def updatePolySeqs(mFN, readsFN, alignFN):
tim = bioLibCG.cgTimer()
tim.start()
variousAs = ["A" * x for x in range(1,20)]
variousGs = ["G" * x for x in range(1,20)]
variousTs = ["T" * x for x in range(1,20)]
variousCs = ["C" * x for x in range(1,20)]
letter_variousLetters = [ ("A", variousAs),
("G", variousGs),
("T", variousTs),
("C", variousCs)]
checkRange = range(1,8)
NX = cgNexusFlat.Nexus(mFN, miR)
NX.load(['sequence', 'polySeqs'])
#print 'load micro', tim.split()
reads = cgNexusFlat.quickTable(('read','string', '.', 1))
rNX = cgNexusFlat.Nexus(readsFN, reads)
rNX.load(['read'])
#print 'load reads', tim.split()
aNX = cgNexusFlat.Nexus(alignFN, cgAlignment)
aNX.load(['sID', 'tID'])
#print 'load alignments', tim.split()
for id in aNX.ids:
theRead = rNX.read[aNX.sID[id]]
mID = aNX.tID[id]
microSeq = NX.sequence[mID]
#may be a read for expression, but wont count...
if theRead in microSeq: continue
#just for expression
if microSeq == theRead:
print tabIt(microSeq, theRead, 0, 0, "N")
#first check full
elif microSeq in theRead and (len(theRead) != len(microSeq)):
tail = theRead.split(microSeq)[1]
for let, variousLetters in letter_variousLetters:
if tail in variousLetters:
print tabIt(microSeq, theRead, 0, len(tail), let)
#now check trimmed (cant do [:-0])
else:
for i in checkRange:
if microSeq[:-i] in theRead and (len(theRead) != len(microSeq[:-i])):
tail = theRead.split(microSeq[:-i])[1]
for let, variousLetters in letter_variousLetters:
if tail in variousLetters:
print tabIt(microSeq, theRead, i, len(tail), let)
print "TRIMMED"
break #dont trim after the first trimmed one works
def plotBar(analysisFN):
polyTable = cgNexusFlat.quickTable(
('microSeq', 'string', '.', 0), ('readSeq', 'string', '.', 1),
('trimAmount', 'int', 0, 2), ('tailLength', 'int', 0, 3),
('letter', 'string', 'N', 4))
polyNX = cgNexusFlat.Nexus(analysisFN, polyTable, ids=False)
polyNX.load(['trimAmount', 'tailLength', 'letter'])
letter_trim_tLen_count = {}
for id in polyNX.ids:
let = polyNX.letter[id]
trim = polyNX.trimAmount[id]
tLen = polyNX.tailLength[id]
prevCount = letter_trim_tLen_count.setdefault(let, {}).setdefault(
trim, {}).setdefault(tLen, 0)
letter_trim_tLen_count[let][trim][tLen] = prevCount + 1
letters = ["A", "T", "C", "G"]
ind = range(1, 5)
width = 0.35 # the width of the bars
for trim in range(0, 8):
for tLen in range(0, 7):
plotNum = (trim * 7) + tLen
plt.subplot(8, 7, plotNum)
letAmount = []
for let in letters:
letAmount.append(letter_trim_tLen_count[let].setdefault(
trim, {}).setdefault(tLen, 0))
#normalize
theMax = max(letAmount)
if theMax == 0:
letAmount = [0, 0, 0, 0]
else:
letAmount = [float(x + .01) / theMax for x in letAmount]
plt.bar(ind, letAmount, width)
plt.title('%s,%s %s' % (trim, tLen, theMax))
plt.xticks([], [''])
plt.yticks([], [''])
plt.show()
def plotBar(analysisFN):
polyTable = cgNexusFlat.quickTable(('microSeq', 'string', '.', 0),
('readSeq', 'string', '.', 1),
('trimAmount', 'int', 0, 2),
('tailLength', 'int', 0, 3),
('letter', 'string', 'N', 4))
polyNX = cgNexusFlat.Nexus(analysisFN, polyTable, ids = False)
polyNX.load(['trimAmount', 'tailLength', 'letter'])
letter_trim_tLen_count = {}
for id in polyNX.ids:
let = polyNX.letter[id]
trim = polyNX.trimAmount[id]
tLen = polyNX.tailLength[id]
prevCount = letter_trim_tLen_count.setdefault(let, {}).setdefault(trim, {}).setdefault(tLen, 0)
letter_trim_tLen_count[let][trim][tLen] = prevCount + 1
letters = ["A", "T", "C", "G"]
ind = range(1,5)
width = 0.35 # the width of the bars
for trim in range(0, 8):
for tLen in range(0,7):
plotNum = (trim * 7) + tLen
plt.subplot(8,7, plotNum)
letAmount = []
for let in letters:
letAmount.append(letter_trim_tLen_count[let].setdefault(trim, {}).setdefault(tLen, 0))
#normalize
theMax = max(letAmount)
if theMax == 0:
letAmount = [0,0,0,0]
else:
letAmount = [float(x+ .01)/theMax for x in letAmount]
plt.bar(ind, letAmount, width)
plt.title('%s,%s %s' % (trim, tLen, theMax) )
plt.xticks([], [''])
plt.yticks([], [''])
plt.show()