void ReadAlignment(int alignmentIndex, AlignmentCandidate<FASTASequence, FASTASequence> &alignment) {
CmpAlignment cmpAln;
ReadAlignment(alignmentIndex, cmpAln);
string refSequence;
string readSequence;
readSequence.resize(cmpAln.alignmentArray.size());
refSequence.resize(cmpAln.alignmentArray.size());
ByteAlignmentToQueryString(&cmpAln.alignmentArray[0], cmpAln.alignmentArray.size(), &readSequence[0]);
ByteAlignmentToRefString(&cmpAln.alignmentArray[0], cmpAln.alignmentArray.size(), &refSequence[0]);
string ungappedRead, ungappedRef;
RemoveGaps(readSequence, ungappedRead);
RemoveGaps(refSequence, ungappedRef);
GappedStringsToAlignment(readSequence, refSequence, alignment);
FASTASequence qAlignedSeq, rAlignedSeq;
qAlignedSeq.seq = (Nucleotide*) &ungappedRead[0];
qAlignedSeq.length = ungappedRead.size();
rAlignedSeq.seq = (Nucleotide*) &ungappedRef[0];
rAlignedSeq.length = ungappedRef.size();
alignment.tAlignedSeq.Copy(rAlignedSeq);
alignment.qAlignedSeq.Copy(qAlignedSeq);
unsigned int qStart = cmpAln.GetQueryStart();
unsigned int tStart = cmpAln.GetRefStart();
alignment.tPos = cmpAln.GetRefStart();
alignment.qPos = cmpAln.GetQueryStart();
alignment.nIns = cmpAln.GetNInsertions();
alignment.nDel = cmpAln.GetNDeletions();
alignment.nMatch = cmpAln.GetNMatch();
alignment.nMismatch=cmpAln.GetNMismatch();
alignment.qStrand= 0;
alignment.tStrand = cmpAln.GetTStrand();
alignment.pctSimilarity = ((float)alignment.nMatch) / (alignment.nMatch + alignment.nMismatch + alignment.nIns + alignment.nDel);
alignment.mapQV = cmpAln.GetMapQV();
}
void ReadAlignmentArray(int alignmentIndex, ByteAlignment &alignmentArray) {
CmpAlignment cmpAlignment;
alnInfoGroup.ReadCmpAlignment(alignmentIndex, cmpAlignment);
//
// Cache some stats about the read, and where it was aligned to.
//
int queryStart = cmpAlignment.GetQueryStart();
int queryEnd = cmpAlignment.GetQueryEnd();
int refGroupId = cmpAlignment.GetRefGroupId();
int alnGroupId = cmpAlignment.GetAlnGroupId();
int refGroupIndex = refGroupIdToArrayIndex[refGroupId];
if (alnGroupIdToReadGroupName.find(alnGroupId) ==
alnGroupIdToReadGroupName.end()) {
cout << "INTERNAL ERROR! Could not find read group name for alignment "
<< "group with Id " << alnGroupId << "." << endl;
assert(0);
}
string readGroupName = alnGroupIdToReadGroupName[alnGroupId];
if (refAlignGroups[refGroupIndex]->experimentNameToIndex.find(readGroupName) ==
refAlignGroups[refGroupIndex]->experimentNameToIndex.end()) {
cout << "Internal ERROR! The read group name " << readGroupName << " is specified as part of "
<< " the path in alignment " << alignmentIndex
<< " though it does not exist in the ref align group specified for this alignment." << endl;
assert(0);
}
int readGroupIndex = refAlignGroups[refGroupIndex]->experimentNameToIndex[readGroupName];
HDFCmpExperimentGroup* expGroup = refAlignGroups[refGroupIndex]->readGroups[readGroupIndex];
int offsetBegin = cmpAlignment.GetOffsetBegin();
int offsetEnd = cmpAlignment.GetOffsetEnd();
int alignedSequenceLength = offsetEnd - offsetBegin;
if (alignedSequenceLength >= 0) {
alignmentArray.resize(alignedSequenceLength);
}
else {
return;
}
//
// Read the alignment string. All alignments
//
refAlignGroups[refGroupIndex]->readGroups[readGroupIndex]->alignmentArray.Read(offsetBegin,
offsetEnd,
&alignmentArray[0]);
}
void ImportReadFromCmpH5(int alignmentIndex, SMRTSequence &read) {
CmpAlignment cmpAlignment;
alnInfoGroup.ReadCmpAlignment(alignmentIndex, cmpAlignment);
//
// Cache some stats about the read, and where it was aligned to.
//
int queryStart = cmpAlignment.GetQueryStart();
int queryEnd = cmpAlignment.GetQueryEnd();
read.holeNumber = cmpAlignment.GetHoleNumber();
int refGroupId = cmpAlignment.GetRefGroupId();
int alnGroupId = cmpAlignment.GetAlnGroupId();
int refGroupIndex = refGroupIdToArrayIndex[refGroupId];
if (alnGroupIdToReadGroupName.find(alnGroupId) == alnGroupIdToReadGroupName.end()) {
cout << "INTERNAL ERROR! Could not find read group name for alignment "
<< "group with Id " << alnGroupId << "." << endl;
assert(0);
}
string readGroupName = alnGroupIdToReadGroupName[alnGroupId];
if (refAlignGroups[refGroupIndex]->experimentNameToIndex.find(readGroupName) ==
refAlignGroups[refGroupIndex]->experimentNameToIndex.end()) {
cout << "Internal ERROR! The read group name " << readGroupName << " is specified as part of "
<< " the path in alignment " << alignmentIndex
<< " though it does not exist in the ref align group specified for this alignment." << endl;
assert(0);
}
int readGroupIndex = refAlignGroups[refGroupIndex]->experimentNameToIndex[readGroupName];
HDFCmpExperimentGroup* expGroup = refAlignGroups[refGroupIndex]->readGroups[readGroupIndex];
int offsetBegin = cmpAlignment.GetOffsetBegin();
int offsetEnd = cmpAlignment.GetOffsetEnd();
int alignedSequenceLength = offsetEnd - offsetBegin;
string alignedSequence;
string readSequence;
vector<unsigned char> byteAlignment;
if (alignedSequenceLength >= 0) {
alignedSequence.resize(alignedSequenceLength);
byteAlignment.resize(alignedSequenceLength);
}
//
// Read the alignment string. All alignments
//
refAlignGroups[refGroupIndex]->readGroups[readGroupIndex]->alignmentArray.Read(offsetBegin,
offsetEnd,
&byteAlignment[0]);
//
// Convert to something we can compare easily.
//
ByteAlignmentToQueryString(&byteAlignment[0], byteAlignment.size(), &alignedSequence[0]);
//
// Initialize the sequence of the read.
//
RemoveGaps(alignedSequence, readSequence);
//
// Make space for the sequence and all fields.
//
read.length = readSequence.size();
read.Allocate(read.length);
memcpy(read.seq, readSequence.c_str(), readSequence.size() * sizeof(char));
vector<int> baseToAlignmentMap;
CreateSequenceToAlignmentMap(byteAlignment, baseToAlignmentMap);
//
// Read in the quality values
//
vector<unsigned char> storedQVArray;
vector<UChar> qvValues;
vector<HalfWord> frameValues;
int length = offsetEnd - offsetBegin;
qvValues.resize(length);
frameValues.resize(length);
int i;
if (expGroup->experimentGroup.ContainsObject("QualityValue")) {
expGroup->qualityValue.Read(offsetBegin, offsetEnd, &qvValues[0]);
StoreQualityValueFromAlignment(qvValues, baseToAlignmentMap, &read.qual.data[0]);
int i;
for (i= 0; i < read.length; i++) {
assert(read.qual[i] < 100);
}
}
if (expGroup->experimentGroup.ContainsObject("InsertionQV")) {
expGroup->insertionQV.Read(offsetBegin, offsetEnd, &qvValues[0]);
StoreQualityValueFromAlignment(qvValues, baseToAlignmentMap, &read.insertionQV.data[0]);
}
if (expGroup->experimentGroup.ContainsObject("SubstitutionQV")) {
expGroup->substitutionQV.Read(offsetBegin, offsetEnd, &qvValues[0]);
StoreQualityValueFromAlignment(qvValues, baseToAlignmentMap, &read.substitutionQV.data[0]);
}
if (expGroup->experimentGroup.ContainsObject("DeletionQV")) {
expGroup->deletionQV.Read(offsetBegin, offsetEnd, &qvValues[0]);
StoreQualityValueFromAlignment(qvValues, baseToAlignmentMap, &read.deletionQV.data[0]);
}
if (expGroup->experimentGroup.ContainsObject("DeletionTag")) {
vector<char> deletionTagValues;
deletionTagValues.resize(offsetEnd-offsetBegin);
expGroup->deletionTag.Read(offsetBegin, offsetEnd, &deletionTagValues[0]);
StoreQualityValueFromAlignment(deletionTagValues, baseToAlignmentMap, read.deletionTag);
}
if (expGroup->experimentGroup.ContainsObject("SubstitutionTag")) {
vector<char> substitutionTagValues;
substitutionTagValues.resize(offsetEnd-offsetBegin);
expGroup->substitutionTag.Read(offsetBegin, offsetEnd, &substitutionTagValues[0]);
StoreQualityValueFromAlignment(substitutionTagValues, baseToAlignmentMap, read.substitutionTag);
}
if (expGroup->experimentGroup.ContainsObject("PulseIndex")) {
vector<uint32_t> pulseIndexValues;
pulseIndexValues.resize(offsetEnd-offsetBegin);
expGroup->pulseIndex.Read(offsetBegin, offsetEnd, &pulseIndexValues[0]);
StoreQualityValueFromAlignment(pulseIndexValues, baseToAlignmentMap, read.pulseIndex);
}
if (expGroup->experimentGroup.ContainsObject("PreBaseFrames")) {
expGroup->preBaseFrames.Read(offsetBegin, offsetEnd, &frameValues[0]);
StoreQualityValueFromAlignment(frameValues, baseToAlignmentMap, read.preBaseFrames);
}
if (expGroup->experimentGroup.ContainsObject("WidthInFrames")) {
expGroup->widthInFrames.Read(offsetBegin, offsetEnd, &frameValues[0]);
StoreQualityValueFromAlignment(frameValues, baseToAlignmentMap, read.widthInFrames);
}
}