def getG4(G4DetectedInGene, inputfile, dicoParam):
"""Merges all windows from a gene upper thresholds, it will be predicted G4.
This function browse all windows returned by G4RNA Screener and
keep only those over the thresholds. It also merge overlapping
windows. If two windows upper thresholds are separated with one window
which is under thresholds, the two windows will not be merge and will be
concidered as 2 pG4.
:param G4DetectedInGene: pG4 with its scores and sequence, predicted in
genes.
:type G4DetectedInGene: dictionary
:param inputfile: name of the outputfile of G4RNA screener.
:type inputfile: string
:param dicoParam: all parameters given to G4RNA screener.
:type dicoParam: dictionary
:returns: G4DetectedInGene, contains all predicted G4 in genes,
with there scores.
:rtype: dictionary
"""
oldPassed = False # boolean about the previous windows
passed = False # true if over threshold, elsewise false
inputfile = open(inputfile, 'r')
for line in inputfile:
if (re.search('^[0-9]', line)):
# if the line is not the header of the file
dicoLine = readLineG4ScreenerLoc(line)
if (dicoLine['cGcC'] >= dicoParam['cGcC']
and dicoLine['g4H'] >= dicoParam['g4H']
and dicoLine['g4NN'] >= dicoParam['g4NN']):
# if window over thresholds
passed = True
if (oldPassed != passed):
# first windows over the threshold
descriptionOverThreshold = dicoLine['Description']
# assignation of description for this window
sequenceG4 = dicoLine['WindowSeq']
oldPassed = passed
if dicoLine['Strand'] == '1':
startG4 = dicoLine['WindowStart']
endG4 = dicoLine['WindowEnd']
else:
startG4 = dicoLine['locationEnd'] - \
(dicoLine['WindowStart'] -\
dicoLine['locationStart'])
endG4 = startG4 - (len(dicoLine['WindowSeq'])) + 1
listeCGcC = [dicoLine['cGcC']]
listeG4Hunter = [dicoLine['g4H']]
listeG4NN = [dicoLine['g4NN']]
else:
# not the first windows above the thresholds
sequenceG4 = rF.addWindowToG4Seq(sequenceG4,
dicoLine['WindowSeq'],
dicoParam['Step'],
dicoParam['Window'])
if dicoLine['Strand'] == '1':
endG4 = dicoLine['WindowEnd']
else:
endG4 = startG4 - (len(sequenceG4)) + 1
listeCGcC.append(dicoLine['cGcC'])
listeG4Hunter.append(dicoLine['g4H'])
listeG4NN.append(dicoLine['g4NN'])
if (dicoLine['cGcC'] < dicoParam['cGcC']
or dicoLine['g4H'] < dicoParam['g4H']
or dicoLine['g4NN'] < dicoParam['g4NN']
or descriptionOverThreshold != dicoLine['Description']):
# one of the score is under his threshold
# or this windows is from another gene
passed = False # update
if (oldPassed != passed):
# last windows before under the thresolds
meanCGcC = rF.mean(listeCGcC)
meanG4Hunter = rF.mean(listeG4Hunter)
meanG4NN = rF.mean(listeG4NN)
oldPassed = passed # update
if descriptionOverThreshold != dicoLine['Description']:
headerG4 = rF.createIdpG4rShuffle(
descriptionOverThreshold, startG4, endG4,
dicoLine['Strand'])
else:
headerG4 = rF.createIdpG4rShuffle(
dicoLine['Description'], startG4, endG4,
dicoLine['Strand'])
if headerG4 not in G4DetectedInGene and dicoLine['Strand']:
G4DetectedInGene[headerG4] = [
str(meanCGcC),
str(meanG4Hunter), sequenceG4,
str(meanG4NN)
]
inputfile.close()
return G4DetectedInGene
def ReturnG4InCircu(G4Detected, inputfile, dicoParam):
"""Merges all windows from a gene upper thresholds, it will be predicted G4.
This function browses all windows returned by G4RNA Screener and
keep only those over the thresholds. It also merges overlapping
windows. If two windows upper thresholds are separated with one window
which is under thresholds, the two windows will not be merge and will be
concidered as 2 pG4.
:param G4Detected: pG4r with its scores and sequence, predicted in
genes.
:type G4Detected: dictionary
:param inputfile: name of the outputfile containing the liste of
informations for each circular.
:type inputfile: string
:param dicoParam: all parameters given to G4RNA screener.
:type dicoParam: dictionary
:returns: G4Detected, contains all predicted G4 in genes, with there scores.
:rtype: dictionary
"""
rnaType = 'circu'
location = 'ExonNC'
oldPassed = False
passed = False
overtakingSequence = False
first = False
tmp = []
upstream = ''
downstream = ''
condition = False
descriptionOverThreshold = ''
inputfile = open(inputfile, "r")
for line in inputfile:
if re.search('^[0-9]', line):
words = line.split('\t')
if len(words) == 8:
dL = readLineCircu(words)
if (dL['cGcC'] >= dicoParam['cGcC']
and dL['g4H'] >= dicoParam['g4H']
and dL['g4NN'] >= dicoParam['g4NN']):
passed = True
if (oldPassed != passed):
# if it's the first windows, beginning if the passage
descriptionOverThreshold = dL['Description']
sequenceG4 = dL['Sequence']
oldPassed = passed
if dL['Strand'] == '+':
startG4 = dL['wStart']
# same as startWindow of G4 screener
endG4 = dL['wEnd']
# same as endWindow of G4 screener
else:
startG4 = dL['circuEnd'] - dL['wStart'] + 1
endG4 = startG4 - len(dL['Sequence']) + 1
listeCGcC = [dL['cGcC']]
listeG4Hunter = [dL['g4H']]
listeG4NN = [dL['g4NN']]
else:
# if it's not the first windows above the thresholds
sequenceG4 = rF.addWindowToG4Seq(
sequenceG4, dL['Sequence'], dicoParam['Step'],
dicoParam['Window'])
if dL['Strand'] == '+':
endG4 = dL['wEnd']
else:
endG4 = startG4 - len(sequenceG4) + 1
listeCGcC.append(dL['cGcC'])
listeG4Hunter.append(dL['g4H'])
listeG4NN.append(dL['g4NN'])
if (dL['cGcC'] < dicoParam['cGcC']
or dL['g4H'] < dicoParam['g4H']
or dL['g4NN'] < dicoParam['g4NN']
or descriptionOverThreshold != dL['Description']):
# one of the score is under his threshold
# or this windows is from another gene
passed = False
if (oldPassed != passed):
# last windows before under the thresolds
meanCGcC = rF.mean(listeCGcC)
meanG4Hunter = rF.mean(listeG4Hunter)
meanG4NN = rF.mean(listeG4NN)
oldPassed = passed # update
headerG4 = rF.createIdG4(dL['circuId'], startG4, endG4,
dL['Strand'])
if dL['Strand'] == '+':
if (startG4 > dL['circuStart']
and startG4 < dL['circuEnd']
and endG4 > dL['circuEnd']):
# overlap
condition = True
else:
if (startG4 > dL['circuStart']
and startG4 < dL['circuEnd']
and endG4 < dL['circuStart']): # overlap
condition = True
if condition == True: # if overlap
location = 'junction'
if dL['Strand'] == '+':
endG4 = dL['circuStart'] + \
(endG4 - dL['circuEnd'])
else:
endG4 = dL['circuEnd'] - \
(dL['circuStart'] - endG4)
headerG4 = rF.createIdG4(dL['circuId'], startG4,
endG4, dL['Strand'])
G4Detected[headerG4] = [
str(meanCGcC),
str(meanG4Hunter), sequenceG4,
str(meanG4NN), location, rnaType
]
else:
if (G4Detected.has_key(headerG4) == False
and dL['Strand'] != None):
G4Detected[headerG4] = [
str(meanCGcC),
str(meanG4Hunter), sequenceG4,
str(meanG4NN), location, rnaType
]
condition = False
location = 'ExonNC'
inputfile.close()
return G4Detected
def returnG4InJunction(G4DetectedInJunction, inputfile, dicoParam):
"""Merges all windows from junctions upper thresholds, it will be pG4.
This function browses all windows returned by G4RNA Screener and
keep only those over the thresholds. It also merge overlapping
windows. If two windows upper thresholds are separated with one window
which is under thresholds, the two windows will not be merge and will be
concidered as 2 pG4.
:param G4DetectedInJunction: pG4 with its scores and sequence, those pG4
are predicted in junctions.
:type G4DetectedInJunction: dictionary
:param inputfile: name of the outputfile of G4RNA screener.
:type inputfile: string
:param dicoParam: all parameters given to G4RNA screener.
:type dicoParam: dictionary
:returns: G4DetectedInJunction, contains all pG4 in juctions,
with there scores.
:rtype: dictionary
"""
oldPassed = False
passed = False
descriptionOverThreshold = ''
inputfile = open(inputfile, 'r')
for line in inputfile:
if (re.search('^[0-9]', line)):
# if the line is not the header of the file
dicoLine = readLineG4ScreenerJun(line, 'Junction')
if (dicoLine['cGcC'] >= dicoParam['cGcC']
and dicoLine['g4H'] >= dicoParam['g4H']
and dicoLine['g4NN'] >= dicoParam['g4NN']
and dicoLine['WindowStart']):
# window over thresholds
onJunction = False
passed = True
if (oldPassed != passed):
# first windows over thresholds
descriptionOverThreshold = dicoLine['Description']
sequenceG4 = dicoLine['WindowSeq']
oldPassed = passed # update
startFirstWindow = dicoLine['locationStart']
endFirstWindow = dicoLine['locationEnd']
startG4 = dicoLine['WindowStart']
endG4 = dicoLine['WindowEnd']
listeCGcC = [dicoLine['cGcC']]
listeG4Hunter = [dicoLine['g4H']]
listeG4NN = [dicoLine['g4NN']]
else:
# not the first windows above the thresholds
sequenceG4 = rF.addWindowToG4Seq(sequenceG4,
dicoLine['WindowSeq'],
dicoParam['Step'],
dicoParam['Window'])
endG4 = dicoLine['WindowEnd']
listeCGcC.append(dicoLine['cGcC'])
listeG4Hunter.append(dicoLine['g4H'])
listeG4NN.append(dicoLine['g4NN'])
if (dicoLine['cGcC'] < dicoParam['cGcC']
or dicoLine['g4H'] < dicoParam['g4H']
or dicoLine['g4NN'] < dicoParam['g4NN']
or descriptionOverThreshold != dicoLine['Description']):
# one of the score is under his threshold
# or this windows is from another gene
passed = False
if (oldPassed != passed):
# last windows before under the thresolds
meanCGcC = rF.mean(listeCGcC)
meanG4Hunter = rF.mean(listeG4Hunter)
meanG4NN = rF.mean(listeG4NN)
oldPassed = passed
if (startG4 > 40 or endG4 < 160):
if descriptionOverThreshold != dicoLine['Description']:
description = descriptionOverThreshold.split(
':')[0] + ':junction:' + ':'.join(
descriptionOverThreshold.split(':')[2:])
else:
description = dicoLine['Description'].split(
':')[0] + ':junction:' + ':'.join(
dicoLine['Description'].split(':')[2:])
if description not in G4DetectedInJunction:
G4DetectedInJunction[description] = [
str(meanCGcC),
str(meanG4Hunter), sequenceG4,
str(meanG4NN)
]
inputfile.close()
return G4DetectedInJunction