def checkDiFrequency(dataSeqsFN, simSeqsFN, outFN):
dataSeqs = cgDL.listFromColumns(dataSeqsFN, [0], ['string'])
simSeqs = cgDL.listFromColumns(simSeqsFN, [0], ['string'])
def returnDiFreq(seqs):
#collect di count
di_freq = {}
totalFrames = 0.0
for seq in seqs:
dis = bioLibCG.returnFrames(seq, 2)
totalFrames += len(dis)
for di in dis:
di_freq[di] = di_freq.get(di, 0) + 1.0
#convert to frequencey
di_freq = dict((x, di_freq[x] / totalFrames) for x in di_freq)
return di_freq
data_freqs = returnDiFreq(dataSeqs)
sim_freqs = returnDiFreq(simSeqs)
with open(outFN, 'w') as f:
for di in data_freqs:
f.write('%s\t%s\t%s\n' %
(di, data_freqs.get(di, 0.0), sim_freqs.get(di, 0.0)))
def checkDiFrequency(dataSeqsFN, simSeqsFN, outFN):
dataSeqs = cgDL.listFromColumns(dataSeqsFN, [0], ['string'])
simSeqs = cgDL.listFromColumns(simSeqsFN, [0], ['string'])
def returnDiFreq(seqs):
#collect di count
di_freq = {}
totalFrames = 0.0
for seq in seqs:
dis = bioLibCG.returnFrames(seq, 2)
totalFrames += len(dis)
for di in dis:
di_freq[di] = di_freq.get(di, 0) + 1.0
#convert to frequencey
di_freq = dict( (x, di_freq[x]/totalFrames) for x in di_freq)
return di_freq
data_freqs = returnDiFreq(dataSeqs)
sim_freqs = returnDiFreq(simSeqs)
with open(outFN, 'w') as f:
for di in data_freqs:
f.write('%s\t%s\t%s\n' % (di, data_freqs.get(di, 0.0), sim_freqs.get(di, 0.0)))
def plotTotalSNR(fN):
fig = PLT.figure()
ax1 = fig.add_subplot(111, projection='3d')
Xs, Ys, dZs = cgDL.listFromColumns(fN, [0,1,2], ['float', 'float', 'float'], naToZero = True)
#xpos = [1,2,3]
#ypos = [1,2,3]
#zpos = [0,0,0]
Zs = [0] * len(dZs)
dZs = [1 if x == 0.0 else x for x in dZs]
dx = dy = [.2] * len(Zs)
ax1.bar3d(Xs, Ys, Zs, dx, dy, dZs, color='#8E4585', zsort = 'max')
PLT.show()
def getMicroHistExpression(microFN, fqFile, outFN):
microSeqs = cgDL.listFromColumns(microFN, [0], ['string'])
microSeq_count = dict( (seq, 0) for seq in microSeqs)
f = open(fqFile, 'r')
for line in f:
possibleSeq = line.strip()
if possibleSeq in microSeq_count:
microSeq_count[possibleSeq] += 1
f.close()
with open(outFN, 'w') as f:
for seq, count in microSeq_count.iteritems():
f.write('%s\t%s\n' % (seq, count))
def checkMaskEnds(maskPerLineFN):
masks = cgDL.listFromColumns(maskPerLineFN, [0], ['string'])
index_numMM = dict((i, 0) for i in range(10))
for mask in masks:
mask = mask[::-1]
for i, char in enumerate(mask):
if i == 10: break
if char == 'X':
index_numMM[i] += 1
for i, num in index_numMM.items():
print i, num
def getMicroHistExpression(microFN, fqFile, outFN):
microSeqs = cgDL.listFromColumns(microFN, [0], ['string'])
microSeq_count = dict((seq, 0) for seq in microSeqs)
f = open(fqFile, 'r')
for line in f:
possibleSeq = line.strip()
if possibleSeq in microSeq_count:
microSeq_count[possibleSeq] += 1
f.close()
with open(outFN, 'w') as f:
for seq, count in microSeq_count.iteritems():
f.write('%s\t%s\n' % (seq, count))
def checkMaskEnds(maskPerLineFN):
masks = cgDL.listFromColumns(maskPerLineFN, [0], ['string'])
index_numMM = dict((i,0) for i in range(10))
for mask in masks:
mask = mask[::-1]
for i, char in enumerate(mask):
if i == 10: break
if char == 'X':
index_numMM[i] += 1
for i, num in index_numMM.items():
print i, num
def grepBC(grepList, inFile, column, word=False):
grepList = cgDL.listFromColumns(grepList, [column], ["int"])
f = open(inFile, "r")
for line in f:
ls = line.strip().split("\t")
for w in grepList:
if word:
if w == ls[column]:
print line,
else:
if w in ls[column]:
print line,
f.close()
def plotTotalSNR(fN):
fig = PLT.figure()
ax1 = fig.add_subplot(111, projection='3d')
Xs, Ys, dZs = cgDL.listFromColumns(fN, [0, 1, 2],
['float', 'float', 'float'],
naToZero=True)
#xpos = [1,2,3]
#ypos = [1,2,3]
#zpos = [0,0,0]
Zs = [0] * len(dZs)
dZs = [1 if x == 0.0 else x for x in dZs]
dx = dy = [.2] * len(Zs)
ax1.bar3d(Xs, Ys, Zs, dx, dy, dZs, color='#8E4585', zsort='max')
PLT.show()
def generateLeftRightChimera(oRNAFN, outFNBase):
'''make map of 4mer --> left, 4mer rights
NOTE: left and right do not include middle!'''
#FN is id/sequence
sequences = cgDL.listFromColumns(oRNAFN, [1], ['string'])
#update left/rights
fourMer_left = {}
fourMer_right = {}
for seq in sequences:
middle = seq[9:13]
left, right = seq[:9], seq[13:]
fourMer_left.setdefault(middle, set()).add(left)
fourMer_right.setdefault(middle, set()).add(right)
#lefts and rights mapped by middle 4
for side in ['left', 'right']:
outDict = eval('fourMer_%s' % side) #uh oh! EVAL TIME!
with open(outFNBase + '.' + side, 'w') as f:
for mer, seqs in outDict.iteritems():
print mer, seqs
f.write('%s\t%s\n' % (mer, ','.join(seqs)))
def maskAll(fN):
allSeqs = cgDL.listFromColumns(fN, [2], ['string'])
allMasks = [multiMask(x, list(), True) for x in allSeqs]