// Given a SMRT sequence and a subread interval, make the subread.
// Input:
// smrtRead - a SMRT sequence
// subreadInterval - a subread interval
// params - mapping parameters
// Output:
// subreadSequence - the constructed subread
void MakeSubreadOfInterval(SMRTSequence & subreadSequence,
SMRTSequence & smrtRead,
ReadInterval & subreadInterval,
MappingParameters & params)
{
int start = subreadInterval.start;
int end = subreadInterval.end;
assert(smrtRead.length >= subreadSequence.length);
smrtRead.MakeSubreadAsMasked(subreadSequence, start, end);
if (!params.preserveReadTitle) {
smrtRead.SetSubreadTitle(subreadSequence,
subreadSequence.SubreadStart(),
subreadSequence.SubreadEnd());
}
else {
subreadSequence.CopyTitle(smrtRead.title);
}
subreadSequence.zmwData = smrtRead.zmwData;
}
void MakeVirtualRead(SMRTSequence & smrtRead,
const vector<SMRTSequence> & subreads)
{
assert(subreads.size() > 0);
DNALength hqStart = 0, hqEnd = 0;
for(auto subread: subreads) {
hqStart = min(DNALength(subread.SubreadStart()), hqStart);
hqEnd = max(DNALength(subread.SubreadEnd()), hqEnd);
}
smrtRead.Free();
smrtRead.Allocate(hqEnd);
memset(smrtRead.seq, 'N', sizeof(char) * hqEnd);
smrtRead.lowQualityPrefix = hqStart;
smrtRead.lowQualitySuffix = smrtRead.length - hqEnd;
smrtRead.highQualityRegionScore = subreads[0].highQualityRegionScore;
smrtRead.HoleNumber(subreads[0].HoleNumber());
stringstream ss;
ss << SMRTTitle(subreads[0].GetTitle()).MovieName() << "/" << subreads[0].HoleNumber();
smrtRead.CopyTitle(ss.str());
for (auto subread: subreads) {
memcpy(&smrtRead.seq[subread.SubreadStart()],
&subread.seq[0], sizeof(char) * subread.length);
}
}
void SMRTSequence::SetSubreadTitle(SMRTSequence &subread, DNALength subreadStart, DNALength subreadEnd) {
stringstream titleStream;
titleStream << title << "/"<< subreadStart << "_" << subreadEnd;
subread.CopyTitle(titleStream.str());
}
int main(int argc, char* argv[]) {
string inputFileName, outputFileName;
if (argc < 2) {
PrintUsage();
exit(0);
}
vector<string> inputFileNames;
inputFileName = argv[1];
outputFileName = argv[2];
int argi = 3;
RegionTable regionTable;
string regionsFOFNName = "";
vector<string> regionFileNames;
bool splitSubreads = true;
bool useCCS = false;
int minSubreadLength = 1;
while (argi < argc) {
if (strcmp(argv[argi], "-regionTable") == 0) {
regionsFOFNName = argv[++argi];
}
else if (strcmp(argv[argi], "-noSplitSubreads") == 0) {
splitSubreads = false;
}
else if (strcmp(argv[argi], "-minSubreadLength") == 0) {
minSubreadLength = atoi(argv[++argi]);
}
else if (strcmp(argv[argi], "-useccsdenovo") == 0) {
useCCS = true;
}
else {
PrintUsage();
cout << "ERROR! Option " << argv[argi] << " is not supported." << endl;
}
argi++;
}
if (FileOfFileNames::IsFOFN(inputFileName)) {
FileOfFileNames::FOFNToList(inputFileName, inputFileNames);
}
else {
inputFileNames.push_back(inputFileName);
}
if (regionsFOFNName == "") {
regionFileNames = inputFileNames;
}
else {
if (FileOfFileNames::IsFOFN(regionsFOFNName)) {
FileOfFileNames::FOFNToList(regionsFOFNName, regionFileNames);
}
else {
regionFileNames.push_back(regionsFOFNName);
}
}
ofstream fastaOut;
CrucialOpen(outputFileName, fastaOut);
int plsFileIndex;
HDFRegionTableReader hdfRegionReader;
AfgBasWriter afgWriter;
afgWriter.Initialize(outputFileName);
for (plsFileIndex = 0; plsFileIndex < inputFileNames.size(); plsFileIndex++) {
if (splitSubreads) {
hdfRegionReader.Initialize(regionFileNames[plsFileIndex]);
hdfRegionReader.ReadTable(regionTable);
regionTable.SortTableByHoleNumber();
}
ReaderAgglomerate reader;
// reader.SkipReadQuality(); // should have been taken care of by *Filter modules
if (useCCS){
reader.UseCCS();
} else {
reader.IgnoreCCS();
}
reader.Initialize(inputFileNames[plsFileIndex]);
CCSSequence seq;
int seqIndex = 0;
int numRecords = 0;
vector<ReadInterval> subreadIntervals;
while (reader.GetNext(seq)){
++seqIndex;
if (splitSubreads == false) {
if (seq.length >= minSubreadLength) {
afgWriter.Write(seq);
}
seq.Free();
continue;
}
DNALength hqReadStart, hqReadEnd;
int score;
GetReadTrimCoordinates(seq, seq.zmwData, regionTable, hqReadStart, hqReadEnd, score);
subreadIntervals.clear(); // clear old, new intervals are appended.
CollectSubreadIntervals(seq,®ionTable, subreadIntervals);
if (seq.length == 0 and subreadIntervals.size() > 0) {
cout << "WARNING! A high quality interval region exists for a read of length 0." <<endl;
cout << " The offending ZMW number is " << seq.zmwData.holeNumber << endl;
seq.Free();
continue;
}
for (int intvIndex = 0; intvIndex < subreadIntervals.size(); intvIndex++) {
SMRTSequence subreadSequence;
int subreadStart = subreadIntervals[intvIndex].start > hqReadStart ?
subreadIntervals[intvIndex].start : hqReadStart;
int subreadEnd = subreadIntervals[intvIndex].end < hqReadEnd ?
subreadIntervals[intvIndex].end : hqReadEnd;
int subreadLength = subreadEnd - subreadStart;
if (subreadLength < minSubreadLength) continue;
subreadSequence.subreadStart = subreadStart;
subreadSequence.subreadEnd = subreadEnd;
subreadSequence.ReferenceSubstring(seq, subreadStart, subreadLength);
stringstream titleStream;
titleStream << seq.title << "/" << subreadIntervals[intvIndex].start
<< "_" << subreadIntervals[intvIndex].end;
subreadSequence.CopyTitle(titleStream.str());
afgWriter.Write(subreadSequence);
delete[] subreadSequence.title;
}
seq.Free();
}
reader.Close();
hdfRegionReader.Close();
}
}