//1100110 record
//------------------------------------------------------------------------------
int main(int argc, char *argv[]){
int i;
if(! std_input_check(argc, argv))
return EXIT_FAILURE;
gl_pageID = 1;
init_programs();
for (i=0 ; i<gl_program_count; i++){
printf("pid:%i \t initML: %i \t totalML:%i \t totalPages: %i P#:%i \t Flag:%i \n",\
gl_programs[i].pid, \
gl_programs[i].init_memorylocation, \
gl_programs[i].total_memorylocation, \
gl_programs[i].total_pages, \
gl_programs[i].pagetable[gl_programs[0].init_memorylocation].pagenumber,\
gl_programs[i].pagetable[gl_programs[0].init_memorylocation].validbid);
//gl_programs[0] because it was the border point
}
run_programs();
}
void Biceps::readMGF()
{
std::string filename = genOp_.mgfAbsPath.string();
std::cout <<"Now reading " << filename << std::endl;
fasta_.setNumTags(genOp_.numTags);
std::string linebuffer; //< buffering the textlines in the mgf file
bool validSpectrum = true; //<So we can just skip if it's invalid.
//Now checking for the filetype.
bool dosformat = isDosFile(filename);
std::ifstream mgfFile;
mgfFile.open(filename.c_str(), ios::binary); //<The Filename of the spectraFile
bool inBeginIons = 0; bool inPeakList = 0; //status, where are we in the spectrum?
//Only for buffer reasons, what's the charge etc. of the current spectrum?
int charge = 0;
float precursormass = 0;
string title="";
std::vector<int> indices; //to remember the corresponding spectrum-indices of valid results
std::vector<string> titles; //remember the spectrum titles
size_t numGoodscores = 0;
ofstream bufferfile;
ofstream results;
string tempresultname = std::string("Biceps.tempResults.") + genOp_.mgfShortName + string(".txt");
results.open(tempresultname.c_str(), ios::binary); //Biceps_results will handle the final output.
while(std::getline(mgfFile, linebuffer, '\n') )
{
//first case, we're at the beginning of a new spectrum-subsection
if(linebuffer.substr(0, 8) == "BEGIN IO"){
mgfId_++;
if (inBeginIons)
{
//throw runtime_error(("BEGIN IONS tag found without previous BEGIN IONS being closed at" + lexical_cast<string>(size_t(mgfFile.tellg())-linebuffer.length()-1) + "\n"));
cout << "BEGIN IONS tag found without previous BEGIN IONS being closed at" + lexical_cast<string>(size_t(mgfFile.tellg())-linebuffer.length()-1) + "\n" << std::endl;
cout << "will begin new spectrum here" << std::endl;
bufferfile.close();
}
validSpectrum = true;
inBeginIons = true;
if (bufferfile.is_open()) bufferfile.close();
bufferfile.open(genOp_.sSpectrumFN.c_str(), fstream::trunc | ios::binary);
if (!bufferfile.good()) { cerr << "Problem writing to buffer, check your writing rights please." << std::endl;}
if (dosformat)
{
bufferfile << linebuffer.substr(0,linebuffer.size()-1).c_str() << "\n";
}
else
{
bufferfile << linebuffer.c_str() << "\n";
}
//continue;
} //if BEGIN IO - Beginning of a spectrum
//second case
else if (linebuffer.substr(0, 8) != ("END IONS")) //we are in a spectrum, parse the lines.
{
if (!inBeginIons) continue;
try
{
if (!inPeakList)
{
if (linebuffer.substr(0, 6) == ("TITLE="))
{
// if a title is found, use it as the id instead of the index
//spectrum.id = lineStr.substr(6);
title = linebuffer.substr(6);
titles.push_back(title);
}
else if (linebuffer.substr(0, 8) == ("PEPMASS="))
{
string pepMassStr = linebuffer.substr(8);
pepMassStr = pepMassStr.substr(0,pepMassStr.find_first_of("\t "));
//bal::trim(pepMassStr);
precursormass= lexical_cast<float>(pepMassStr);
linebuffer = linebuffer.substr(0,linebuffer.find_first_of("\t "));
// selectedIon.set(MS_m_z, mz);
}
else if (linebuffer.substr(0, 7) == ("CHARGE="))
{
string pepChargeStr = linebuffer.substr(7);
//bal::trim_if(pepChargeStr, bal::is_any_of("+- \t\r"));
size_t rest = pepChargeStr.find_first_not_of("0123456789",0);
if (rest != string::npos)
pepChargeStr.erase(rest);
charge = lexical_cast<size_t>(pepChargeStr);
if (charge > 3)
{
std::cout << "chargevalue " << charge << " is unhandled, skipping." << std::endl;
validSpectrum = false;
}
// selectedIon.set(MS_charge_state, charge);
}
else if(linebuffer.find('=') != string::npos)
{
continue; // ignored attribute
}
else
{
if (inBeginIons)
inPeakList = true;
}
} //if (!inPeakList)
} // if try
catch(const std::exception& e)
{
//throw runtime_error(("[SpectrumList_MGF::parseSpectrum] Error parsing line at offset " +
// lexical_cast<string>(size_t(mgfFile.tellg())-linebuffer.length()-1) + ": " + linebuffer + "\n"));
cout << "[SpectrumList_MGF::parseSpectrum] Error parsing line at offset " +
lexical_cast<string>(size_t(mgfFile.tellg())-linebuffer.length()-1) + ": " + linebuffer + "\n" << std::endl;
inBeginIons = false;
validSpectrum = false;
}
if (dosformat)
{
bufferfile << linebuffer.substr(0,linebuffer.size()-1).c_str() << "\n";
}
else
{
bufferfile << linebuffer.c_str() << "\n";
}
//continue;
} //else if (not end of spectrum)
else //END IONS
{
//We found the last line of a spectrum, now it's time to finish the mgf buffer file and
//do pepnovo/directag and pepsplice afterwards.
if (!inBeginIons)
{
//incorrect mgf file, please fix.
//throw runtime_error(("END IONS tag found without opening BEGIN IONS at" + lexical_cast<string>(size_t(mgfFile.tellg())-linebuffer.length()-1) + "\n")); //
std::cout << "END IONS tag found without opening BEGIN IONS at" + lexical_cast<string>(size_t(mgfFile.tellg())-linebuffer.length()-1) + "\n" << std::endl;
validSpectrum = false;
}
inBeginIons = false;
inPeakList = false;
if (dosformat)
{
bufferfile << linebuffer.substr(0,linebuffer.size()-1).c_str() << "\n";
}
else
{
bufferfile << linebuffer.c_str() << "\n";
}
bufferfile.close();
if (validSpectrum == false) //either charge too high or spectrum invalid
{
continue; //just go to the next line
}
fasta_.initializeMGF(charge, precursormass);
cout << "Analyzing spectrum " << mgfId_ << endl;
//Start analyzing by calling run_programs, which will run directag/pepnovo and then pepsplice afterwards
bool success = run_programs(); //if an actual result was found, return true, else return false;
if (success){
Pepsplice::PepspliceResult & res = pepResults_.back();
indices.push_back(mgfId_);
titles.push_back(title);
//if it's good, write the result to a file
writeResult(results, res, mgfId_, title);
//pepres.pop_back();
++numGoodscores;
}
} //END IONS, parsed one spectrum, call other programs //end else
} //while File.good
assert(pepResults_.size() == numGoodscores);//check if all the results are in pepres
ofstream resfile;
string resfileName = string("Biceps.gmm.") + genOp_.mgfShortName + string(".txt");
resfile.open(resfileName.c_str(), ios::trunc | ios::binary);
//temporary score-output to call bic on.
for(size_t i = 0; i < pepResults_.size(); ++i)
{
resfile << pepResults_[i].score << "\n";
}
resfile << std::flush;
resfile.close();
//now beginning BIC part
std::vector<double> mu,sigma;
std::vector<int> labels;
double cutoff = -99999999;
//final output
if (pepResults_.size() > 10){
gmm_bic(2,numGoodscores,resfileName.c_str(), mu, sigma, labels);
cutoff = findCutoff(mu, sigma, 2);
}
Biceps::writeCompleteResult(pepResults_, indices, titles, labels, mu, sigma, cutoff);
Biceps::writeFasta(pepResults_);
}
