def convert(sequenceData, lastTranscriptID, fivePrimeSequence, fivePrimeUTRs,
cdss, stopCodon, direction):
cdsSequence = ""
cdssAndStopCodon = cdss[:]
for cds in cdssAndStopCodon:
cdsSequence += gencode.readSequence(sequenceData, cds['start'],
cds['end'], direction)
stopCodonSequence = ""
if stopCodon:
#stop codon is always last
cdssAndStopCodon.append(stopCodon)
stopCodonSequence += gencode.readSequence(sequenceData,
stopCodon['start'],
stopCodon['end'], direction)
sequence = fivePrimeSequence + cdsSequence + stopCodonSequence
EnsemblID = lastTranscriptID.split(
'.'
)[0] # transcript stripped of version number to match RefSeq dictionary transcript
uORF_number = 1
while lee_TIS_dict.get((EnsemblID, uORF_number)):
RefSeqID, position_to_aTIS, annotation, RLTM_RCHX, codon, uORF_number = lee_TIS_dict.get(
(EnsemblID, uORF_number))
position = position_to_aTIS + sequence.find(cdsSequence)
# Now find the closest codon in the transcript sequence
offSet = 0
lst_pos = []
codon_position = sequence.find(codon, offSet)
offSet = codon_position + 1
while codon_position != -1:
lst_pos.append(codon_position)
codon_position = sequence.find(codon, offSet)
offSet = codon_position + 1
best_position = closest_match(lst_pos, position)
#startPosition, endPosition = conversion_parts(fivePrimeUTRs, cdssAndStopCodon, best_position - 1, best_position + 3, direction)
record = fivePrimeUTRs[0]
uORF_ID = RefSeqID + ':' + annotation + '_' + codon
if abs(best_position - position) <= 5: # 5 is an arbitrary threshold
writeUORFOutput(uORF_ID, fivePrimeUTRs, cdssAndStopCodon,
best_position, best_position + 3, direction,
RLTM_RCHX)
# 5 is an arbitrary threshold for matching based on me seeing that most of the elements were off by 1 or 0, which is what I expect
#leeToFritschFile.write(record['chromosome'] + '\t' + str(startPosition) + '\t' + str(endPosition) + '\t' + \
# RefSeqID + ':' + annotation + '_' + codon + '\t' + RLTM_RCHX + '\t' + \
# direction + '\n')
else:
tooFarleeuorfsFile.write(record['chromosome'] + '\t' + str(best_position-1) + '\t' + str(best_position + 3) + '\t' + \
RefSeqID + ':' + annotation + '_' + codon + '\t' + RLTM_RCHX + '\t' + \
direction + '\t' + str(best_position - position) + '\n')
uORF_number += 1
def transcriptCallback(records, sequenceData, lastTranscriptID, fivePrimeUTRs, fivePrimeContent, cdss, stopCodon, direction):
cdsSequence = ""
for cds in cdss:
cdsSequence += gencode.readSequence(sequenceData, cds['start'], cds['end'], direction)
cdsAnnotationFile.write(cds['line'])
stopCodonSequence = ""
if stopCodon:
stopCodonSequence += gencode.readSequence(sequenceData, stopCodon['start'], stopCodon['end'], direction)
#print 'error! No transcript ID for ' + str(lastTranscriptID)
if cdsSequence:
cdsSequenceFile.write(">" + str(records[0]['geneID']) + "|" + str(lastTranscriptID) + "|" + str(records[0]["geneName"]) + "\n")
cdsSequenceFile.write(cdsSequence + stopCodonSequence + "\n")
def transcriptCallback(records, sequenceData, lastTranscriptID, fivePrimeUTRs,
fivePrimeContent, cdss, stopCodon, direction):
#Scan intron's, ignore first cds block
for cds in cdss[1:]:
if direction == '+':
startPosition = cds['start'] - 6
endPosition = cds['start'] + 2
elif direction == '-':
startPosition = cds['end'] - 2
endPosition = cds['end'] + 6
#Not in bounds of sequence data, skip
if endPosition > len(sequenceData):
print 'once'
print cds['start']
print cds['end']
print cds
print len(sequenceData)
continue
sequence = gencode.readSequence(sequenceData, startPosition,
endPosition, direction)
#NAGNAGNAG
lowerPosition = 'NA'
higherPosition = 'NA'
if sequence[1:3][0] == 'A' or sequence[1:3][1] == 'G':
if direction == '+':
lowerPosition = str(startPosition + 1)
elif direction == '-':
higherPosition = str(endPosition - 2)
if sequence[7:9][0] == 'A' or sequence[7:9][1] == 'G':
if direction == '+':
higherPosition = str(startPosition + 7)
elif direction == '-':
lowerPosition = str(endPosition - 8)
if "AG" == sequence[4:6]:
spliceAcceptorCase = 'SpliceAcceptorIsAG'
else:
spliceAcceptorCase = 'SpliceAcceptorIsNotAG'
agCase = None
if "AG" == sequence[1:3] or "AG" == sequence[7:9]:
agCase = 'FoundAGOnSides'
elif sequence[1:3][0] == 'A' or sequence[1:3][1] == 'G' or sequence[
7:9][0] == 'A' or sequence[7:9][1] == 'G':
agCase = 'FoundEitherAOrGOnSides'
if agCase:
nagAnnotationFile.write("\t".join([
records[0]['chromosome'], 'mayur', 'NAGNAG', records[0]
['geneID'], records[0]['geneName'], records[0]['geneType'],
lastTranscriptID, records[0]['transcriptType'], direction,
str(startPosition),
str(endPosition), sequence, lowerPosition, higherPosition,
spliceAcceptorCase, agCase
]) + "\n")
def findKozak2(outputFile, sequenceData, transcript, cdss, stopCodon,
direction):
cdsSequence = ""
cdssAndStopCodon = cdss[:]
for cds in cdssAndStopCodon:
cdsSequence += gencode.readSequence(sequenceData, cds['start'],
cds['end'], direction)
stopCodonSequence = ""
if stopCodon:
#stop codon is always last
cdssAndStopCodon.append(stopCodon)
stopCodonSequence += gencode.readSequence(sequenceData,
stopCodon['start'],
stopCodon['end'], direction)
afterCDSAndStopCodonSequence = ""
genomeRecord = {
'chromosome': cdssAndStopCodon[0]['chromosome'],
'type': 'genome',
'transcriptName': cdssAndStopCodon[0]['transcriptName'],
'geneID': cdssAndStopCodon[0]['geneID'],
'geneName': cdssAndStopCodon[0]['geneName'],
'transcriptStatus': cdssAndStopCodon[0]['transcriptStatus'],
'geneStatus': cdssAndStopCodon[0]['geneStatus'],
'levelNumber': ''
}
sequence = cdsSequence + stopCodonSequence + afterCDSAndStopCodonSequence
atgSequenceLength = len(cdsSequence)
atgOffsetToSearch = 0
sequenceLength = len(sequence)
transcriptNumber = 1
genomeLength = 300
while atgOffsetToSearch < atgSequenceLength - 2:
atgIndex = cdsSequence.find('ATG', atgOffsetToSearch)
if atgIndex == -1:
break
atgOffsetToSearch = atgIndex + 1
if atgIndex < atgSequenceLength - 2:
endIndex = atgIndex + 3
while endIndex < sequenceLength - 2:
threeCharacterSequence = sequence[endIndex:endIndex + 3]
if threeCharacterSequence == 'TAG' or threeCharacterSequence == 'TAA' or threeCharacterSequence == 'TGA':
kozakSequence = sequence[atgIndex:endIndex + 3]
writeKozakOutput2(
outputFile,
transcript + ".kozak_cds." + str(transcriptNumber),
cdssAndStopCodon, genomeRecord, atgIndex, endIndex + 3,
direction)
transcriptNumber += 1
found = True
break
endIndex += 3
if endIndex >= sequenceLength - 2 and sequenceLength < len(
sequenceData):
if direction == '+':
afterCDSAndStopCodonSequence = gencode.readSequence(
sequenceData, cdssAndStopCodon[-1]['end'] + 1,
min(cdssAndStopCodon[-1]['end'] + genomeLength,
len(sequenceData)), direction)
genomeRecord['start'] = cdssAndStopCodon[-1]['end'] + 1
genomeRecord['end'] = len(sequenceData)
elif direction == '-':
afterCDSAndStopCodonSequence = gencode.readSequence(
sequenceData,
max(cdssAndStopCodon[-1]['start'] - genomeLength,
1), cdssAndStopCodon[-1]['start'] - 1,
direction)
genomeRecord['start'] = 1
genomeRecord['end'] = cdssAndStopCodon[-1]['start'] - 1
sequence = cdsSequence + stopCodonSequence + afterCDSAndStopCodonSequence
sequenceLength = len(sequence)
genomeLength += 300
def findKozak(utrOutputFile, cdsOutputFile, transcript, sequenceData,
fivePrimeSequence, fivePrimeUTRs, cdss, stopCodon, direction):
cdsSequence = ""
for cds in cdss:
cdsSequence += gencode.readSequence(sequenceData, cds['start'],
cds['end'], direction)
beforeFivePrimeSequence = ""
afterCDSSequence = ""
if direction == "+":
if fivePrimeUTRs[0]['start'] - 9 > 0:
beforeFivePrimeSequence = gencode.readSequence(
sequenceData, fivePrimeUTRs[0]['start'] - 9,
fivePrimeUTRs[0]['start'] - 1, direction)
if cdss[-1]['end'] + 7 <= len(sequenceData):
afterCDSSequence = gencode.readSequence(sequenceData,
cdss[-1]['end'] + 1,
cdss[-1]['end'] + 7,
direction)
else:
if fivePrimeUTRs[-1]['start'] - 9 > 0:
beforeFivePrimeSequence = gencode.readSequence(
sequenceData, fivePrimeUTRs[-1]['start'] - 9,
fivePrimeUTRs[-1]['start'] - 1, direction)
if cdss[0]['end'] + 7 <= len(sequenceData):
afterCDSSequence = gencode.readSequence(sequenceData,
cdss[0]['end'] + 1,
cdss[0]['end'] + 7,
direction)
sequence = beforeFivePrimeSequence + fivePrimeSequence + cdsSequence + afterCDSSequence
atgOffsetToSearch = len(beforeFivePrimeSequence)
transcriptVersion = 1
while atgOffsetToSearch < len(beforeFivePrimeSequence) + len(
fivePrimeSequence) - 2:
atgIndex = sequence.find('ATG', atgOffsetToSearch)
if atgIndex == -1:
break
if atgIndex < len(beforeFivePrimeSequence) + len(
fivePrimeSequence) - 2:
writeKozakOutput(utrOutputFile, fivePrimeUTRs[0], atgIndex,
atgIndex - len(beforeFivePrimeSequence) + 1,
len(fivePrimeSequence), sequence, "kozak_utr",
transcript, transcriptVersion, direction)
transcriptVersion += 1
atgOffsetToSearch = atgIndex + 1
transcriptVersion = 1
atgOffsetToSearch = len(beforeFivePrimeSequence) + len(fivePrimeSequence)
while atgOffsetToSearch < len(beforeFivePrimeSequence) + len(
fivePrimeSequence) + len(cdsSequence) - 2:
atgIndex = sequence.find('ATG', atgOffsetToSearch)
if atgIndex == -1:
break
if atgIndex < len(beforeFivePrimeSequence) + len(
fivePrimeSequence) + len(cdsSequence) - 2:
writeKozakOutput(
cdsOutputFile, cdss[0], atgIndex, atgIndex -
len(fivePrimeSequence) - len(beforeFivePrimeSequence) + 1,
len(cdsSequence), sequence, "kozak_cds", transcript,
transcriptVersion, direction)
transcriptVersion += 1
atgOffsetToSearch = atgIndex + 1
def findUORF(outputFile, outputSequenceFile, sequenceData, divisibleByThree,
transcript, fivePrimeSequence, fivePrimeUTRs, cdss, stopCodon,
direction):
cdsSequence = ""
cdssAndStopCodon = cdss[:]
for cds in cdssAndStopCodon:
cdsSequence += gencode.readSequence(sequenceData, cds['start'],
cds['end'], direction)
stopCodonSequence = ""
if stopCodon:
#stop codon is always last
cdssAndStopCodon.append(stopCodon)
stopCodonSequence += gencode.readSequence(sequenceData,
stopCodon['start'],
stopCodon['end'], direction)
sequence = fivePrimeSequence + cdsSequence + stopCodonSequence
atgOffsetToSearch = 0
atgSequenceLength = len(fivePrimeSequence)
sequenceLength = len(sequence)
transcriptNumber = 1
while atgOffsetToSearch < atgSequenceLength - 2:
atgIndex = fivePrimeSequence.find(StartCodon, atgOffsetToSearch)
if atgIndex == -1:
break
atgOffsetToSearch = atgIndex + 1
if atgIndex < atgSequenceLength - 2:
endIndex = atgIndex + 3
while endIndex < sequenceLength - 2:
threeCharacterSequence = sequence[endIndex:endIndex + 3]
if threeCharacterSequence == 'TAG' or threeCharacterSequence == 'TAA' or threeCharacterSequence == 'TGA':
if not divisibleByThree or (endIndex - atgIndex) % 3 == 0:
uorfSequence = sequence[atgIndex:endIndex + 3]
howFarReached = ""
if endIndex + 3 <= len(fivePrimeSequence):
howFarReached = "UTRonly"
elif endIndex + 3 <= len(fivePrimeSequence) + len(
cdsSequence):
howFarReached = "CDSpartial"
else:
howFarReached = "CDSfull"
record = fivePrimeUTRs[0]
outputSequenceFile.write(record["chromosome"] + "|" +
record["geneID"] + "|" +
transcript + ".uORF_" +
StartCodon + "." +
str(transcriptNumber) + "|" +
direction + "|" +
record["geneName"] + "|" +
howFarReached + "\n")
outputSequenceFile.write(uorfSequence + "\n")
writeUORFOutput(
outputFile, transcript + ".uORF_" + StartCodon +
"." + str(transcriptNumber), fivePrimeUTRs,
cdssAndStopCodon, atgIndex, endIndex + 3,
direction)
transcriptNumber += 1
break
endIndex += 1
