ExitCodes main_(int, const char**)
{
// instance specific location of settings in INI file (e.g. 'TOPP_Skeleton:1:')
String ini_location;
// path to the log file
String logfile(getStringOption_("log"));
String xtandem_executable(getStringOption_("xtandem_executable"));
String inputfile_name;
String outputfile_name;
//-------------------------------------------------------------
// parsing parameters
//-------------------------------------------------------------
inputfile_name = getStringOption_("in");
writeDebug_(String("Input file: ") + inputfile_name, 1);
if (inputfile_name == "")
{
writeLog_("No input file specified. Aborting!");
printUsage_();
return ILLEGAL_PARAMETERS;
}
outputfile_name = getStringOption_("out");
writeDebug_(String("Output file: ") + outputfile_name, 1);
if (outputfile_name == "")
{
writeLog_("No output file specified. Aborting!");
printUsage_();
return ILLEGAL_PARAMETERS;
}
// write input xml file
String temp_directory = QDir::toNativeSeparators((File::getTempDirectory() + "/" + File::getUniqueName() + "/").toQString()); // body for the tmp files
{
QDir d;
d.mkpath(temp_directory.toQString());
}
String input_filename(temp_directory + "_tandem_input_file.xml");
String tandem_input_filename(temp_directory + "_tandem_input_file.mzData");
String tandem_output_filename(temp_directory + "_tandem_output_file.xml");
String tandem_taxonomy_filename(temp_directory + "_tandem_taxonomy_file.xml");
//-------------------------------------------------------------
// Validate user parameters
//-------------------------------------------------------------
if (getIntOption_("min_precursor_charge") > getIntOption_("max_precursor_charge"))
{
LOG_ERROR << "Given charge range is invalid: max_precursor_charge needs to be >= min_precursor_charge." << std::endl;
return ILLEGAL_PARAMETERS;
}
//-------------------------------------------------------------
// reading input
//-------------------------------------------------------------
String db_name(getStringOption_("database"));
if (!File::readable(db_name))
{
String full_db_name;
try
{
full_db_name = File::findDatabase(db_name);
}
catch (...)
{
printUsage_();
return ILLEGAL_PARAMETERS;
}
db_name = full_db_name;
}
PeakMap exp;
MzMLFile mzml_file;
mzml_file.getOptions().addMSLevel(2); // only load msLevel 2
mzml_file.setLogType(log_type_);
mzml_file.load(inputfile_name, exp);
if (exp.getSpectra().empty())
{
throw OpenMS::Exception::FileEmpty(__FILE__, __LINE__, __FUNCTION__, "Error: No MS2 spectra in input file.");
}
// determine type of spectral data (profile or centroided)
SpectrumSettings::SpectrumType spectrum_type = exp[0].getType();
if (spectrum_type == SpectrumSettings::RAWDATA)
{
if (!getFlag_("force"))
{
throw OpenMS::Exception::IllegalArgument(__FILE__, __LINE__, __FUNCTION__, "Error: Profile data provided but centroided MS2 spectra expected. To enforce processing of the data set the -force flag.");
}
}
// we need to replace the native id with a simple numbering schema, to be able to
// map the IDs back to the spectra (RT, and MZ information)
Size native_id(0);
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
it->setNativeID(++native_id);
}
// We store the file in mzData file format, because MGF files somehow produce in most
// of the cases IDs with charge 2+. We do not use the input file directly
// because XTandem sometimes stumbles over misleading substrings in the filename,
// e.g. mzXML ...
MzDataFile mzdata_outfile;
mzdata_outfile.store(tandem_input_filename, exp);
XTandemInfile infile;
infile.setInputFilename(tandem_input_filename);
infile.setOutputFilename(tandem_output_filename);
ofstream tax_out(tandem_taxonomy_filename.c_str());
tax_out << "<?xml version=\"1.0\"?>" << "\n";
tax_out << "\t<bioml label=\"x! taxon-to-file matching list\">" << "\n";
tax_out << "\t\t<taxon label=\"OpenMS_dummy_taxonomy\">" << "\n";
tax_out << "\t\t\t<file format=\"peptide\" URL=\"" << db_name << "\" />" << "\n";
tax_out << "\t</taxon>" << "\n";
tax_out << "</bioml>" << "\n";
tax_out.close();
infile.setTaxonomyFilename(tandem_taxonomy_filename);
if (getStringOption_("precursor_error_units") == "Da")
{
infile.setPrecursorMassErrorUnit(XTandemInfile::DALTONS);
}
else
{
infile.setPrecursorMassErrorUnit(XTandemInfile::PPM);
}
if (getStringOption_("fragment_error_units") == "Da")
{
infile.setFragmentMassErrorUnit(XTandemInfile::DALTONS);
}
else
{
infile.setFragmentMassErrorUnit(XTandemInfile::PPM);
}
if (getStringOption_("default_input_file") != "")
{
infile.load(getStringOption_("default_input_file"));
infile.setDefaultParametersFilename(getStringOption_("default_input_file"));
}
else
{
String default_file = File::find("CHEMISTRY/XTandem_default_input.xml");
infile.load(default_file);
infile.setDefaultParametersFilename(default_file);
}
infile.setPrecursorMassTolerancePlus(getDoubleOption_("precursor_mass_tolerance"));
infile.setPrecursorMassToleranceMinus(getDoubleOption_("precursor_mass_tolerance"));
infile.setFragmentMassTolerance(getDoubleOption_("fragment_mass_tolerance"));
infile.setMaxPrecursorCharge(getIntOption_("max_precursor_charge"));
infile.setNumberOfThreads(getIntOption_("threads"));
infile.setModifications(ModificationDefinitionsSet(getStringList_("fixed_modifications"), getStringList_("variable_modifications")));
infile.setTaxon("OpenMS_dummy_taxonomy");
infile.setOutputResults(getStringOption_("output_results"));
infile.setMaxValidEValue(getDoubleOption_("max_valid_expect"));
infile.setCleavageSite(getStringOption_("cleavage_site"));
infile.setNumberOfMissedCleavages(getIntOption_("missed_cleavages"));
infile.setRefine(getFlag_("refinement"));
infile.setSemiCleavage(getFlag_("semi_cleavage"));
bool allow_isotope_error = getStringOption_("allow_isotope_error") == "yes" ? true : false;
infile.setAllowIsotopeError(allow_isotope_error);
infile.write(input_filename);
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
int status = QProcess::execute(xtandem_executable.toQString(), QStringList(input_filename.toQString())); // does automatic escaping etc...
if (status != 0)
{
writeLog_("XTandem problem. Aborting! Calling command was: '" + xtandem_executable + " \"" + input_filename + "\"'.\nDoes the !XTandem executable exist?");
// clean temporary files
if (this->debug_level_ < 2)
{
File::removeDirRecursively(temp_directory);
LOG_WARN << "Set debug level to >=2 to keep the temporary files at '" << temp_directory << "'" << std::endl;
}
else
{
LOG_WARN << "Keeping the temporary files at '" << temp_directory << "'. Set debug level to <2 to remove them." << std::endl;
}
return EXTERNAL_PROGRAM_ERROR;
}
vector<ProteinIdentification> protein_ids;
ProteinIdentification protein_id;
vector<PeptideIdentification> peptide_ids;
// read the output of X!Tandem and write it to idXML
XTandemXMLFile tandem_output;
tandem_output.setModificationDefinitionsSet(ModificationDefinitionsSet(getStringList_("fixed_modifications"), getStringList_("variable_modifications")));
// find the file, because XTandem extends the filename with a timestamp we do not know (exactly)
StringList files;
File::fileList(temp_directory, "_tandem_output_file*.xml", files);
if (files.size() != 1)
{
throw Exception::FileNotFound(__FILE__, __LINE__, __PRETTY_FUNCTION__, tandem_output_filename);
}
tandem_output.load(temp_directory + files[0], protein_id, peptide_ids);
// now put the RTs into the peptide_ids from the spectrum ids
for (vector<PeptideIdentification>::iterator it = peptide_ids.begin(); it != peptide_ids.end(); ++it)
{
UInt id = (Int)it->getMetaValue("spectrum_id");
--id; // native IDs were written 1-based
if (id < exp.size())
{
it->setRT(exp[id].getRT());
double pre_mz(0.0);
if (!exp[id].getPrecursors().empty()) pre_mz = exp[id].getPrecursors()[0].getMZ();
it->setMZ(pre_mz);
//it->removeMetaValue("spectrum_id");
}
else
{
LOG_ERROR << "XTandemAdapter: Error: id '" << id << "' not found in peak map!" << endl;
}
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
// handle the search parameters
ProteinIdentification::SearchParameters search_parameters;
search_parameters.db = getStringOption_("database");
search_parameters.charges = "+" + String(getIntOption_("min_precursor_charge")) + "-+" + String(getIntOption_("max_precursor_charge"));
ProteinIdentification::PeakMassType mass_type = ProteinIdentification::MONOISOTOPIC;
search_parameters.mass_type = mass_type;
search_parameters.fixed_modifications = getStringList_("fixed_modifications");
search_parameters.variable_modifications = getStringList_("variable_modifications");
search_parameters.missed_cleavages = getIntOption_("missed_cleavages");
search_parameters.peak_mass_tolerance = getDoubleOption_("fragment_mass_tolerance");
search_parameters.precursor_tolerance = getDoubleOption_("precursor_mass_tolerance");
protein_id.setSearchParameters(search_parameters);
protein_id.setSearchEngineVersion("");
protein_id.setSearchEngine("XTandem");
protein_ids.push_back(protein_id);
IdXMLFile().store(outputfile_name, protein_ids, peptide_ids);
/// Deletion of temporary files
if (this->debug_level_ < 2)
{
File::removeDirRecursively(temp_directory);
LOG_WARN << "Set debug level to >=2 to keep the temporary files at '" << temp_directory << "'" << std::endl;
}
else
{
LOG_WARN << "Keeping the temporary files at '" << temp_directory << "'. Set debug level to <2 to remove them." << std::endl;
}
// some stats
LOG_INFO << "Statistics:\n"
<< " identified MS2 spectra: " << peptide_ids.size() << " / " << exp.size() << " = " << int(peptide_ids.size() * 100.0 / exp.size()) << "% (with e-value < " << String(getDoubleOption_("max_valid_expect")) << ")" << std::endl;
return EXECUTION_OK;
}
ExitCodes main_(int, const char**)
{
// parsing parameters
String in(getStringOption_("in"));
String feature_in(getStringOption_("feature_in"));
String out(getStringOption_("out"));
double precursor_mass_tolerance(getDoubleOption_("precursor_mass_tolerance"));
// reading input
FileHandler fh;
FileTypes::Type in_type = fh.getType(in);
PeakMap exp;
fh.loadExperiment(in, exp, in_type, log_type_, false, false);
exp.sortSpectra();
FeatureMap feature_map;
if (feature_in != "")
{
FeatureXMLFile().load(feature_in, feature_map);
}
// calculations
FeatureFinderAlgorithmIsotopeWavelet iso_ff;
Param ff_param(iso_ff.getParameters());
ff_param.setValue("max_charge", getIntOption_("max_charge"));
ff_param.setValue("intensity_threshold", getDoubleOption_("intensity_threshold"));
iso_ff.setParameters(ff_param);
FeatureFinder ff;
ff.setLogType(ProgressLogger::NONE);
PeakMap exp2 = exp;
exp2.clear(false);
for (PeakMap::ConstIterator it = exp.begin(); it != exp.end(); ++it)
{
if (it->size() != 0)
{
exp2.addSpectrum(*it);
}
}
exp = exp2;
exp.updateRanges();
// TODO check MS2 and MS1 counts
ProgressLogger progresslogger;
progresslogger.setLogType(log_type_);
progresslogger.startProgress(0, exp.size(), "Correcting precursor masses");
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
progresslogger.setProgress(exp.end() - it);
if (it->getMSLevel() != 2)
{
continue;
}
// find first MS1 scan of the MS/MS scan
PeakMap::Iterator ms1_it = it;
while (ms1_it != exp.begin() && ms1_it->getMSLevel() != 1)
{
--ms1_it;
}
if (ms1_it == exp.begin() && ms1_it->getMSLevel() != 1)
{
writeLog_("Did not find a MS1 scan to the MS/MS scan at RT=" + String(it->getRT()));
continue;
}
if (ms1_it->size() == 0)
{
writeDebug_("No peaks in scan at RT=" + String(ms1_it->getRT()) + String(", skipping"), 1);
continue;
}
PeakMap::Iterator ms2_it = ms1_it;
++ms2_it;
while (ms2_it != exp.end() && ms2_it->getMSLevel() == 2)
{
// first: error checks
if (ms2_it->getPrecursors().empty())
{
writeDebug_("Warning: found no precursors of spectrum RT=" + String(ms2_it->getRT()) + ", skipping it.", 1);
++ms2_it;
continue;
}
else if (ms2_it->getPrecursors().size() > 1)
{
writeLog_("Warning: found more than one precursor of spectrum RT=" + String(ms2_it->getRT()) + ", using first one.");
}
Precursor prec = *ms2_it->getPrecursors().begin();
double prec_pos = prec.getMZ();
PeakMap new_exp;
// now excise small region from the MS1 spec for the feature finder (isotope pattern must be covered...)
PeakSpectrum zoom_spec;
for (PeakSpectrum::ConstIterator pit = ms1_it->begin(); pit != ms1_it->end(); ++pit)
{
if (pit->getMZ() > prec_pos - 3 && pit->getMZ() < prec_pos + 3)
{
zoom_spec.push_back(*pit);
}
}
new_exp.addSpectrum(zoom_spec);
new_exp.updateRanges();
FeatureMap features, seeds;
ff.run("isotope_wavelet", new_exp, features, ff_param, seeds);
if (features.empty())
{
writeDebug_("No features found for scan RT=" + String(ms1_it->getRT()), 1);
++ms2_it;
continue;
}
double max_int(numeric_limits<double>::min());
double min_dist(numeric_limits<double>::max());
Size max_int_feat_idx(0);
for (Size i = 0; i != features.size(); ++i)
{
if (fabs(features[i].getMZ() - prec_pos) < precursor_mass_tolerance &&
features[i].getIntensity() > max_int)
{
max_int_feat_idx = i;
max_int = features[i].getIntensity();
min_dist = fabs(features[i].getMZ() - prec_pos);
}
}
writeDebug_(" max_int=" + String(max_int) + " mz=" + String(features[max_int_feat_idx].getMZ()) + " charge=" + String(features[max_int_feat_idx].getCharge()), 5);
if (min_dist < precursor_mass_tolerance)
{
prec.setMZ(features[max_int_feat_idx].getMZ());
prec.setCharge(features[max_int_feat_idx].getCharge());
vector<Precursor> precs;
precs.push_back(prec);
ms2_it->setPrecursors(precs);
writeDebug_("Correcting precursor mass of spectrum RT=" + String(ms2_it->getRT()) + " from " + String(prec_pos) + " to " + String(prec.getMZ()) + " (z=" + String(prec.getCharge()) + ")", 1);
}
++ms2_it;
}
it = --ms2_it;
}
progresslogger.endProgress();
// writing output
fh.storeExperiment(out, exp, log_type_);
return EXECUTION_OK;
}
ExitCodes main_(int, const char **) override
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
String in = getStringOption_("in");
String out = getStringOption_("out");
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
PeakMap exp;
MzMLFile f;
f.load(in, exp);
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
//determine maximum peak
exp.updateRanges();
double max = exp.getMaxInt() / 100.0;
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
if (it->getMSLevel() < 2)
{
for (PeakMap::SpectrumType::Iterator it2 = it->begin(); it2 != it->end(); ++it2)
{
it2->setIntensity(it2->getIntensity() / max);
}
}
}
/// @todo add chromatogram support for normalization, e.g. for MRM stuff (Andreas)
/*
vector<MSChromatogram > chroms = exp.getChromatograms();
double sum(0);
for (vector<MSChromatogram >::iterator it = chroms.begin(); it != chroms.end(); ++it)
{
for (MSChromatogram::Iterator it2 = it->begin(); it2 != it->end(); ++it2)
{
sum += it2->getIntensity();
}
}
for (vector<MSChromatogram >::iterator it = chroms.begin(); it != chroms.end(); ++it)
{
for (MSChromatogram::Iterator it2 = it->begin(); it2 != it->end(); ++it2)
{
it2->setIntensity(it2->getIntensity() / sum * 1000000.0);
}
}
exp.setChromatograms(chroms);
*/
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(exp, getProcessingInfo_(DataProcessing::NORMALIZATION));
f.store(out, exp);
return EXECUTION_OK;
}
