'differ. Quitting.', file=sys.stderr)
exit(1)
NewConsensusAsString = ''
NewRefAsString = ''
for ConsensusBase, RefBase in itertools.izip(ConsensusAsString, RefAsString):
if RefBase == '-' and (ConsensusBase == '-' or ConsensusBase == '?'):
continue
else:
NewConsensusAsString += ConsensusBase
NewRefAsString += RefBase
ConsensusAsString = NewConsensusAsString
RefAsString = NewRefAsString
ref.seq = Seq.Seq(RefAsString)
# Replaces gaps that border "no coverage" by "no coverage".
ConsensusAsString = PropagateNoCoverageChar(ConsensusAsString)
# Check all seq IDs are unique.
IDsOfSeqsToBeAdded = [seq.id for seq in \
SeqIO.parse(open(args.OtherSeqsToBeAdded),'fasta')]
if consensus.id in IDsOfSeqsToBeAdded:
print('A sequence in',
args.OtherSeqsToBeAdded,
'is called',
consensus.id + ', like the consensus in',
args.SeqPairWithMissingCov +
'. Rename to avoid such a clash. Quitting.',
file=sys.stderr)
exit(1)
# Align
if CallMostCommon:
BaseToCall = CallAmbigBaseIfNeeded(BasesWithMaxCount, coverage)
else:
CountToCallBase = coverage * args.MinFracToCall
# This next 'if' would also be covered by the 'else', but the explicit
# 'if' scope is faster and is what is usually needed.
if MaxCount * len(BasesWithMaxCount) >= CountToCallBase:
BaseToCall = CallAmbigBaseIfNeeded(BasesWithMaxCount, coverage)
else:
BaseToCall = CallEnoughBases(counts, CountToCallBase, coverage)
consensus += BaseToCall
# Replaces gaps that border "no coverage" by "no coverage".
consensus = PropagateNoCoverageChar(consensus)
# Skip positions at which the ref has a gap and the consensus has a gap or
# missing cov.
if not args.ref_seq_missing:
NewConsensus = ''
NewRefSeq = ''
for ConsensusBase, RefBase in itertools.izip(consensus, RefSeq):
if RefBase == GapChar and (ConsensusBase == '?' or ConsensusBase == \
GapChar):
continue
NewConsensus += ConsensusBase
NewRefSeq += RefBase
consensus = NewConsensus
RefSeq = NewRefSeq
RefPosition = PositionInSeqToAdd
if not RefInMainHasUniqueInsertion:
SeqToAdd_WithGaps += SeqBase
ReferenceWithoutInsertions += RefSeqFromPair[PositionInSeqToAdd]
PositionInFinalAln += 1
TranslationRecord += '\n' +str(PositionInFinalAln) +',' +\
str(RefPosition+1) +',' +SeqBase
else:
TranslationRecord += '\n-,' + str(RefPosition + 1) + ',' + SeqBase
PositionInSeqToAdd += 1
if args.log_file != None:
with open(args.log_file, 'w') as f:
f.write(TranslationRecord)
FinalSeqToAdd = PropagateNoCoverageChar(SeqToAdd_WithGaps)
# Thanks Stackoverflow:
def insert_newlines(string, every=FastaSeqLineLength):
lines = []
for i in range(0, len(string), every):
lines.append(string[i:i + every])
return '\n'.join(lines)
print('>' + SeqToAddName)
print(insert_newlines(FinalSeqToAdd))
#print('>'+RefSeqName+'_UniqueInsertionsExcised')
#print(insert_newlines(ReferenceWithoutInsertions))
# Make a list, of the same length of the alignment, of integers: each one
# counting the number of contigs with coverage there. Gaps inside contigs get
# counted as coverage; gaps between contigs get a count of 0.
ContigCoverageByPosition = [0 for n in range(0,AlignmentLength)]
for [start,end] in ContigStartsAndEnds.values():
for position in range(start,end+1):
ContigCoverageByPosition[position] += 1
# Compare contigs to consensus to count positions in the four categories
# ('cats') described in the help above. Convert all bases to upper case, and
# replace any gap char that neighbours a '?' char in the consensus by a '?'.
if ConsensusName != None:
categories = []
ConsensusSeq = ConsensusSeq.upper()
ConsensusSeq = PropagateNoCoverageChar(ConsensusSeq)
FlattenedContigsSeq = FlattenedContigsSeq.upper()
for ContigName in ContigDict:
ContigDict[ContigName] = ContigDict[ContigName].upper()
if CheckContigSNPs:
NumPosLongestContigCorrect = 0
NumPosLongestContigIncorrect = 0
for pos in range(0,AlignmentLength):
if ContigCoverageByPosition[pos] < 2:
continue
ContigBases = []
for ContigName, ContigSeq in ContigDict.items():
StartOfContig, EndOfContig = ContigStartsAndEnds[ContigName]
if StartOfContig <= pos <= EndOfContig:
ContigBase = ContigSeq[pos]
BaseToCall = CallAmbigBaseIfNeeded(BasesWithMaxCount, coverage)
else:
CountToCallBase = coverage * args.MinFracToCall
# This next 'if' would also be covered by the 'else', but the explicit
# 'if' scope is faster and is what is usually needed.
if MaxCount * len(BasesWithMaxCount) >= CountToCallBase:
BaseToCall = CallAmbigBaseIfNeeded(BasesWithMaxCount, coverage)
else:
BaseToCall = CallEnoughBases(counts, CountToCallBase, coverage)
consensus += BaseToCall
# Replaces gaps that border "no coverage" by "no coverage".
if not args.keep_gaps_by_missing:
consensus = PropagateNoCoverageChar(consensus)
# Skip positions at which the ref has a gap and the consensus has a gap or
# missing cov.
if not args.ref_seq_missing:
NewConsensus = ''
NewRefSeq = ''
for ConsensusBase, RefBase in zip(consensus, RefSeq):
if RefBase == GapChar and (ConsensusBase == '?' or ConsensusBase == \
GapChar):
continue
NewConsensus += ConsensusBase
NewRefSeq += RefBase
consensus = NewConsensus
RefSeq = NewRefSeq