int readerToMeth(BamTools::BamReader & reader1,
BamTools::BamReader & reader2,
std::map<std::string, std::vector<std::string> > & lociMeth1,
std::map<std::string, std::vector<std::string> > & lociMeth2,
BamTools::RefVector::const_iterator i,
int d,
const BamTools::RefVector refs,
std::string sample)
{
const int r1=reader1.GetReferenceID(i->RefName);
const int r2=reader2.GetReferenceID(i->RefName);
const int rl=i->RefLength;
if(reader1.SetRegion(r1,0, r1, rl) & reader2.SetRegion(r2,0, r2, rl))
{
std::cerr << "Processing " << i->RefName << std::endl;
BamTools::BamAlignment al;
while (reader1.GetNextAlignment(al)){
//std::cout << al.RefID << std::endl;
byread(al, d, lociMeth1, refs, sample);
}
while (reader2.GetNextAlignment(al)){
//std::cout << al.RefID << std::endl;
byread(al, d, lociMeth2, refs, sample);
}
}
reader1.Rewind();
reader2.Rewind();
return 0;
}
std::string getQuickStats(const std::string &bamFile, std::map< std::string, int > &keyLen, unsigned int &nFlowFZ, unsigned int &nFlowZM) {
std::string errMsg = "";
BamTools::BamReader bamReader;
if(!bamReader.Open(bamFile)) {
errMsg += "Failed to open bam " + bamFile + "\n";
return(errMsg);
}
BamTools::SamHeader samHeader = bamReader.GetHeader();
for (BamTools::SamReadGroupIterator itr = samHeader.ReadGroups.Begin(); itr != samHeader.ReadGroups.End(); ++itr ) {
if(itr->HasID())
keyLen[itr->ID] = itr->HasKeySequence() ? itr->KeySequence.length() : 0;
if(itr->HasFlowOrder())
nFlowZM = std::max(nFlowZM,(unsigned int) itr->FlowOrder.length());
}
BamTools::BamAlignment alignment;
std::vector<uint16_t> flowIntFZ;
while(bamReader.GetNextAlignment(alignment)) {
if(alignment.GetTag("FZ", flowIntFZ))
nFlowFZ = flowIntFZ.size();
break;
}
bamReader.Close();
if(nFlowFZ==0)
std::cout << "NOTE: bam file has no flow signals in FZ tag: " + bamFile + "\n";
if(nFlowZM==0)
std::cout << "NOTE: bam file has no flow signals in ZM tag: " + bamFile + "\n";
return(errMsg);
}
void bamParser::parse(Reads & reads, string & filename,
vector<string> & chrs_to_parse) {
BamTools::BamReader bam;
BamTools::BamAlignment read;
string chr;
uint64_t readCnt = 0;
uint32_t meanReadLen = 0;
if (!(bam.Open(filename))) {
throw FileNotGood(filename);
}
const BamTools::RefVector refvec = bam.GetReferenceData();
while (bam.GetNextAlignment(read)) {
chr = getR1Chr(read, refvec);
if (isGoodRead(read)) {
if (isChrToParse(chrs_to_parse, chr)) {
updateAvgReadLength(readCnt, meanReadLen, read);
insertRead(read, reads, chr);
}
}
}
reads.setReadlength(meanReadLen);
}
bool getNextAlignment(BamTools::BamAlignment &alignment, BamTools::BamReader &bamReader, const std::map<std::string, int> &groupID, std::vector< BamTools::BamAlignment > &alignmentSample, std::map<std::string, int> &wellIndex, unsigned int nSample) {
if(nSample > 0) {
// We are randomly sampling, so next read should come from the sample that was already taken from the bam file
if(alignmentSample.size() > 0) {
alignment = alignmentSample.back();
alignmentSample.pop_back();
alignment.BuildCharData();
return(true);
} else {
return(false);
}
} else {
// No random sampling, so we're either returning everything or we're looking for specific read names
bool storeRead = false;
while(bamReader.GetNextAlignment(alignment)) {
if(groupID.size() > 0) {
std::string thisReadGroupID = "";
if( !alignment.GetTag("RG", thisReadGroupID) || (groupID.find(thisReadGroupID)==groupID.end()) );
continue;
}
storeRead=true;
if(wellIndex.size() > 0) {
// We are filtering by position, so check if we should skip or keep the read
int thisCol,thisRow;
if(1 != ion_readname_to_rowcol(alignment.Name.c_str(), &thisRow, &thisCol))
std::cerr << "Error parsing read name: " << alignment.Name << "\n";
std::stringstream wellIdStream;
wellIdStream << thisCol << ":" << thisRow;
std::map<std::string, int>::iterator wellIndexIter;
wellIndexIter = wellIndex.find(wellIdStream.str());
if(wellIndexIter != wellIndex.end()) {
// If the read ID matches we should keep, unless its a duplicate
if(wellIndexIter->second >= 0) {
storeRead=true;
wellIndexIter->second=-1;
} else {
storeRead=false;
std::cerr << "WARNING: found extra instance of readID " << wellIdStream.str() << ", keeping only first\n";
}
} else {
// read ID is not one we should keep
storeRead=false;
}
}
if(storeRead)
break;
}
return(storeRead);
}
}
uint32_t BamAlignmentReader::GetReadLength(const std::string& bamPath)
{
uint32_t bamReadLength = 300;
BamTools::BamReader bamReader;
if (!bamReader.Open(bamPath))
{
throw "Unable to open bam file";
}
BamTools::BamAlignment bamAlignment;
while(bamReader.GetNextAlignment(bamAlignment))
{
if (bamAlignment.IsPrimaryAlignment())
{
bamReadLength = bamAlignment.QueryBases.size();
break;
}
}
bamReader.Close();
return bamReadLength;
}
void Config::InitializationClustering() {
struct stat st;
if(stat(Workspace.c_str(),&st) == 0 and st.st_mode and S_IFDIR != 0) Log("[Warning] Workspace directory already present");
else if (mkdir(Workspace.c_str(), 0755) != 0) {
Log("[Error] Could not create workspace directory: " + Workspace);
exit(1);
}
RunningTasksFile = Workspace + "/" + FilePrefix + "running.tasks";
StatsFile = Workspace + "/" + FilePrefix + "stats";
BinClusterFile = Workspace + "/" + FilePrefix + "bpc";
clusterFile = new ClusterFile(BinClusterFile);
clusterDir = Workspace + "/clusters/";
if(stat(clusterDir.c_str(),&st) == 0 and st.st_mode and S_IFDIR != 0) Log("[Warning] Cluster directory already present");
else if (mkdir(clusterDir.c_str(), 0755) != 0) {
Log("[Error] Could not create cluster directory: " + clusterDir);
exit(1);
}
insertsizeDir = Workspace + "/insertsize/";
if(stat(insertsizeDir.c_str(),&st) == 0 and st.st_mode and S_IFDIR != 0) Log("[Warning] Insertsize directory already present");
else if (mkdir(insertsizeDir.c_str(), 0755) != 0) {
Log("[Error] Could not create insertsize directory: " + insertsizeDir);
exit(1);
}
coverageDir = Workspace + "/coverage/";
if(stat(coverageDir.c_str(),&st) == 0 and st.st_mode and S_IFDIR != 0) Log("[Warning] Coverage directory already present");
else if (mkdir(coverageDir.c_str(), 0755) != 0) {
Log("[Error] Could not create coverage directory: " + coverageDir);
exit(1);
}
if (!ForwardBam.empty() && !ReverseBam.empty() && PairedBam.empty()) {
UsePairedBam = false;
} else if (ForwardBam.empty() && ReverseBam.empty() && !PairedBam.empty()) {
UsePairedBam = true;
} else {
Log("[Error] No correct bam file(s)");
exit(1);
}
BamTools::BamAlignment alignment;
BamTools::BamReader BamReader;
if (UsePairedBam) {
BamReader.Open(PairedBam);
if (not BamReader.IsOpen()) {
Log("[Error] Could not open paired bam");
exit(1);
}
if (PairedIndex.empty()) {
if (not BamReader.LocateIndex(BamTools::BamIndex::STANDARD)) {
PairedIndex = PairedBam.substr(0,PairedBam.find_last_of(".bam")-3) + ".bai";
BamReader.OpenIndex(PairedIndex);
}
if (not BamReader.HasIndex()) {
Log("[Error] No index for bamfile");
exit(1);
}
}
BamTools::SamHeader header = BamReader.GetHeader();
for (BamTools::SamReadGroupIterator it = header.ReadGroups.Begin(); it != header.ReadGroups.End(); it++) {
BamTools::SamReadGroup* readgroup = &*it;
readNameConverter.TrimName(readgroup->ID);
readNameConverter.AddReadGroup(readgroup->ID);
}
long int count = 0;
while (BamReader.GetNextAlignment(alignment)) {
string RG;
if (alignment.GetTag("RG", RG)) {
if (not NameTrim.empty()) readNameConverter.TrimName(RG);
if (readNameConverter.AddReadGroup(RG)) {
Log("[Warning] Readgroup '" + RG + "' found in reads but not in header");
count = 0;
}
}
count++;
if (count > 10000) break;
}
BamReader.Close();
} else {
BamReader.Open(ForwardBam);
if (not BamReader.IsOpen()) {
Log("[Error] Could not open first/forward bam");
exit(1);
}
if (ForwardIndex.empty()) {
if (not BamReader.LocateIndex(BamTools::BamIndex::STANDARD)) {
ForwardIndex = ForwardBam.substr(0,ForwardBam.find_last_of(".bam")-3) + ".bai";
BamReader.OpenIndex(ForwardIndex);
}
if (not BamReader.HasIndex()) {
Log("[Error] No index for forward bamfile");
exit(1);
}
}
BamTools::SamHeader forwardheader = BamReader.GetHeader();
for (BamTools::SamReadGroupIterator it = forwardheader.ReadGroups.Begin(); it != forwardheader.ReadGroups.End(); it++) {
BamTools::SamReadGroup* readgroup = &*it;
readNameConverter.TrimName(readgroup->ID);
readNameConverter.AddReadGroup(readgroup->ID);
}
long int count = 0;
while (BamReader.GetNextAlignment(alignment)) {
string RG;
if (alignment.GetTag("RG", RG)) {
if (!NameTrim.empty()) readNameConverter.TrimName(RG);
if (readNameConverter.AddReadGroup(RG)) {
Log("[Warning] Readgroup '" + RG + "' found in forward reads but not in header");
count = 0;
}
}
count++;
if (count > 10000) break;
}
BamReader.Close();
BamReader.Open(ReverseBam);
if (not BamReader.IsOpen()) {
Log("[Error] Could not open second/reverse bam");
exit(1);
}
if (ReverseIndex.empty()) {
if (not BamReader.LocateIndex(BamTools::BamIndex::STANDARD)) {
ReverseIndex = ReverseBam.substr(0,ReverseBam.find_last_of(".bam")-3) + ".bai";
BamReader.OpenIndex(ReverseIndex);
}
if (not BamReader.HasIndex()) {
Log("[Error] No index for reverse bamfile");
exit(1);
}
}
BamTools::SamHeader reverseheader = BamReader.GetHeader();
for (BamTools::SamReadGroupIterator it = reverseheader.ReadGroups.Begin(); it != reverseheader.ReadGroups.End(); it++) {
BamTools::SamReadGroup* readgroup = &*it;
readNameConverter.TrimName(readgroup->ID);
if (readNameConverter.AddReadGroup(readgroup->ID)) {
Log("[Warning] Readgroup '" + readgroup->ID + "' found in reverse but not in forward");
}
}
count = 0;
while (BamReader.GetNextAlignment(alignment)) {
string RG;
if (alignment.GetTag("RG", RG)) {
if (!NameTrim.empty()) readNameConverter.TrimName(RG);
if (readNameConverter.AddReadGroup(RG)) {
Log("[Warning] Readgroup '" + RG + "' found in reverse reads but not in header");
count = 0;
}
}
count++;
if (count > 10000) break;
}
BamReader.Close();
}
for(map<string, int>::iterator it = readNameConverter.ReadGroups.begin(); it!=readNameConverter.ReadGroups.end(); ++it) {
ostringstream logBuffer;
logBuffer << "Readgroup found: " << it->second << " - " << it->first;
Log(logBuffer.str());
}
writeConfigFile(Workspace + FilePrefix + "config");
}