void Normalizer::filterPeakMap(PeakMap & exp)
{
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
filterSpectrum(*it);
}
}
void CompNovoIdentificationCID::getIdentifications(vector<PeptideIdentification> & pep_ids, const PeakMap & exp)
{
Size count(1);
for (PeakMap::ConstIterator it = exp.begin(); it != exp.end(); ++it, ++count)
{
//cerr << count << "/" << exp.size() << endl;
PeptideIdentification id;
// TODO check if both CID and ETD is present;
PeakSpectrum CID_spec(*it);
id.setRT(it->getRT());
id.setMZ(it->getPrecursors().begin()->getMZ());
subspec_to_sequences_.clear();
permute_cache_.clear();
decomp_cache_.clear();
getIdentification(id, CID_spec);
//cerr << "size_of id=" << id.getHits().size() << endl;
pep_ids.push_back(id);
//++it;
//
//if (count == 10)
//{
//return;
//}
}
return;
}
void MassTraceDetection::run(PeakMap::ConstAreaIterator& begin,
PeakMap::ConstAreaIterator& end,
std::vector<MassTrace>& found_masstraces)
{
PeakMap map;
MSSpectrum<Peak1D> current_spectrum;
if (begin == end)
{
return;
}
for (; begin != end; ++begin)
{
// AreaIterator points on novel spectrum?
if (begin.getRT() != current_spectrum.getRT())
{
// save new spectrum in map
if (current_spectrum.getRT() != -1)
{
map.addSpectrum(current_spectrum);
}
current_spectrum.clear(false);
current_spectrum.setRT(begin.getRT());
}
current_spectrum.push_back(*begin);
}
map.addSpectrum(current_spectrum);
run(map, found_masstraces);
}
void ParentPeakMower::filterPeakMap(PeakMap & exp)
{
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
filterSpectrum(*it);
}
}
// lists of peptide hits in "maps" will be sorted
bool MapAlignmentAlgorithmIdentification::getRetentionTimes_(
PeakMap& experiment, SeqToList& rt_data)
{
for (PeakMap::Iterator exp_it = experiment.begin();
exp_it != experiment.end(); ++exp_it)
{
getRetentionTimes_(exp_it->getPeptideIdentifications(), rt_data);
}
// duplicate annotations should not be possible -> no need to remove them
return false;
}
void MzMLFile::load(const String& filename, PeakMap& map)
{
map.reset();
//set DocumentIdentifier
map.setLoadedFileType(filename);
map.setLoadedFilePath(filename);
Internal::MzMLHandler handler(map, filename, getVersion(), *this);
handler.setOptions(options_);
safeParse_(filename, &handler);
}
void WindowMower::filterPeakMap(PeakMap & exp)
{
bool sliding = (String)param_.getValue("movetype") == "slide" ? true : false;
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
if (sliding)
{
filterPeakSpectrumForTopNInSlidingWindow(*it);
} else
{
filterPeakSpectrumForTopNInJumpingWindow(*it);
}
}
}
ExitCodes main_(int, const char **)
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
String in_spectra = getStringOption_("in_spectra");
String in_identifications = getStringOption_("in_identifications");
String outfile = getStringOption_("model_output_file");
Int precursor_charge = getIntOption_("precursor_charge");
//-------------------------------------------------------------
// init SvmTheoreticalSpectrumGeneratorTrainer
//-------------------------------------------------------------
SvmTheoreticalSpectrumGeneratorTrainer trainer;
Param param = getParam_().copy("algorithm:", true);
String write_files = getFlag_("write_training_files") ? "true" : "false";
param.setValue("write_training_files", write_files);
trainer.setParameters(param);
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
PeakMap map;
MzMLFile().load(in_spectra, map);
std::vector<PeptideIdentification> pep_ids;
std::vector<ProteinIdentification> prot_ids;
String tmp_str;
IdXMLFile().load(in_identifications, prot_ids, pep_ids, tmp_str);
IDMapper idmapper;
Param par;
par.setValue("rt_tolerance", 0.001);
par.setValue("mz_tolerance", 0.001);
idmapper.setParameters(par);
idmapper.annotate(map, pep_ids, prot_ids);
//generate vector of annotations
std::vector<AASequence> annotations;
PeakMap::iterator it;
for (it = map.begin(); it != map.end(); ++it)
{
annotations.push_back(it->getPeptideIdentifications()[0].getHits()[0].getSequence());
}
trainer.trainModel(map, annotations, outfile, precursor_charge);
return EXECUTION_OK;
}
ExitCodes main_(int, const char **) override
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
//input/output files
String in(getStringOption_("in"));
String out(getStringOption_("out"));
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
PeakMap exp;
MzMLFile f;
f.setLogType(log_type_);
f.load(in, exp);
//-------------------------------------------------------------
// if meta data arrays are present, remove them and warn
//-------------------------------------------------------------
if (exp.clearMetaDataArrays())
{
writeLog_("Warning: Spectrum meta data arrays cannot be sorted. They are deleted.");
}
//-------------------------------------------------------------
// filter
//-------------------------------------------------------------
Param filter_param = getParam_().copy("algorithm:", true);
writeDebug_("Used filter parameters", filter_param, 3);
BernNorm filter;
filter.setParameters(filter_param);
filter.filterPeakMap(exp);
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(exp, getProcessingInfo_(DataProcessing::FILTERING));
f.store(out, exp);
return EXECUTION_OK;
}
CachedmzML cacheFile(std::string & tmp_filename, PeakMap& exp)
{
NEW_TMP_FILE(tmp_filename);
// Load experiment
MzMLFile().load(OPENMS_GET_TEST_DATA_PATH("MzMLFile_1.mzML"), exp);
TEST_EQUAL(exp.getNrSpectra() > 0, true)
TEST_EQUAL(exp.getNrChromatograms() > 0, true)
// Cache the experiment to a temporary file
CachedmzML cache;
cache.writeMemdump(exp, tmp_filename);
// Create the index from the given file
cache.createMemdumpIndex(tmp_filename);
return cache;
}
bool IDEvaluationBase::addSearchFile(const String& file_name)
{
MSSpectrum<> points;
if (!loadCurve(file_name, points)) return false;
data_.addSpectrum(points);
PeakMap* exp = new PeakMap();
exp->addSpectrum(points);
spec_1d_->canvas()->addLayer(SpectrumCanvas::ExperimentSharedPtrType(exp));
spec_1d_->canvas()->setLayerName(spec_1d_->canvas()->getLayerCount() - 1, points.getMetaValue("search_engine"));
// set intensity mode (after spectrum has been added!)
setIntensityMode((int) SpectrumCanvas::IM_SNAP);
return true;
}
// version for label-free linkers
void XQuestResultXMLFile::writeXQuestXMLSpec(String out_file, String base_name, const std::vector< std::vector< OPXLDataStructs::CrossLinkSpectrumMatch > >& all_top_csms, const PeakMap& spectra)
{
// String spec_xml_filename = base_name + "_matched.spec.xml";
// XML Header
std::ofstream spec_xml_file;
std::cout << "Writing spec.xml to " << out_file << std::endl;
spec_xml_file.open(out_file.c_str(), std::ios::trunc); // ios::app = append to file, ios::trunc = overwrites file
// TODO write actual data
spec_xml_file << "<?xml version=\"1.0\" encoding=\"UTF-8\"?><xquest_spectra compare_peaks_version=\"3.4\" date=\"Tue Nov 24 12:41:18 2015\" author=\"Thomas Walzthoeni,Oliver Rinner\" homepage=\"http://proteomics.ethz.ch\" resultdir=\"aleitner_M1012_004_matched\" deffile=\"xquest.def\" >" << std::endl;
// collect indices of spectra, that need to be written out
std::vector <Size> spectrum_indices;
for (Size i = 0; i < all_top_csms.size(); ++i)
{
if (!all_top_csms[i].empty())
{
if (all_top_csms[i][0].scan_index_light < spectra.size())
{
spectrum_indices.push_back(all_top_csms[i][0].scan_index_light);
}
}
}
// loop over list of indices and write out spectra
for (Size i = 0; i < spectrum_indices.size(); ++i)
{
String spectrum_light_name = base_name + ".light." + spectrum_indices[i];
String spectrum_heavy_name = base_name + ".heavy." + spectrum_indices[i];
String spectrum_name = spectrum_light_name + String("_") + spectrum_heavy_name;
// 4 Spectra resulting from a light/heavy spectra pair. Write for each spectrum, that is written to xquest.xml (should be all considered pairs, or better only those with at least one sensible Hit, meaning a score was computed)
spec_xml_file << "<spectrum filename=\"" << spectrum_light_name << ".dta" << "\" type=\"light\">" << std::endl;
spec_xml_file << getxQuestBase64EncodedSpectrum_(spectra[spectrum_indices[i]], String(""));
spec_xml_file << "</spectrum>" << std::endl;
spec_xml_file << "<spectrum filename=\"" << spectrum_heavy_name << ".dta" << "\" type=\"heavy\">" << std::endl;
spec_xml_file << getxQuestBase64EncodedSpectrum_(spectra[spectrum_indices[i]], String(""));
spec_xml_file << "</spectrum>" << std::endl;
String spectrum_common_name = spectrum_name + String("_common.txt");
spec_xml_file << "<spectrum filename=\"" << spectrum_common_name << "\" type=\"common\">" << std::endl;
spec_xml_file << getxQuestBase64EncodedSpectrum_(spectra[spectrum_indices[i]], spectrum_light_name + ".dta," + spectrum_heavy_name + ".dta");
spec_xml_file << "</spectrum>" << std::endl;
String spectrum_xlink_name = spectrum_name + String("_xlinker.txt");
spec_xml_file << "<spectrum filename=\"" << spectrum_xlink_name << "\" type=\"xlinker\">" << std::endl;
spec_xml_file << getxQuestBase64EncodedSpectrum_(spectra[spectrum_indices[i]], spectrum_light_name + ".dta," + spectrum_heavy_name + ".dta");
spec_xml_file << "</spectrum>" << std::endl;
}
spec_xml_file << "</xquest_spectra>" << std::endl;
spec_xml_file.close();
return;
}
vector<vector<Size> > PScore::calculateRankMap(const PeakMap& peak_map, double mz_window)
{
vector<std::vector<Size> > rank_map; // note: ranks are zero based
rank_map.reserve(peak_map.size());
for (Size i = 0; i != peak_map.size(); ++i)
{
const PeakSpectrum& spec = peak_map[i];
vector<double> mz;
vector<double> intensities;
for (Size j = 0; j != spec.size(); ++j)
{
mz.push_back(spec[j].getMZ());
intensities.push_back(spec[j].getIntensity());
}
rank_map.push_back(calculateIntensityRankInMZWindow(mz, intensities, mz_window));
}
return rank_map;
}
Int main(int argc, const char ** argv)
{
if (argc < 2) return 1;
// the path to the data should be given on the command line
String tutorial_data_path(argv[1]);
QApplication app(argc, const_cast<char **>(argv));
PeakMap exp;
exp.resize(1);
DTAFile().load(tutorial_data_path + "/data/Tutorial_Spectrum1D.dta", exp[0]);
LayerData::ExperimentSharedPtrType exp_sptr(new PeakMap(exp));
Spectrum1DWidget * widget = new Spectrum1DWidget(Param(), 0);
widget->canvas()->addLayer(exp_sptr);
widget->show();
return app.exec();
} //end of main
void getPrecursors_(const PeakMap & exp, vector<Precursor> & precursors, vector<double> & precursors_rt)
{
for (Size i = 0; i != exp.size(); ++i)
{
vector<Precursor> pcs = exp[i].getPrecursors();
if (pcs.empty())
{
continue;
}
vector<double> pcs_rt(pcs.size(), exp[i].getRT());
copy(pcs.begin(), pcs.end(), back_inserter(precursors));
copy(pcs_rt.begin(), pcs_rt.end(), back_inserter(precursors_rt));
}
}
void getSwathFile(PeakMap& exp, int nr_swathes=32, bool ms1=true)
{
if (ms1)
{
MSSpectrum s;
s.setMSLevel(1);
Peak1D p; p.setMZ(100); p.setIntensity(200);
s.push_back(p);
exp.addSpectrum(s);
}
for (int i = 0; i< nr_swathes; i++)
{
MSSpectrum s;
s.setMSLevel(2);
std::vector<Precursor> prec(1);
prec[0].setIsolationWindowLowerOffset(12.5);
prec[0].setIsolationWindowUpperOffset(12.5);
prec[0].setMZ(400 + i*25 + 12.5);
s.setPrecursors(prec);
Peak1D p; p.setMZ(101 + i); p.setIntensity(201 + i);
s.push_back(p);
exp.addSpectrum(s);
}
}
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;
}
START_SECTION((static FileTypes::Type getType(const String &filename)))
FileHandler tmp;
TEST_EQUAL(tmp.getType(OPENMS_GET_TEST_DATA_PATH("header_file.h")), FileTypes::UNKNOWN)
TEST_EQUAL(tmp.getType(OPENMS_GET_TEST_DATA_PATH("class_test_infile.txt")), FileTypes::TXT)
TEST_EQUAL(tmp.getType(OPENMS_GET_TEST_DATA_PATH("IdXMLFile_whole.idXML")), FileTypes::IDXML)
TEST_EQUAL(tmp.getType(OPENMS_GET_TEST_DATA_PATH("ConsensusXMLFile.consensusXML")), FileTypes::CONSENSUSXML)
TEST_EQUAL(tmp.getType(OPENMS_GET_TEST_DATA_PATH("TransformationXMLFile_1.trafoXML")), FileTypes::TRANSFORMATIONXML)
TEST_EQUAL(tmp.getType(OPENMS_GET_TEST_DATA_PATH("FileHandler_toppas.toppas")), FileTypes::TOPPAS)
TEST_EQUAL(tmp.getType(OPENMS_GET_TEST_DATA_PATH("pepnovo.txt")), FileTypes::TXT)
TEST_EXCEPTION(Exception::FileNotFound, tmp.getType("/bli/bla/bluff"))
END_SECTION
START_SECTION((template < class PeakType > bool loadExperiment(const String &filename, MSExperiment< PeakType > &exp, FileTypes::Type force_type=FileTypes::UNKNOWN, ProgressLogger::LogType log=ProgressLogger::NONE, const bool compute_hash=true)))
FileHandler tmp;
PeakMap exp;
TEST_EQUAL(tmp.loadExperiment("test.bla", exp), false)
TEST_EQUAL(tmp.loadExperiment(OPENMS_GET_TEST_DATA_PATH("DTAFile_test.dta"), exp), true)
TEST_EQUAL(tmp.loadExperiment(OPENMS_GET_TEST_DATA_PATH("MzDataFile_1.mzData"), exp), true)
TEST_REAL_SIMILAR(exp[1][0].getPosition()[0], 110)
TEST_REAL_SIMILAR(exp[1][1].getPosition()[0], 120)
TEST_REAL_SIMILAR(exp[1][2].getPosition()[0], 130)
// starts with 110, so this one should skip the first
tmp.getOptions().setMZRange(DRange<1>(115, 1000));
TEST_EQUAL(tmp.loadExperiment(OPENMS_GET_TEST_DATA_PATH("MzDataFile_1.mzData"), exp), true)
TEST_REAL_SIMILAR(exp[1][0].getPosition()[0], 120)
TEST_REAL_SIMILAR(exp[1][1].getPosition()[0], 130)
tmp.getOptions() = PeakFileOptions();
// Wrong assignment of the mono-isotopic mass for precursors are assumed:
// - if precursor_mz matches the mz of a non-monoisotopic feature mass trace
// - and in the case that believe_charge is true: if feature_charge matches the precursor_charge
// In the case of wrong mono-isotopic assignment several options for correction are available:
// keep_original will create a copy of the precursor and tandem spectrum for the new mono-isotopic mass trace and retain the original one
// all_matching_features does this not for only the closest feature but all features in a question
set<Size> correctToNearestFeature(const FeatureMap& features, PeakMap & exp, double rt_tolerance_s = 0.0, double mz_tolerance = 0.0, bool ppm = true, bool believe_charge = false, bool keep_original = false, bool all_matching_features = false, int max_trace = 2)
{
set<Size> corrected_precursors;
// for each precursor/MS2 find all features that are in the given tolerance window (bounding box + rt tolerances)
// if believe_charge is set, only add features that match the precursor charge
map<Size, set<Size> > scan_idx_to_feature_idx;
for (Size scan = 0; scan != exp.size(); ++scan)
{
// skip non-tandem mass spectra
if (exp[scan].getMSLevel() != 2 || exp[scan].getPrecursors().empty()) continue;
// extract precusor / MS2 information
const double pc_mz = exp[scan].getPrecursors()[0].getMZ();
const double rt = exp[scan].getRT();
const int pc_charge = exp[scan].getPrecursors()[0].getCharge();
for (Size f = 0; f != features.size(); ++f)
{
// feature is incompatible if believe_charge is set and charges don't match
if (believe_charge && features[f].getCharge() != pc_charge) continue;
// check if precursor/MS2 position overlap with feature
if (overlaps_(features[f], rt, pc_mz, rt_tolerance_s))
{
scan_idx_to_feature_idx[scan].insert(f);
}
}
}
// filter sets to retain compatible features:
// if precursor_mz = feature_mz + n * feature_charge (+/- mz_tolerance) a feature is compatible, others are removed from the set
for (map<Size, set<Size> >::iterator it = scan_idx_to_feature_idx.begin(); it != scan_idx_to_feature_idx.end(); ++it)
{
const Size scan = it->first;
const double pc_mz = exp[scan].getPrecursors()[0].getMZ();
const double mz_tolerance_da = ppm ? pc_mz * mz_tolerance * 1e-6 : mz_tolerance;
// Note: This is the "delete while iterating" pattern so mind the pre- and postincrement
for (set<Size>::iterator sit = it->second.begin(); sit != it->second.end(); )
{
if (!compatible_(features[*sit], pc_mz, mz_tolerance_da, max_trace))
{
it->second.erase(sit++);
}
else
{
++sit;
}
}
}
// remove entries with no compatible features (empty sets).
// Note: This is the "delete while iterating" pattern so mind the pre- and postincrement
for (map<Size, set<Size> >::iterator it = scan_idx_to_feature_idx.begin(); it != scan_idx_to_feature_idx.end(); )
{
if (it->second.empty())
{
scan_idx_to_feature_idx.erase(it++);
}
else
{
++it;
}
}
if (debug_level_ > 0)
{
LOG_INFO << "Number of precursors with compatible features: " << scan_idx_to_feature_idx.size() << endl;
}
if (!all_matching_features)
{
// keep only nearest features in set
for (map<Size, set<Size> >::iterator it = scan_idx_to_feature_idx.begin(); it != scan_idx_to_feature_idx.end(); ++it)
{
const Size scan = it->first;
const double pc_rt = exp[scan].getRT();
double min_distance = 1e16;
set<Size>::iterator best_feature = it->second.begin();
// determine nearest/best feature
for (set<Size>::iterator sit = it->second.begin(); sit != it->second.end(); ++sit)
{
const double current_distance = fabs(pc_rt - features[*sit].getRT());
if (current_distance < min_distance)
{
min_distance = current_distance;
best_feature = sit;
}
}
// delete all except the nearest/best feature
// Note: This is the "delete while iterating" pattern so mind the pre- and postincrement
for (set<Size>::iterator sit = it->second.begin(); sit != it->second.end(); )
{
if (sit != best_feature)
{
it->second.erase(sit++);
}
else
{
++sit;
}
}
}
}
// depending on all_matching_features option, only the nearest or all features are contained in the sets
// depending on options: move/copy corrected precursor and tandem spectrum
if (keep_original)
{
// duplicate spectra for each feature in set and adapt precursor_mz and precursor_charge to feature_mz and feature_charge
for (map<Size, set<Size> >::iterator it = scan_idx_to_feature_idx.begin(); it != scan_idx_to_feature_idx.end(); ++it)
{
const Size scan = it->first;
MSSpectrum<> spectrum = exp[scan];
corrected_precursors.insert(scan);
for (set<Size>::iterator f_it = it->second.begin(); f_it != it->second.end(); ++f_it)
{
spectrum.getPrecursors()[0].setMZ(features[*f_it].getMZ());
spectrum.getPrecursors()[0].setCharge(features[*f_it].getCharge());
exp.addSpectrum(spectrum);
}
}
}
else
{
// set precursor_mz and _charge to the feature_mz and _charge
for (map<Size, set<Size> >::iterator it = scan_idx_to_feature_idx.begin(); it != scan_idx_to_feature_idx.end(); ++it)
{
const Size scan = it->first;
exp[scan].getPrecursors()[0].setMZ(features[*it->second.begin()].getMZ());
exp[scan].getPrecursors()[0].setCharge(features[*it->second.begin()].getCharge());
corrected_precursors.insert(scan);
}
}
return corrected_precursors;
}
set<Size> correctToNearestMS1Peak(PeakMap & exp, double mz_tolerance, bool ppm, vector<double> & deltaMZs, vector<double> & mzs, vector<double> & rts)
{
set<Size> corrected_precursors;
// load experiment and extract precursors
vector<Precursor> precursors; // precursor
vector<double> precursors_rt; // RT of precursor MS2 spectrum
vector<Size> precursor_scan_index;
getPrecursors_(exp, precursors, precursors_rt, precursor_scan_index);
for (Size i = 0; i != precursors_rt.size(); ++i)
{
// get precursor rt
double rt = precursors_rt[i];
// get precursor MZ
double mz = precursors[i].getMZ();
//cout << rt << " " << mz << endl;
// get precursor spectrum
MSExperiment<Peak1D>::ConstIterator rt_it = exp.RTBegin(rt - 1e-8);
// store index of MS2 spectrum
UInt precursor_spectrum_idx = rt_it - exp.begin();
// get parent (MS1) of precursor spectrum
rt_it = exp.getPrecursorSpectrum(rt_it);
if (rt_it->getMSLevel() != 1)
{
LOG_WARN << "Error: no MS1 spectrum for this precursor" << endl;
}
//cout << rt_it->getRT() << " " << rt_it->size() << endl;
// find peak (index) closest to expected position
Size nearest_peak_idx = rt_it->findNearest(mz);
// get actual position of closest peak
double nearest_peak_mz = (*rt_it)[nearest_peak_idx].getMZ();
// calculate error between expected and actual position
double nearestPeakError = ppm ? abs(nearest_peak_mz - mz)/mz * 1e6 : abs(nearest_peak_mz - mz);
// check if error is small enough
if (nearestPeakError < mz_tolerance)
{
// sanity check: do we really have the same precursor in the original and the picked spectrum
if (fabs(exp[precursor_spectrum_idx].getPrecursors()[0].getMZ() - mz) > 0.0001)
{
LOG_WARN << "Error: index is referencing different precursors in original and picked spectrum." << endl;
}
// cout << mz << " -> " << nearest_peak_mz << endl;
double deltaMZ = nearest_peak_mz - mz;
deltaMZs.push_back(deltaMZ);
mzs.push_back(mz);
rts.push_back(rt);
// correct entries
Precursor corrected_prec = precursors[i];
corrected_prec.setMZ(nearest_peak_mz);
exp[precursor_spectrum_idx].getPrecursors()[0] = corrected_prec;
corrected_precursors.insert(precursor_spectrum_idx);
}
}
return corrected_precursors;
}
void filterByFoldChange(const PeakMap& exp1, const PeakMap& exp2,
const vector<double>& pc_ms2_rts, const vector<double>& pc_mzs,
const double rttol, const double mztol, double fold_change,
vector<double>& control_XIC_larger,
vector<double>& treatment_XIC_larger,
vector<double>& indifferent_XICs)
{
assert(pc_mzs.size() == pc_ms2_rts.size());
// search for each EIC and add up
for (Size i = 0; i < pc_mzs.size(); ++i)
{
//cerr << "start" << endl;
double pc_ms2_rt = pc_ms2_rts[i];
double pc_mz = pc_mzs[i];
//std::cerr << "Rt" << cm[i].getRT() << " mz: " << cm[i].getMZ() << " R " << cm[i].getMetaValue("rank") << "\n";
double mz_da = mztol * pc_mzs[i] / 1e6; // mz tolerance in Dalton
double rt_start = pc_ms2_rts[i] - rttol / 2.0;
// get area iterator (is MS1 only!) for rt and mz window
PeakMap::ConstAreaIterator it1 = exp1.areaBeginConst(pc_ms2_rt - rttol / 2, pc_ms2_rt + rttol / 2, pc_mz - mz_da, pc_mz + mz_da);
PeakMap::ConstAreaIterator it2 = exp2.areaBeginConst(pc_ms2_rt - rttol / 2, pc_ms2_rt + rttol / 2, pc_mz - mz_da, pc_mz + mz_da);
// determine maximum number of MS1 scans in retention time window
set<double> rts1;
set<double> rts2;
for (; it1 != exp1.areaEndConst(); ++it1)
{
rts1.insert(it1.getRT());
}
for (; it2 != exp2.areaEndConst(); ++it2)
{
rts2.insert(it2.getRT());
}
Size length = std::max(rts1.size(), rts2.size()) / 2.0;
//cout << length << endl;
if (length == 0)
{
cerr << "WARNING: no MS1 scans in retention time window found in both maps (mz: " << pc_mzs[i] << " / rt: " << pc_ms2_rts[i] << ")" << endl;
continue;
}
vector<double> XIC1(length, 0.0);
vector<double> XIC2(length, 0.0);
it1 = exp1.areaBeginConst(pc_ms2_rt - rttol / 2, pc_ms2_rt + rttol / 2, pc_mz - mz_da, pc_mz + mz_da);
it2 = exp2.areaBeginConst(pc_ms2_rt - rttol / 2, pc_ms2_rt + rttol / 2, pc_mz - mz_da, pc_mz + mz_da);
for (; it1 != exp1.areaEndConst(); ++it1)
{
double relative_rt = (it1.getRT() - rt_start) / rttol;
Size bin = relative_rt * (length - 1);
XIC1[bin] += it1->getIntensity();
if (bin >= length)
{
bin = length - 1;
}
}
for (; it2 != exp2.areaEndConst(); ++it2)
{
double relative_rt = (it2.getRT() - rt_start) / rttol;
Size bin = relative_rt * (length - 1);
if (bin >= length)
{
bin = length - 1;
}
XIC2[bin] += it2->getIntensity();
}
double total_itensity1 = std::accumulate(XIC1.begin(), XIC1.end(), 0.0);
double total_itensity2 = std::accumulate(XIC2.begin(), XIC2.end(), 0.0);
double ratio = total_itensity2 / (total_itensity1 + 1);
//cout << pc_ms2_rt << "/" << pc_mz << " has ratio: " << ratio << " determined on " << length << " bins" << endl;
if (ratio < 1.0 / fold_change)
{
control_XIC_larger.push_back(pc_ms2_rt);
}
else if (ratio > fold_change)
{
treatment_XIC_larger.push_back(pc_ms2_rt);
}
else
{
indifferent_XICs.push_back(pc_ms2_rt);
continue;
}
/*
for (Size k = 0; k != length; ++k)
{
cout << k << ": " << rt_start + rttol / length * k << ": " << XIC1[k] << " " << XIC2[k] << endl;
}
*/
}
cout << "control larger: " << control_XIC_larger.size() << " treatment larger: " << treatment_XIC_larger.size() << " indifferent: " << indifferent_XICs.size() << endl;
return;
}
ExitCodes main_(int, const char **) override
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
//input/output files
String in(getStringOption_("in"));
String out(getStringOption_("out"));
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
PeakMap exp;
MzMLFile f;
f.setLogType(log_type_);
PeakFileOptions options;
options.clearMSLevels();
options.addMSLevel(2);
f.getOptions() = options;
f.load(in, exp);
writeDebug_("Data set contains " + String(exp.size()) + " spectra", 1);
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
vector<PeptideIdentification> pep_ids;
CompNovoIdentificationCID comp_novo_id;
// set the options
Param algorithm_param = getParam_().copy("algorithm:", true);
comp_novo_id.setParameters(algorithm_param);
comp_novo_id.getIdentifications(pep_ids, exp);
algorithm_param = comp_novo_id.getParameters();
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
DateTime now = DateTime::now();
String date_string = now.get();
String identifier("CompNovoCID_" + date_string);
for (vector<PeptideIdentification>::iterator it = pep_ids.begin(); it != pep_ids.end(); ++it)
{
it->assignRanks();
it->setIdentifier(identifier);
}
vector<ProteinIdentification> prot_ids;
ProteinIdentification prot_id;
prot_id.setIdentifier(identifier);
prot_id.setDateTime(now);
StringList ms_runs;
exp.getPrimaryMSRunPath(ms_runs);
prot_id.setPrimaryMSRunPath(ms_runs);
ProteinIdentification::SearchParameters search_parameters;
search_parameters.charges = "+2-+3";
if (algorithm_param.getValue("tryptic_only").toBool())
{
search_parameters.digestion_enzyme = *(ProteaseDB::getInstance()->getEnzyme("Trypsin"));
}
else
{
search_parameters.digestion_enzyme = *(ProteaseDB::getInstance()->getEnzyme("no cleavage"));
}
search_parameters.mass_type = ProteinIdentification::MONOISOTOPIC;
search_parameters.fixed_modifications = algorithm_param.getValue("fixed_modifications");
search_parameters.variable_modifications = algorithm_param.getValue("variable_modifications");
search_parameters.missed_cleavages = (UInt)algorithm_param.getValue("missed_cleavages");
search_parameters.fragment_mass_tolerance = (double)algorithm_param.getValue("fragment_mass_tolerance");
search_parameters.precursor_mass_tolerance = (double)algorithm_param.getValue("precursor_mass_tolerance");
search_parameters.fragment_mass_tolerance_ppm = false;
search_parameters.precursor_mass_tolerance_ppm = false;
prot_id.setSearchParameters(search_parameters);
prot_id.setSearchEngineVersion("0.9beta");
prot_id.setSearchEngine("CompNovo");
prot_ids.push_back(prot_id);
IdXMLFile().store(out, prot_ids, pep_ids);
return EXECUTION_OK;
}
ExitCodes main_(int, const char**)
{
//-------------------------------------------------------------
// parsing parameters
//-------------------------------------------------------------
String in(getStringOption_("in"));
String out(getStringOption_("out"));
String pair_in(getStringOption_("pair_in"));
String feature_out(getStringOption_("feature_out"));
double precursor_mass_tolerance(getDoubleOption_("precursor_mass_tolerance"));
double RT_tolerance(getDoubleOption_("RT_tolerance"));
double expansion_range(getDoubleOption_("expansion_range"));
Size max_isotope(getIntOption_("max_isotope"));
Int debug(getIntOption_("debug"));
//-------------------------------------------------------------
// reading input
//-------------------------------------------------------------
PeakMap exp;
MzMLFile().load(in, exp);
exp.sortSpectra();
exp.updateRanges();
// read pair file
ifstream is(pair_in.c_str());
String line;
vector<SILAC_pair> pairs;
while (getline(is, line))
{
line.trim();
if (line.empty() || line[0] == '#')
{
continue;
}
vector<String> split;
line.split(' ', split);
if (split.size() != 4)
{
cerr << "missformated line ('" << line << "') should be (space separated) 'm/z-light m/z-heavy charge rt'" << endl;
}
SILAC_pair p;
p.mz_light = split[0].toDouble();
p.mz_heavy = split[1].toDouble();
p.charge = split[2].toInt();
p.rt = split[3].toDouble();
pairs.push_back(p);
}
is.close();
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
ConsensusMap results_map;
results_map.getFileDescriptions()[0].label = "light";
results_map.getFileDescriptions()[0].filename = in;
results_map.getFileDescriptions()[1].label = "heavy";
results_map.getFileDescriptions()[1].filename = in;
FeatureFinderAlgorithmIsotopeWavelet iso_ff;
Param ff_param(iso_ff.getParameters());
ff_param.setValue("max_charge", 3);
ff_param.setValue("intensity_threshold", -1.0);
iso_ff.setParameters(ff_param);
FeatureFinder ff;
ff.setLogType(ProgressLogger::NONE);
vector<SILACQuantitation> quantlets;
FeatureMap all_features;
for (PeakMap::ConstIterator it = exp.begin(); it != exp.end(); ++it)
{
if (it->size() == 0 || it->getMSLevel() != 1 || !it->getInstrumentSettings().getZoomScan())
{
continue;
}
PeakSpectrum new_spec = *it;
// get spacing from data
double min_spacing(numeric_limits<double>::max());
double last_mz(0);
for (PeakSpectrum::ConstIterator pit = new_spec.begin(); pit != new_spec.end(); ++pit)
{
if (pit->getMZ() - last_mz < min_spacing)
{
min_spacing = pit->getMZ() - last_mz;
}
last_mz = pit->getMZ();
}
writeDebug_("Min-spacing=" + String(min_spacing), 1);
// split the spectrum into two subspectra, by using different hypothesis of
// the SILAC pairs
Size idx = 0;
for (vector<SILAC_pair>::const_iterator pit = pairs.begin(); pit != pairs.end(); ++pit, ++idx)
{
// in RT window?
if (fabs(it->getRT() - pit->rt) >= RT_tolerance)
{
continue;
}
// now excise the two ranges for the pair, complete isotope distributions of both, light and heavy
PeakSpectrum light_spec, heavy_spec;
light_spec.setRT(it->getRT());
heavy_spec.setRT(it->getRT());
for (PeakSpectrum::ConstIterator sit = it->begin(); sit != it->end(); ++sit)
{
double mz(sit->getMZ());
if (mz - (pit->mz_light - precursor_mass_tolerance) > 0 &&
(pit->mz_light + (double)max_isotope * Constants::NEUTRON_MASS_U / (double)pit->charge + precursor_mass_tolerance) - mz > 0)
{
light_spec.push_back(*sit);
}
if (mz - (pit->mz_heavy - precursor_mass_tolerance) > 0 &&
(pit->mz_heavy + (double)max_isotope * Constants::NEUTRON_MASS_U / (double)pit->charge + precursor_mass_tolerance) - mz > 0)
{
heavy_spec.push_back(*sit);
}
}
// expand light spectrum
Peak1D p;
p.setIntensity(0);
if (light_spec.size() > 0)
{
double lower_border = light_spec.begin()->getMZ() - expansion_range;
for (double pos = light_spec.begin()->getMZ(); pos > lower_border; pos -= min_spacing)
{
p.setMZ(pos);
light_spec.insert(light_spec.begin(), p);
}
double upper_border = light_spec.begin()->getMZ() - expansion_range;
for (double pos = light_spec.rbegin()->getMZ(); pos < upper_border; pos += min_spacing)
{
p.setMZ(pos);
light_spec.push_back(p);
}
}
if (heavy_spec.size() > 0)
{
// expand heavy spectrum
double lower_border = heavy_spec.begin()->getMZ() - expansion_range;
for (double pos = heavy_spec.begin()->getMZ(); pos > lower_border; pos -= min_spacing)
{
p.setMZ(pos);
heavy_spec.insert(heavy_spec.begin(), p);
}
double upper_border = heavy_spec.begin()->getMZ() - expansion_range;
for (double pos = heavy_spec.rbegin()->getMZ(); pos < upper_border; pos += min_spacing)
{
p.setMZ(pos);
heavy_spec.push_back(p);
}
}
// create experiments for feature finding
PeakMap new_exp_light, new_exp_heavy;
new_exp_light.addSpectrum(light_spec);
new_exp_heavy.addSpectrum(heavy_spec);
if (debug > 9)
{
MzMLFile().store(String(it->getRT()) + "_debugging_light.mzML", new_exp_light);
MzMLFile().store(String(it->getRT()) + "_debugging_heavy.mzML", new_exp_heavy);
}
writeDebug_("Spectrum-id: " + it->getNativeID() + " @ " + String(it->getRT()) + "s", 1);
new_exp_light.updateRanges();
new_exp_heavy.updateRanges();
FeatureMap feature_map_light, feature_map_heavy, seeds;
if (light_spec.size() > 0)
{
ff.run("isotope_wavelet", new_exp_light, feature_map_light, ff_param, seeds);
}
writeDebug_("#light_features=" + String(feature_map_light.size()), 1);
if (heavy_spec.size() > 0)
{
ff.run("isotope_wavelet", new_exp_heavy, feature_map_heavy, ff_param, seeds);
}
writeDebug_("#heavy_features=" + String(feature_map_heavy.size()), 1);
// search if feature maps to m/z value of pair
vector<MatchedFeature> light, heavy;
for (FeatureMap::const_iterator fit = feature_map_light.begin(); fit != feature_map_light.end(); ++fit)
{
all_features.push_back(*fit);
light.push_back(MatchedFeature(*fit, idx));
}
for (FeatureMap::const_iterator fit = feature_map_heavy.begin(); fit != feature_map_heavy.end(); ++fit)
{
all_features.push_back(*fit);
heavy.push_back(MatchedFeature(*fit, idx));
}
if (!heavy.empty() && !light.empty())
{
writeDebug_("Finding best feature pair out of " + String(light.size()) + " light and " + String(heavy.size()) + " heavy matching features.", 1);
// now find "good" matches, means the pair with the smallest m/z deviation
Feature best_light, best_heavy;
double best_deviation(numeric_limits<double>::max());
Size best_idx(pairs.size());
for (vector<MatchedFeature>::const_iterator fit1 = light.begin(); fit1 != light.end(); ++fit1)
{
for (vector<MatchedFeature>::const_iterator fit2 = heavy.begin(); fit2 != heavy.end(); ++fit2)
{
if (fit1->idx != fit2->idx || fit1->f.getCharge() != fit2->f.getCharge() ||
fabs(fit1->f.getMZ() - pairs[fit1->idx].mz_light) > precursor_mass_tolerance ||
fabs(fit2->f.getMZ() - pairs[fit2->idx].mz_heavy) > precursor_mass_tolerance)
{
continue;
}
double deviation(0);
deviation = fabs((fit1->f.getMZ() - pairs[fit1->idx].mz_light) - (fit2->f.getMZ() - pairs[fit2->idx].mz_heavy));
if (deviation < best_deviation && deviation < precursor_mass_tolerance)
{
best_light = fit1->f;
best_heavy = fit2->f;
best_idx = fit1->idx;
}
}
}
if (best_idx == pairs.size())
{
continue;
}
writeDebug_("Ratio: " + String(best_heavy.getIntensity() / best_light.getIntensity()), 1);
ConsensusFeature SILAC_feature;
SILAC_feature.setMZ((best_light.getMZ() + best_heavy.getMZ()) / 2.0);
SILAC_feature.setRT((best_light.getRT() + best_heavy.getRT()) / 2.0);
SILAC_feature.insert(0, best_light);
SILAC_feature.insert(1, best_heavy);
results_map.push_back(SILAC_feature);
quantlets.push_back(SILACQuantitation(best_light.getIntensity(), best_heavy.getIntensity(), best_idx));
}
}
}
// now calculate the final quantitation values from the quantlets
Map<Size, vector<SILACQuantitation> > idx_to_quantlet;
for (vector<SILACQuantitation>::const_iterator it = quantlets.begin(); it != quantlets.end(); ++it)
{
idx_to_quantlet[it->idx].push_back(*it);
}
for (Map<Size, vector<SILACQuantitation> >::ConstIterator it1 = idx_to_quantlet.begin(); it1 != idx_to_quantlet.end(); ++it1)
{
SILAC_pair silac_pair = pairs[it1->first];
// simply add up all intensities and calculate the final ratio
double light_sum(0), heavy_sum(0);
vector<double> light_ints, heavy_ints, ratios;
for (vector<SILACQuantitation>::const_iterator it2 = it1->second.begin(); it2 != it1->second.end(); ++it2)
{
light_sum += it2->light_intensity;
light_ints.push_back(it2->light_intensity);
heavy_sum += it2->heavy_intensity;
heavy_ints.push_back(it2->heavy_intensity);
ratios.push_back(it2->heavy_intensity / it2->light_intensity * (it2->heavy_intensity + it2->light_intensity));
}
double absdev_ratios = Math::absdev(ratios.begin(), ratios.begin() + (ratios.size()) / (heavy_sum + light_sum));
cout << "Ratio: " << silac_pair.mz_light << " <-> " << silac_pair.mz_heavy << " @ " << silac_pair.rt << " s, ratio(h/l) " << heavy_sum / light_sum << " +/- " << absdev_ratios << " (#scans for quantation: " << String(it1->second.size()) << " )" << endl;
}
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
if (feature_out != "")
{
FeatureXMLFile().store(feature_out, all_features);
}
writeDebug_("Writing output", 1);
ConsensusXMLFile().store(out, results_map);
return EXECUTION_OK;
}
e_ptr->filterSpectrum(spec);
TEST_EQUAL(spec.size(), 121)
p.setValue("threshold", 10.0);
e_ptr->setParameters(p);
e_ptr->filterSpectrum(spec);
TEST_EQUAL(spec.size(), 14)
END_SECTION
START_SECTION((void filterPeakMap(PeakMap& exp)))
DTAFile dta_file;
PeakSpectrum spec;
dta_file.load(OPENMS_GET_TEST_DATA_PATH("Transformers_tests.dta"), spec);
PeakMap pm;
pm.addSpectrum(spec);
TEST_EQUAL(pm.begin()->size(), 121)
Param p(e_ptr->getParameters());
p.setValue("threshold", 1.0);
e_ptr->setParameters(p);
e_ptr->filterPeakMap(pm);
TEST_EQUAL(pm.begin()->size(), 121)
p.setValue("threshold", 10.0);
e_ptr->setParameters(p);
e_ptr->filterPeakMap(pm);
TEST_EQUAL(pm.begin()->size(), 14)
ExitCodes main_(int, const char **)
{
//input and output file names ..
String in = getStringOption_("in");
String out = getStringOption_("out");
//prevent loading of fragment spectra
PeakFileOptions options;
options.setMSLevels(vector<Int>(1, 1));
//reading input data
MzMLFile f;
f.getOptions() = options;
f.setLogType(log_type_);
PeakMap exp;
f.load(in, exp);
exp.updateRanges();
//no seeds supported
FeatureMap<> seeds;
//setup of FeatureFinder
FeatureFinder ff;
ff.setLogType(log_type_);
// A map for the resulting features
FeatureMap<> features;
// get parameters specific for the feature finder
Param feafi_param = getParam_().copy("algorithm:", true);
writeDebug_("Parameters passed to FeatureFinder", feafi_param, 3);
// Apply the feature finder
ff.run(FeatureFinderAlgorithmIsotopeWavelet<Peak1D, Feature>::getProductName(), exp, features, feafi_param, seeds);
features.applyMemberFunction(&UniqueIdInterface::setUniqueId);
// DEBUG
if (debug_level_ > 10)
{
FeatureMap<>::Iterator it;
for (it = features.begin(); it != features.end(); ++it)
{
if (!it->isMetaEmpty())
{
vector<String> keys;
it->getKeys(keys);
LOG_INFO << "Feature " << it->getUniqueId() << endl;
for (Size i = 0; i < keys.size(); i++)
{
LOG_INFO << " " << keys[i] << " = " << it->getMetaValue(keys[i]) << endl;
}
}
}
}
//-------------------------------------------------------------
// writing files
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(features, getProcessingInfo_(DataProcessing::QUANTITATION));
// write features to user specified output file
FeatureXMLFile map_file;
map_file.store(out, features);
return EXECUTION_OK;
}
ptr = new MascotGenericFile();
TEST_NOT_EQUAL(ptr, nullPointer)
}
END_SECTION
START_SECTION(virtual ~MascotGenericFile())
{
delete ptr;
}
END_SECTION
ptr = new MascotGenericFile();
START_SECTION((template < typename MapType > void load(const String &filename, MapType &exp)))
{
PeakMap exp;
ptr->load(OPENMS_GET_TEST_DATA_PATH("MascotInfile_test.mascot_in"), exp);
TEST_EQUAL(exp.size(), 1)
TEST_EQUAL(exp.begin()->size(), 9)
}
END_SECTION
START_SECTION((void store(std::ostream &os, const String &filename, const PeakMap &experiment, bool compact = false)))
{
PeakMap exp;
ptr->load(OPENMS_GET_TEST_DATA_PATH("MascotInfile_test.mascot_in"), exp);
// handling of modifications:
Param params = ptr->getParameters();
params.setValue("fixed_modifications", ListUtils::create<String>("Carbamidomethyl (C),Phospho (S)"));
ExitCodes main_(int, const char **)
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
in = getStringOption_("in");
out = getStringOption_("out");
String process_option = getStringOption_("processOption");
Param filter_param = getParam_().copy("algorithm:", true);
writeDebug_("Parameters passed to filter", filter_param, 3);
SavitzkyGolayFilter sgolay;
sgolay.setLogType(log_type_);
sgolay.setParameters(filter_param);
if (process_option == "lowmemory")
{
return doLowMemAlgorithm(sgolay);
}
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
MzMLFile mz_data_file;
mz_data_file.setLogType(log_type_);
PeakMap exp;
mz_data_file.load(in, exp);
if (exp.empty() && exp.getChromatograms().size() == 0)
{
LOG_WARN << "The given file does not contain any conventional peak data, but might"
" contain chromatograms. This tool currently cannot handle them, sorry.";
return INCOMPATIBLE_INPUT_DATA;
}
//check for peak type (profile data required)
if (!exp.empty() && PeakTypeEstimator().estimateType(exp[0].begin(), exp[0].end()) == SpectrumSettings::PEAKS)
{
writeLog_("Warning: OpenMS peak type estimation indicates that this is not profile data!");
}
//check if spectra are sorted
for (Size i = 0; i < exp.size(); ++i)
{
if (!exp[i].isSorted())
{
writeLog_("Error: Not all spectra are sorted according to peak m/z positions. Use FileFilter to sort the input!");
return INCOMPATIBLE_INPUT_DATA;
}
}
//check if chromatograms are sorted
for (Size i = 0; i < exp.getChromatograms().size(); ++i)
{
if (!exp.getChromatogram(i).isSorted())
{
writeLog_("Error: Not all chromatograms are sorted according to peak m/z positions. Use FileFilter to sort the input!");
return INCOMPATIBLE_INPUT_DATA;
}
}
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
sgolay.filterExperiment(exp);
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(exp, getProcessingInfo_(DataProcessing::SMOOTHING));
mz_data_file.store(out, exp);
return EXECUTION_OK;
}
DTAFile dta_file;
PeakSpectrum spec;
dta_file.load(OPENMS_GET_TEST_DATA_PATH("Transformers_tests.dta"), spec);
TEST_REAL_SIMILAR((spec.begin() + 40)->getIntensity(), 37.5)
e_ptr->filterSpectrum(spec);
TEST_REAL_SIMILAR((spec.begin() + 40)->getIntensity(), sqrt(37.5))
END_SECTION
START_SECTION((void filterPeakMap(PeakMap& exp)))
DTAFile dta_file;
PeakSpectrum spec;
dta_file.load(OPENMS_GET_TEST_DATA_PATH("Transformers_tests.dta"), spec);
PeakMap pm;
pm.addSpectrum(spec);
TEST_REAL_SIMILAR((pm.begin()->begin() + 40)->getIntensity(), 37.5)
e_ptr->filterPeakMap(pm);
TEST_REAL_SIMILAR((pm.begin()->begin() + 40)->getIntensity(), sqrt(37.5))
END_SECTION
START_SECTION((void filterPeakSpectrum(PeakSpectrum& spectrum)))
DTAFile dta_file;
PeakSpectrum spec;
dta_file.load(OPENMS_GET_TEST_DATA_PATH("Transformers_tests.dta"), spec);
TEST_REAL_SIMILAR((spec.begin() + 40)->getIntensity(), 37.5)
START_SECTION(~CachedmzML())
{
delete ptr;
}
END_SECTION
// see also MSDataCachedConsumer_test.cpp -> consumeSpectrum
// this is a complete test of the caching object
START_SECTION(( [EXTRA] testCaching))
{
std::string tmp_filename;
NEW_TMP_FILE(tmp_filename);
// Load experiment
PeakMap exp;
MzMLFile().load(OPENMS_GET_TEST_DATA_PATH("MzMLFile_1.mzML"), exp);
TEST_EQUAL(exp.getNrSpectra() > 0, true)
TEST_EQUAL(exp.getNrChromatograms() > 0, true)
// Cache the experiment to a temporary file
CachedmzML cache;
cache.writeMemdump(exp, tmp_filename);
// Check whether spectra were written to disk correctly...
{
// Create the index from the given file
cache.createMemdumpIndex(tmp_filename);
std::vector<std::streampos> spectra_index = cache.getSpectraIndex();
TEST_EQUAL(spectra_index.size(), 4)
std::ifstream ifs_(tmp_filename.c_str(), std::ios::binary);