bool RapMapIndex::load(std::string& indexPrefix) {
auto logger = spdlog::get("stderrLog");
std::string kmerInfosName = indexPrefix + "kinfo.bin";
std::string eqClassListName = indexPrefix + "eqclass.bin";
std::string eqLabelListName = indexPrefix + "eqlab.bin";
std::string posListName = indexPrefix + "pos.bin";
std::string jfFileName = indexPrefix + "rapidx.jfhash";
std::string txpNameFile = indexPrefix + "txpnames.bin";
std::string txpLenFile = indexPrefix + "txplens.bin";
std::string fwdJumpFile = indexPrefix + "fwdjump.bin";
std::string revJumpFile = indexPrefix + "revjump.bin";
// Load the kmer info list first --- this will
// give us the # of unique k-mers
std::ifstream kmerInfoStream(kmerInfosName, std::ios::binary);
{
logger->info("loading k-mer info list . . .");
ScopedTimer timer;
cereal::BinaryInputArchive kmerInfoArchive(kmerInfoStream);
kmerInfoArchive(kmerInfos);
logger->info("done\n");
}
kmerInfoStream.close();
size_t numDistinctKmers = kmerInfos.size();
{
ScopedTimer timer;
logger->info("loading k-mer => id hash . . . ");
std::ifstream bis(jfFileName);
const SpecialHeader bh(bis);
// mapFile.reset(new jellyfish::mapped_file(jfFileName.c_str()));
const size_t sizeInBytes = bh.size_bytes();
// Load the hash from file
logger->info("\theader format = {}" , bh.format());
logger->info("\t# distinct k-mers = {}" , numDistinctKmers);
logger->info("\thash key len = {}" , bh.key_len());
logger->info("\tcounter len = {}" , bh.counter_len());
logger->info("\tmax reprobe offset = {}" , bh.max_reprobe());
logger->info("\tsize in bytes = {}" , sizeInBytes);
// Allocate the actual storage
rawHashMem.reset(new char[sizeInBytes]);
bis.read(rawHashMem.get(), sizeInBytes);
// We can close the file now
bis.close();
merHash.reset( new FileMerArray(rawHashMem.get(),//mapFile->base() + bh.offset(),
sizeInBytes,
bh.size(),
bh.key_len(),
bh.counter_len(),
bh.max_reprobe(),
bh.matrix()));
// Set the key size
rapmap::utils::my_mer::k(bh.key_len() / 2);
logger->info("done");
}
std::ifstream eqClassStream(eqClassListName, std::ios::binary);
{
logger->info("loading eq classes . . . ");
ScopedTimer timer;
cereal::BinaryInputArchive eqClassArchive(eqClassStream);
eqClassArchive(eqClassList);
logger->info("[{}] classes", eqClassList.size());
logger->info("done");
}
eqClassStream.close();
std::ifstream eqLabelStream(eqLabelListName, std::ios::binary);
{
logger->info("loading eq class labels . . . ");
ScopedTimer timer;
cereal::BinaryInputArchive eqLabelArchive(eqLabelStream);
eqLabelArchive(eqLabelList);
logger->info("[{}] labels", eqLabelList.size());
logger->info("done");
}
eqLabelStream.close();
std::ifstream posStream(posListName, std::ios::binary);
{
logger->info("loading position list . . . ");
ScopedTimer timer;
cereal::BinaryInputArchive posArchive(posStream);
posArchive(posList);
logger->info("[{}] total k-mer positions", posList.size());
logger->info("done");
}
posStream.close();
std::ifstream txpNameStream(txpNameFile, std::ios::binary);
{
logger->info("loading transcript names ");
ScopedTimer timer;
cereal::BinaryInputArchive txpNameArchive(txpNameStream);
txpNameArchive(txpNames);
logger->info("[{}] transcripts in index ", txpNames.size());
logger->info("done ");
}
txpNameStream.close();
std::ifstream txpLenStream(txpLenFile, std::ios::binary);
{
logger->info("loading transcript lengths");
ScopedTimer timer;
cereal::BinaryInputArchive txpLenArchive(txpLenStream);
txpLenArchive(txpLens);
logger->info("[{}] transcripts in index", txpLens.size());
logger->info("done ");
}
txpLenStream.close();
std::ifstream fwdJumpStream(fwdJumpFile, std::ios::binary);
{
logger->info("loading forward jumps");
ScopedTimer timer;
cereal::BinaryInputArchive fwdJumpArchive(fwdJumpStream);
fwdJumpArchive(fwdJumpTable);
logger->info("[{}] forward jumps", fwdJumpTable.size());
logger->info("done ");
}
fwdJumpStream.close();
std::ifstream revJumpStream(revJumpFile, std::ios::binary);
{
logger->info("loading forward jumps");
ScopedTimer timer;
cereal::BinaryInputArchive revJumpArchive(revJumpStream);
revJumpArchive(revJumpTable);
logger->info("[{}] reverse jumps", revJumpTable.size());
logger->info("done ");
}
revJumpStream.close();
return true;
}
void StandardStrings::convertLasToRms(QWidget *parent, const QString& lasFileName, const QString& rmsFileName, const QString& posFileName){
//this assumes that MD/Z is the first log in the las file.
QFile lasFile(lasFileName);
QFile rmsFile(rmsFileName);
QFile posFile(posFileName);
if (!lasFile.open(QFile::ReadOnly | QFile::Text)) {
QMessageBox::warning(parent, QString("Unable to open file"), QString("Could not open the file."), QMessageBox::Ok);
return;
}
if (!posFile.open(QFile::ReadOnly | QFile::Text)) {
QMessageBox::warning(parent, QString("Unable to open file"), QString("Could not open the file."), QMessageBox::Ok);
return;
}
if (!rmsFile.open(QFile::WriteOnly | QFile::Text)) {
QMessageBox::warning(parent, QString("Unable to open file"), QString("Could not open the file."), QMessageBox::Ok);
return;
}
QString wellName;
QTextStream lasStream(&lasFile);
QTextStream rmsStream(&rmsFile);
QTextStream posStream(&posFile);
QString line = posStream.readLine();
while(!line.isNull()){
if(!line.isEmpty()){
if(line[0]==QChar('-')){
line = posStream.readLine();
break;
}
}
line = posStream.readLine();
}
QList<double> xList;
QList<double> yList;
QList<double> mdList;
//
while(!line.isNull()){
QStringList list =line.simplified().split(" ");
xList << list[0].toDouble();
yList << list[1].toDouble();
mdList << list[5].toDouble();
line = posStream.readLine();
}
line = lasStream.readLine();
while(!line.isNull()){
if(line.split(" ").at(0)==QString("WELL")){
wellName=line.split("\"").at(1);
while(!wellName[0].isDigit()){
wellName.remove(0,1);
if(wellName.isEmpty()){
QMessageBox::warning(parent, QString("No well name"), QString("Could not find a valid well name."), QMessageBox::Ok);
return;
}
}
wellName=wellName.remove(" ");
break;
}
line = lasStream.readLine();
}
if(wellName.isEmpty()){
QMessageBox::warning(parent, QString("No well name"), QString("Could not find a valid well name."), QMessageBox::Ok);
return;
}
QString wellX;
while(!line.isNull()){
if(line.split(" ").at(0)==QString("XWELL")){
QStringList wellXList=line.split(" ");
for(int i=0;i<wellXList.size();++i){
if(!wellXList[i].isEmpty()){
bool ok;
wellXList[i].toDouble(&ok);
if(ok){
wellX=wellXList[i];
}
}
}
break;
}
line = lasStream.readLine();
}
if(wellX.isEmpty()){
QMessageBox::warning(parent, QString("No well x-coordinate"), QString("Could not find a valid well x-coordinate."), QMessageBox::Ok);
return;
}
QString wellY;
while(!line.isNull()){
if(line.split(" ").at(0)==QString("YWELL")){
QStringList wellYList=line.split(" ");
for(int i=0;i<wellYList.size();++i){
if(!wellYList[i].isEmpty()){
bool ok;
wellYList[i].toDouble(&ok);
if(ok){
wellY=wellYList[i];
}
}
}
break;
}
line = lasStream.readLine();
}
if(wellY.isEmpty()){
QMessageBox::warning(parent, QString("No well x-coordinate"), QString("Could not find a valid well x-coordinate."), QMessageBox::Ok);
return;
}
while(!line.isNull()){
if(!line.isEmpty()){
if(line[0]==QChar('~')){
lasStream.readLine();//read away the comments
lasStream.readLine();
break;
}
}
line = lasStream.readLine();
}
QStringList logNames;
lasStream.readLine();
line = lasStream.readLine();
while(!line.isEmpty()){
logNames << line.split(".")[0].remove(" ");
line = lasStream.readLine();
}
lasStream.readLine();//~ASCII
line = lasStream.readLine();
int minIndex=0;
rmsStream.QTextStream::setRealNumberNotation(QTextStream::FixedNotation);
if(wellX.size()>wellY.size()){
rmsStream.setRealNumberPrecision(wellX.size());
}
else{
rmsStream.setRealNumberPrecision(wellY.size());
}
rmsStream.setRealNumberPrecision(2);
rmsStream << 1.0 << endl;
rmsStream << "undefined" << endl;
rmsStream << wellName << " " << wellX << " " << wellY << endl;
rmsStream << logNames.size() << endl;
for(int i=0;i<logNames.size();++i){
rmsStream << logNames[i] << " " << "UNK lin" << endl;
}
bool first=true;
while(!line.isEmpty()){
if(!first){
rmsStream << endl;
}
else{
first=false;
}
bool printed=false;
QStringList lineList = line.simplified().split(" ");
for(int j=minIndex;j<mdList.size();++j){
if(mdList[j]==lineList.at(0).toDouble()){
rmsStream << xList[j] << " " << yList[j];
for(int i=0;i<lineList.size();++i){
rmsStream << " " << lineList.at(i);
}
printed=true;
break;
} else if(mdList[j]<lineList.at(0).toDouble()){
minIndex=j;
} else{
bool overRule=false;
while(mdList[j]==mdList[minIndex]){
if(j<mdList.size()-1){
++j;
}
else {
overRule=true;
break;
}
}
double fractionDifference=(lineList.at(0).toDouble()-mdList[minIndex])/(mdList[j]-mdList[minIndex]);
if(overRule){
fractionDifference=0;
}
rmsStream << (xList[j]+fractionDifference*(xList[j]-xList[minIndex])) << " ";
rmsStream << (yList[j]+fractionDifference*(yList[j]-yList[minIndex]));
for(int i=0;i<lineList.size();++i){
rmsStream << " " << lineList.at(i);
}
printed=true;
break;
}
}
if(!printed){
rmsStream << (xList[xList.size()-1]) << " " << (yList[yList.size()-1]);
for(int i=0;i<lineList.size();++i){
rmsStream << " " << lineList.at(i);
}
}
line = lasStream.readLine();
}
}
