i], seqs[j])
lines.append((ids[i], ids[j], t1, t2))
return lines
def AddToCtg(smb, fm, idm, idum, ilinei, submat, abet):
print 'AddToCtg', idum, 'added to ', fm
if ilinei[0] == idm:
matseq = ilinei[2] # the seq that is also in the smb
umatseq = ilinei[3]
else:
matseq = ilinei[3] # the seq that is also in the smb
umatseq = ilinei[2]
# align match seq with the same seq in smb
bv = dynprog.FastSubMatrix(submat, abet, smb[fm][idm], matseq)
scm, ar = dynprog.FastSW(bv, smb[fm][idm], matseq)
t1, t2, vgap, hgap = SWBacktrace2(scm, ar, smb[fm][idm], matseq)
# gap contig
GapCtgByList(smb[fm], hgap)
# create and add new sequence
newseq = GapSeqByList(umatseq, vgap)
newseq = LeadTrailGaps(newseq)
smb[fm][idum] = newseq
st2 += '-'
v -= 1
elif arrow[v, h] == 1:
st1 += '-'
st2 += seq2[h - 1]
h -= 1
elif arrow[v, h] == 2:
st1 += seq1[v - 1]
st2 += seq2[h - 1]
v -= 1
h -= 1
if v == 0 and h == 0:
ok = 0
#reverse the strings
st1 = st1[::-1]
st2 = st2[::-1]
return st1, st2
st1, st2 = dpg.Backtrace(arrow, s1, s2)
print(st1)
print(st2)
import blosum
sq1 = 'KMTIFFMILK'
sq2 = 'NQTIFF'
subvals = dpg.FastSubValues(blosum.BLOSUM50, blosum.PBET, sq1, sq2)
scmat, arrow = dpg.FastSW(subvals, sq1, sq2)
t1, t2 = dpg.SWBacktrace(scmat, arrow, sq1, sq2)
print(subvals, scmat, arrow, t1, t2)
# Create a second string with the form YAYBY where # - Y is a different set of random letters # - each Y can have a different length. # Align the sequences using Smith-Waterman. # The scoring matrix will have two major maximum for the alignments of the A and B regions. # Modify the program to extract both alignments. import dynprog as dpg from blosum import BLOSUM50, PBET import random as rd a = ''.join(rd.choices(PBET, k=12)) b = ''.join(rd.choices(PBET, k=12)) Xstr = stringXAXBX(a, b) Ystr = stringXAXBX(a, b) subvals = dpg.FastSubValues(BLOSUM50, PBET, Xstr, Ystr) scmat, arrow = dpg.FastSW(subvals, Xstr, Ystr) t1, t2, = dpg.SWBacktrace(scmat, arrow, Xstr, Ystr) print(t1, t2) # import numpy as np # import random as rd # import string # N=12 # r1 = "abc-ABC" # r2 = "abd-ABd" # r3 = ''.join(rd.choices(string.ascii_uppercase + string.ascii_lowercase, k=N)) # r4 = ''.join(rd.choices(string.ascii_uppercase + string.ascii_lowercase, k=N)) # r5 = ''.join(rd.choices(string.ascii_uppercase + string.ascii_lowercase, k=N)) # r6 = ''.join(rd.choices(string.ascii_uppercase + string.ascii_lowercase, k=N))