def pacbporflist2similarityarray(pacbps, queryorsbjct, length):
""" """
bea = zeros(length)
for pacbporf in pacbps:
spos = pacbporf._get_original_alignment_pos_start()
epos = pacbporf._get_original_alignment_pos_end()
q, m, s = pacbporf.get_unextended_aligned_protein_sequences()
if queryorsbjct == 'query':
start = spos.query_pos
end = epos.query_pos + 1
seqa = list(q)
ma = list(m)
else:
start = spos.sbjct_pos
end = epos.sbjct_pos + 1
seqa = list(s)
ma = list(m)
for pos in range(len(seqa) - 1, -1, -1):
if seqa[pos] == '-':
seqa.pop(pos)
ma.pop(pos)
# prepare replacement of match string into match score list
matcharray = zeros(end - start)
for pos in range(0, len(ma)):
symbol = ma[pos]
if symbol != ' ':
matcharray[pos] = 1
# update (binary) array
bea[spos.query_pos:epos.query_pos + 1] += matcharray
# correct bea for values > 1
bea = where(greater_equal(bea, 2), 1, bea)
return bea
def pacbporflist2similarityarray(pacbps,queryorsbjct,length):
""" """
bea = zeros(length)
for pacbporf in pacbps:
spos = pacbporf._get_original_alignment_pos_start()
epos = pacbporf._get_original_alignment_pos_end()
q,m,s = pacbporf.get_unextended_aligned_protein_sequences()
if queryorsbjct == 'query':
start = spos.query_pos
end = epos.query_pos + 1
seqa = list(q)
ma = list(m)
else:
start = spos.sbjct_pos
end = epos.sbjct_pos + 1
seqa = list(s)
ma = list(m)
for pos in range(len(seqa)-1,-1,-1):
if seqa[pos] == '-':
seqa.pop(pos)
ma.pop(pos)
# prepare replacement of match string into match score list
matcharray = zeros(end-start)
for pos in range(0,len(ma)):
symbol = ma[pos]
if symbol != ' ':
matcharray[pos] = 1
# update (binary) array
bea[spos.query_pos:epos.query_pos+1] += matcharray
# correct bea for values > 1
bea = where(greater_equal(bea, 2), 1, bea)
return bea
def masked_to_unmasked(mask, remove_mask=False):
"""Returns array mapping indices in orig to indices in ungapped.
Specifically, for each position in orig, returns the index of the position
in the unmasked sequence of the last non-masked character at or before
that index (i.e. if the index corresponds to a masked position, will return
the index of the previous non-masked position since the masked positions
aren't in the unmasked sequence by definition).
If remove_mask is True (the default is False), sets the masked positions
to -1 for easy detection.
"""
result = cumsum(logical_not(mask)) -1
if remove_mask:
result = where(mask, -1, result)
return result
def pca(M):
"Perform PCA on M, return eigenvectors and eigenvalues, sorted."
T, N = shape(M)
# if there are fewer rows T than columns N, use snapshot method
if T < N:
C = dot(M, t(M))
evals, evecsC = eigenvectors(C)
# HACK: make sure evals are all positive
evals = where(evals < 0, 0, evals)
evecs = 1. / sqrt(evals) * dot(t(M), t(evecsC))
else:
# calculate covariance matrix
K = 1. / T * dot(t(M), M)
evals, evecs = eigenvectors(K)
# sort the eigenvalues and eigenvectors, descending order
order = (argsort(evals)[::-1])
evecs = take(evecs, order, 1)
evals = take(evals, order)
return evals, t(evecs)
def pca(M):
from Numeric import take, dot, shape, argsort, where, sqrt, transpose as t
from LinearAlgebra import eigenvectors
"Perform PCA on M, return eigenvectors and eigenvalues, sorted."
T, N = shape(M)
# if there are less rows T than columns N, use
# snapshot method
if T < N:
C = dot(M, t(M))
evals, evecsC = eigenvectors(C)
# HACK: make sure evals are all positive
evals = where(evals < 0, 0, evals)
evecs = 1./sqrt(evals) * dot(t(M), t(evecsC))
else:
# calculate covariance matrix
K = 1./T * dot(t(M), M)
evals, evecs = eigenvectors(K)
# sort the eigenvalues and eigenvectors, decending order
order = (argsort(evals)[::-1])
evecs = take(evecs, order, 1)
evals = take(evals, order)
return evals, t(evecs)
# JS.
binary_denotations = {
',': lambda a, b: b,
'|': operator.or_,
'^': operator.xor,
'&': operator.and_,
'<<': operator.lshift,
'>>': operator.rshift,
'+': operator.add,
'-': operator.sub,
# This isn't correct, but it avoids Numeric's ArithmeticError:
# Integer overflow in multiply.
'*': lambda a, b: (a & (2**15-1)) * (b & (2**15-1)),
# These two have to worry about SIGFPE from division by zero.
'/': lambda a, b: a / where(b == 0, 1, b),
'%': lambda a, b: a % where(b == 0, 1, b),
'&&': lambda a, b: where(a, where(b, 1, 0), 0),
'||': lambda a, b: where(a, 1, where(b, 1, 0)),
'==': lambda a, b: where(a == b, 1, 0),
'!=': lambda a, b: where(a != b, 1, 0),
'<': lambda a, b: where(a < b, 1, 0),
'>': lambda a, b: where(a > b, 1, 0),
'<=': lambda a, b: where(a <= b, 1, 0),
'>=': lambda a, b: where(a >= b, 1, 0),
}
unary_denotations = {
'~': operator.inv,
'-': operator.neg,
def clamp(a, b, c):
"Threshold b between lower limit a and upper limit c."
d = where(a < b, b, a)
return where(d < c, d, c)
# JS.
binary_denotations = {
',': lambda a, b: b,
'|': operator.or_,
'^': operator.xor,
'&': operator.and_,
'<<': operator.lshift,
'>>': operator.rshift,
'+': operator.add,
'-': operator.sub,
# This isn't correct, but it avoids Numeric's ArithmeticError:
# Integer overflow in multiply.
'*': lambda a, b: (a & (2**15 - 1)) * (b & (2**15 - 1)),
# These two have to worry about SIGFPE from division by zero.
'/': lambda a, b: a / where(b == 0, 1, b),
'%': lambda a, b: a % where(b == 0, 1, b),
'&&': lambda a, b: where(a, where(b, 1, 0), 0),
'||': lambda a, b: where(a, 1, where(b, 1, 0)),
'==': lambda a, b: where(a == b, 1, 0),
'!=': lambda a, b: where(a != b, 1, 0),
'<': lambda a, b: where(a < b, 1, 0),
'>': lambda a, b: where(a > b, 1, 0),
'<=': lambda a, b: where(a <= b, 1, 0),
'>=': lambda a, b: where(a >= b, 1, 0),
}
unary_denotations = {
'~': operator.inv,
'-': operator.neg,
'!': lambda x: where(x, 0, 1),
