def FilterAlignedPairForPositions(seq1, seq2, method):
"""given the method, return a set of aligned sequences
only containing certain positions.
Available filters:
all: do nothing.
codon1,codon2,codon3: return 1st, 2nd, 3rd codon positions only.
d4: only changes within fourfold-degenerate sites
"""
l1 = len(seq1)
l2 = len(seq2)
if method == "all":
return seq1, seq2
elif method == "codon1":
return ("".join([seq1[x] for x in range(0, l1, 3)]),
"".join([seq2[x] for x in range(0, l2, 3)]))
elif method == "codon2":
return ("".join([seq1[x] for x in range(1, l1, 3)]),
"".join([seq2[x] for x in range(1, l2, 3)]))
elif method == "codon3":
return ("".join([seq1[x] for x in range(2, l1, 3)]),
"".join([seq2[x] for x in range(2, l2, 3)]))
elif method == "d4":
s1 = []
s2 = []
for x in range(0, l1, 3):
codon1 = seq1[x:x + 3]
codon2 = seq2[x:x + 3]
try:
aa1, deg11, deg12, deg13 = Genomics.GetDegeneracy(codon1)
aa2, deg11, deg22, deg23 = Genomics.GetDegeneracy(codon2)
except KeyError:
continue
if aa1 == aa2 and deg13 == 4 and deg23 == 4:
s1.append(codon1[2])
s2.append(codon2[2])
return "".join(s1), "".join(s2)
def loadSequence(self, sequence, seqtype="na"):
"""load sequence properties from a sequence."""
SequencePropertiesLength.loadSequence(self, sequence, seqtype)
if len(sequence) % 3:
raise ValueError(
'''sequence length is not a multiple of 3 (length=%i)''' %
(len(sequence)))
# uppercase all letters
sequence = sequence.upper()
self.mNStopCodons = 0
# setup counting arrays
# nucleotide counts for each position (is not a sum of the counts
# per degenerate site, as the codon might be intelligible, e.g. GNN).
self.mCounts = [{'A': 0, 'C': 0, 'G': 0, 'T': 0, 'X': 0, 'N': 0},
{'A': 0, 'C': 0, 'G': 0, 'T': 0, 'X': 0, 'N': 0},
{'A': 0, 'C': 0, 'G': 0, 'T': 0, 'X': 0, 'N': 0}]
# nucleotide counts for each position per degeneracy
self.mCountsDegeneracy = []
for x in (0, 1, 2):
xx = []
for y in range(5):
yy = {}
for z in Bio.Alphabet.IUPAC.extended_dna.letters:
yy[z] = 0
xx.append(yy)
self.mCountsDegeneracy.append(xx)
# use generator rather than list to save memory
for codon in (sequence[x:x + 3] for x in xrange(0, len(sequence), 3)):
for x in (0, 1, 2):
self.mCounts[x][codon[x]] += 1
if Genomics.IsStopCodon(codon):
self.mNStopCodons += 1
continue
try:
aa, deg1, deg2, deg3 = Genomics.GetDegeneracy(codon)
degrees = (deg1, deg2, deg3)
for x in range(len(degrees)):
self.mCountsDegeneracy[x][degrees[x]][codon[x]] += 1
except KeyError:
pass
def countSites(model):
"""count number of expected synonymous/nonsynonymous sites in a grammar.
"""
## number of synonymous/non-synonymous sites
n, s = 0.0, 0.0
xpi = model.evaluateTerminalFrequencies()[('COD0', 'COD1', 'COD2')]
pi = {}
for codon, f in xpi.items():
pi["".join(codon).upper()] = f
## translate pi and the matrix to codons
for key, value in pi.items():
del pi[key]
pi["".join(key).upper()] = value
for codon, freq in pi.items():
try:
degeneracy = Genomics.GetDegeneracy(codon)
except KeyError:
continue
for x in range(1, 4):
d = (degeneracy[x] - 1.0) / 3.0
s += freq * d
n += freq * (1.0 - d)
## if degeneracy[x] > 1:
## s += freq
## else:
## n += freq
assert (float("%5.2f" % (n + s)) == 3.0)
## print s / (n+s)
## n = 184.9
## s = 76.1
## t = n + s
## n /= t
## s /= t
## print s / (n+s)
return n, s
def Load(self, in_sequence):
"""load sequence properties from a sequence."""
## uppercase all letters
sequence = in_sequence.upper()
self.mNCodons = len(sequence) / 3
self.mNStopCodons = 0
## setup counting arrays
## counts of amino acids
self.mCountsAA = {}
for x in Bio.Alphabet.IUPAC.extended_protein.letters:
self.mCountsAA[x] = 0
## nucleotide counts for each position (is not a sum of the counts
## per degenerate site, as the codon might be intelligible, e.g. GNN).
self.mCounts = [{
'A': 0,
'C': 0,
'G': 0,
'T': 0,
'X': 0,
'N': 0
}, {
'A': 0,
'C': 0,
'G': 0,
'T': 0,
'X': 0,
'N': 0
}, {
'A': 0,
'C': 0,
'G': 0,
'T': 0,
'X': 0,
'N': 0
}]
## nucleotide counts for each position per degeneracy
self.mCountsDegeneracy = []
self.mLength = len(sequence)
for x in (0, 1, 2):
xx = []
for y in range(5):
yy = {}
for z in Bio.Alphabet.IUPAC.extended_dna.letters:
yy[z] = 0
xx.append(yy)
self.mCountsDegeneracy.append(xx)
for codon in [sequence[x:x + 3] for x in range(0, len(sequence), 3)]:
for x in (0, 1, 2):
self.mCounts[x][codon[x]] += 1
if Genomics.IsStopCodon(codon):
self.mNStopCodons += 1
continue
try:
aa, deg1, deg2, deg3 = Genomics.GetDegeneracy(codon)
degrees = (deg1, deg2, deg3)
for x in range(len(degrees)):
self.mCountsDegeneracy[x][degrees[x]][codon[x]] += 1
self.mCountsAA[aa] += 1
except KeyError:
pass
self.Update()