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;
}
// 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;
}
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 FeatureFinder::run(const String& algorithm_name, PeakMap& input_map, FeatureMap& features, const Param& param, const FeatureMap& seeds)
{
// Nothing to do if there is no data
if ((algorithm_name != "mrm" && input_map.empty()) || (algorithm_name == "mrm" && input_map.getChromatograms().empty()))
{
features.clear(true);
return;
}
// check input
{
// We need updated ranges => check number of peaks
if (algorithm_name != "mrm" && input_map.getSize() == 0)
{
throw Exception::IllegalArgument(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "FeatureFinder needs updated ranges on input map. Aborting.");
}
// We need MS1 data only => check levels
if (algorithm_name != "mrm" && (input_map.getMSLevels().size() != 1 || input_map.getMSLevels()[0] != 1))
{
throw Exception::IllegalArgument(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "FeatureFinder can only operate on MS level 1 data. Please do not use MS/MS data. Aborting.");
}
//Check if the peaks are sorted according to m/z
if (!input_map.isSorted(true))
{
LOG_WARN << "Input map is not sorted by RT and m/z! This is done now, before applying the algorithm!" << std::endl;
input_map.sortSpectra(true);
input_map.sortChromatograms(true);
}
for (Size s = 0; s < input_map.size(); ++s)
{
if (input_map[s].empty())
continue;
if (input_map[s][0].getMZ() < 0)
{
throw Exception::IllegalArgument(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "FeatureFinder can only operate on spectra that contain peaks with positive m/z values. Filter the data accordingly beforehand! Aborting.");
}
}
}
// initialize
if (algorithm_name != "mrm" && algorithm_name != "centroided")
{
// Resize peak flag vector
flags_.resize(input_map.size());
for (Size i = 0; i < input_map.size(); ++i)
{
flags_[i].assign(input_map[i].size(), UNUSED);
}
}
// do the work
if (algorithm_name != "none")
{
FeatureFinderAlgorithm* algorithm = Factory<FeatureFinderAlgorithm>::create(algorithm_name);
algorithm->setParameters(param);
algorithm->setData(input_map, features, *this);
algorithm->setSeeds(seeds);
algorithm->run();
delete(algorithm);
}
if (algorithm_name != "mrm") // mrm works on chromatograms; the next section is only for conventional data
{
//report RT apex spectrum index and native ID for each feature
for (Size i = 0; i < features.size(); ++i)
{
//index
Size spectrum_index = input_map.RTBegin(features[i].getRT()) - input_map.begin();
features[i].setMetaValue("spectrum_index", spectrum_index);
//native id
if (spectrum_index < input_map.size())
{
String native_id = input_map[spectrum_index].getNativeID();
features[i].setMetaValue("spectrum_native_id", native_id);
}
else
{
/// @todo that happens sometimes using IsotopeWaveletFeatureFinder (Rene, Marc, Andreas, Clemens)
std::cerr << "FeatureFinderAlgorithm_impl, line=" << __LINE__ << "; FixMe this cannot be, but happens" << std::endl;
}
}
}
}
// version for labeled linkers
void XQuestResultXMLFile::writeXQuestXMLSpec(String out_file, String base_name, const OPXLDataStructs::PreprocessedPairSpectra& preprocessed_pair_spectra, const std::vector< std::pair<Size, Size> >& spectrum_pairs, 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 <std::pair <Size, 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() && all_top_csms[i][0].scan_index_heavy < spectra.size())
{
spectrum_indices.push_back( std::make_pair(all_top_csms[i][0].scan_index_light, all_top_csms[i][0].scan_index_heavy) );
}
}
}
// loop over list of indices and write out spectra
for (Size i = 0; i < spectrum_indices.size(); ++i)
{
Size scan_index_light = spectrum_indices[i].first;
Size scan_index_heavy = spectrum_indices[i].second;
// TODO more correct alternative
String spectrum_light_name = base_name + ".light." + scan_index_light;
String spectrum_heavy_name = base_name + ".heavy." + scan_index_heavy;
String spectrum_name = spectrum_light_name + String("_") + spectrum_heavy_name;
if (scan_index_light < spectra.size() && scan_index_heavy < spectra.size() && i < preprocessed_pair_spectra.spectra_common_peaks.size() && i < preprocessed_pair_spectra.spectra_xlink_peaks.size())
{
// 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[scan_index_light], 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[scan_index_heavy], String(""));
spec_xml_file << "</spectrum>" << std::endl;
// the preprocessed pair spectra are sorted by another index
// because some pairs do not yield any resonable hits, the index from the spectrum matches or spectrum_indices does not address the right pair anymore
// use find with the pair of spectrum indices to find the correct index for the preprocessed common and cross-linked ion spectra
std::vector<std::pair <Size, Size> >::const_iterator pair_it = std::find(spectrum_pairs.begin(), spectrum_pairs.end(), spectrum_indices[i]);
Size pair_index = std::distance(spectrum_pairs.begin(), pair_it);
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_(preprocessed_pair_spectra.spectra_common_peaks[pair_index], 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_(preprocessed_pair_spectra.spectra_xlink_peaks[pair_index], 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;
}
ExitCodes main_(int, const char **)
{
//-------------------------------------------------------------
// parsing parameters
//-------------------------------------------------------------
String in(getStringOption_("in"));
String out(getStringOption_("out"));
Size num_spots_per_row(getIntOption_("num_spots_per_row"));
double RT_distance(getDoubleOption_("RT_distance"));
//-------------------------------------------------------------
// reading input
//-------------------------------------------------------------
PeakMap exp;
MzMLFile f;
f.setLogType(log_type_);
f.load(in, exp);
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
ProgressLogger pl;
pl.setLogType(log_type_);
pl.startProgress(0, exp.size(), "Assigning pseudo RTs.");
Size num_ms1(0), num_ms1_base(0), row_counter(0);
bool row_to_reverse(false);
double actual_RT(0);
for (Size i = 0; i != exp.size(); ++i)
{
pl.setProgress(i);
if (row_to_reverse)
{
actual_RT = (double)(num_ms1_base + (num_spots_per_row - row_counter)) * RT_distance;
writeDebug_("RT=" + String(actual_RT) + " (modified, row_counter=" + String(row_counter) + ")", 1);
}
else
{
actual_RT = (double)num_ms1 * RT_distance;
writeDebug_("RT=" + String(actual_RT), 1);
}
exp[i].setRT(actual_RT);
if (exp[i].getMSLevel() == 1)
{
if (++row_counter >= num_spots_per_row)
{
row_counter = 0;
if (row_to_reverse)
{
row_to_reverse = false;
}
else
{
row_to_reverse = true;
}
}
++num_ms1;
if (!row_to_reverse)
{
num_ms1_base = num_ms1;
}
}
}
pl.endProgress();
// sort the spectra according to their new RT
exp.sortSpectra();
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
f.store(out, exp);
return EXECUTION_OK;
}
/**
* @brief Applies the peak-picking algorithm to a map (MSExperiment). This
* method picks peaks for each scan in the map consecutively. The resulting
* picked peaks are written to the output map.
*
* @param input input map in profile mode
* @param output output map with picked peaks
* @param boundaries_spec boundaries of the picked peaks in spectra
* @param boundaries_chrom boundaries of the picked peaks in chromatograms
* @param check_spectrum_type if set, checks spectrum type and throws an exception if a centroided spectrum is passed
*/
void PeakPickerHiRes::pickExperiment(const PeakMap& input, PeakMap& output,
std::vector<std::vector<PeakBoundary> >& boundaries_spec,
std::vector<std::vector<PeakBoundary> >& boundaries_chrom,
const bool check_spectrum_type) const
{
// make sure that output is clear
output.clear(true);
// copy experimental settings
static_cast<ExperimentalSettings &>(output) = input;
// resize output with respect to input
output.resize(input.size());
Size progress = 0;
startProgress(0, input.size() + input.getChromatograms().size(), "picking peaks");
if (input.getNrSpectra() > 0)
{
for (Size scan_idx = 0; scan_idx != input.size(); ++scan_idx)
{
if (ms_levels_.empty()) // auto mode
{
SpectrumSettings::SpectrumType spectrum_type = input[scan_idx].getType();
if (spectrum_type == SpectrumSettings::CENTROID)
{
output[scan_idx] = input[scan_idx];
}
else
{
std::vector<PeakBoundary> boundaries_s; // peak boundaries of a single spectrum
pick(input[scan_idx], output[scan_idx], boundaries_s);
boundaries_spec.push_back(boundaries_s);
}
}
else if (!ListUtils::contains(ms_levels_, input[scan_idx].getMSLevel())) // manual mode
{
output[scan_idx] = input[scan_idx];
}
else
{
std::vector<PeakBoundary> boundaries_s; // peak boundaries of a single spectrum
// determine type of spectral data (profile or centroided)
SpectrumSettings::SpectrumType spectrum_type = input[scan_idx].getType();
if (spectrum_type == SpectrumSettings::CENTROID && check_spectrum_type)
{
throw OpenMS::Exception::IllegalArgument(__FILE__, __LINE__, __FUNCTION__, "Error: Centroided data provided but profile spectra expected.");
}
pick(input[scan_idx], output[scan_idx], boundaries_s);
boundaries_spec.push_back(boundaries_s);
}
setProgress(++progress);
}
}
for (Size i = 0; i < input.getChromatograms().size(); ++i)
{
MSChromatogram chromatogram;
std::vector<PeakBoundary> boundaries_c; // peak boundaries of a single chromatogram
pick(input.getChromatograms()[i], chromatogram, boundaries_c);
output.addChromatogram(chromatogram);
boundaries_chrom.push_back(boundaries_c);
setProgress(++progress);
}
endProgress();
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;
}
// 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;
}
ExitCodes main_(int argc, const char** argv)
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
//input/output files
String in(getStringOption_("in")), out(getStringOption_("out"));
FileHandler fh;
FileTypes::Type in_type = fh.getType(in);
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
PeakMap exp;
// keep only MS2 spectra
fh.getOptions().addMSLevel(2);
fh.loadExperiment(in, exp, in_type, log_type_);
writeDebug_(String("Spectra loaded: ") + exp.size(), 2);
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.");
}
}
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
Param mascot_param = getParam_().copy("Mascot_parameters:", true);
MascotGenericFile mgf_file;
Param p;
// TODO: switch this to mzML (much smaller)
p.setValue("internal:format", "Mascot generic", "Sets the format type of the peak list, this should not be changed unless you write the header only.", ListUtils::create<String>("advanced"));
p.setValue("internal:HTTP_format", "true", "Write header with MIME boundaries instead of simple key-value pairs. For HTTP submission only.", ListUtils::create<String>("advanced"));
p.setValue("internal:content", "all", "Use parameter header + the peak lists with BEGIN IONS... or only one of them.", ListUtils::create<String>("advanced"));
mgf_file.setParameters(mascot_param);
// get the spectra into string stream
writeDebug_("Writing MGF file to stream", 1);
stringstream ss;
mgf_file.store(ss, in, exp, true); // write in compact format
// Usage of a QCoreApplication is overkill here (and ugly too), but we just use the
// QEventLoop to process the signals and slots and grab the results afterwards from
// the MascotRemotQuery instance
char** argv2 = const_cast<char**>(argv);
QCoreApplication event_loop(argc, argv2);
MascotRemoteQuery* mascot_query = new MascotRemoteQuery(&event_loop);
Param mascot_query_param = getParam_().copy("Mascot_server:", true);
writeDebug_("Setting parameters for Mascot query", 1);
mascot_query->setParameters(mascot_query_param);
writeDebug_("Setting spectra for Mascot query", 1);
mascot_query->setQuerySpectra(ss.str());
// remove unnecessary spectra
ss.clear();
QObject::connect(mascot_query, SIGNAL(done()), &event_loop, SLOT(quit()));
QTimer::singleShot(1000, mascot_query, SLOT(run()));
writeDebug_("Fire off Mascot query", 1);
event_loop.exec();
writeDebug_("Mascot query finished", 1);
if (mascot_query->hasError())
{
writeLog_("An error occurred during the query: " + mascot_query->getErrorMessage());
delete mascot_query;
return EXTERNAL_PROGRAM_ERROR;
}
// write Mascot response to file
String mascot_tmp_file_name(File::getTempDirectory() + "/" + File::getUniqueName() + "_Mascot_response");
QFile mascot_tmp_file(mascot_tmp_file_name.c_str());
mascot_tmp_file.open(QIODevice::WriteOnly);
mascot_tmp_file.write(mascot_query->getMascotXMLResponse());
mascot_tmp_file.close();
// clean up
delete mascot_query;
vector<PeptideIdentification> pep_ids;
ProteinIdentification prot_id;
// set up mapping between scan numbers and retention times:
MascotXMLFile::RTMapping rt_mapping;
MascotXMLFile::generateRTMapping(exp.begin(), exp.end(), rt_mapping);
// read the response
MascotXMLFile().load(mascot_tmp_file_name, prot_id, pep_ids, rt_mapping);
writeDebug_("Read " + String(pep_ids.size()) + " peptide ids and " + String(prot_id.getHits().size()) + " protein identifications from Mascot", 5);
// for debugging errors relating to unexpected response files
if (this->debug_level_ >= 100)
{
writeDebug_(String("\nMascot Server Response file saved to: '") + mascot_tmp_file_name + "'. If an error occurs, send this file to the OpenMS team.\n", 100);
}
else
{
// delete file
mascot_tmp_file.remove();
}
// keep or delete protein identifications?!
vector<ProteinIdentification> prot_ids;
if (!getFlag_("keep_protein_links"))
{
// remove protein links from peptides
for (Size i = 0; i < pep_ids.size(); ++i)
{
std::vector<PeptideHit> hits = pep_ids[i].getHits();
for (Size h = 0; h < hits.size(); ++h)
{
hits[h].setPeptideEvidences(vector<PeptideEvidence>());
}
pep_ids[i].setHits(hits);
}
// remove proteins
std::vector<ProteinHit> p_hit;
prot_id.setHits(p_hit);
}
prot_ids.push_back(prot_id);
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
IdXMLFile().store(out, prot_ids, pep_ids);
return EXECUTION_OK;
}
ExitCodes
main_(int, const char**)
{
//-------------------------------------------------------------
// general variables and data
//-------------------------------------------------------------
FileHandler fh;
vector<PeptideIdentification> peptide_identifications;
vector<ProteinIdentification> protein_identifications;
//-------------------------------------------------------------
// reading input
//-------------------------------------------------------------
const String in = getStringOption_("in");
ProgressLogger logger;
logger.setLogType(ProgressLogger::CMD);
logger.startProgress(0, 1, "Loading...");
if (File::isDirectory(in))
{
const String in_directory = File::absolutePath(in).ensureLastChar('/');
const String mz_file = getStringOption_("mz_file");
const bool ignore_proteins_per_peptide = getFlag_("ignore_proteins_per_peptide");
UInt i = 0;
FileHandler fh;
FileTypes::Type type;
MSExperiment<Peak1D> msexperiment;
// Note: we had issues with leading zeroes, so let us represent scan numbers as Int (next line used to be map<String, float> num_and_rt;) However, now String::toInt() might throw.
map<Int, float> num_and_rt;
vector<String> NativeID;
// The mz-File (if given)
if (!mz_file.empty())
{
type = fh.getTypeByFileName(mz_file);
fh.loadExperiment(mz_file, msexperiment, type);
for (MSExperiment<Peak1D>::Iterator spectra_it = msexperiment.begin(); spectra_it != msexperiment.end(); ++spectra_it)
{
String(spectra_it->getNativeID()).split('=', NativeID);
try
{
num_and_rt[NativeID[1].toInt()] = spectra_it->getRT();
// cout << "num_and_rt: " << NativeID[1] << " = " << NativeID[1].toInt() << " : " << num_and_rt[NativeID[1].toInt()] << endl; // CG debuggging 2009-07-01
}
catch (Exception::ConversionError& e)
{
writeLog_(String("Error: Cannot read scan number as integer. '") + e.getMessage());
}
}
}
// Get list of the actual Sequest .out-Files
StringList in_files;
if (!File::fileList(in_directory, String("*.out"), in_files))
{
writeLog_(String("Error: No .out files found in '") + in_directory + "'. Aborting!");
}
// Now get to work ...
for (vector<String>::const_iterator in_files_it = in_files.begin(); in_files_it != in_files.end(); ++in_files_it)
{
vector<PeptideIdentification> peptide_ids_seq;
ProteinIdentification protein_id_seq;
vector<double> pvalues_seq;
vector<String> in_file_vec;
SequestOutfile sequest_outfile;
writeDebug_(String("Reading file ") + *in_files_it, 3);
try
{
sequest_outfile.load((String) (in_directory + *in_files_it), peptide_ids_seq, protein_id_seq, 1.0, pvalues_seq, "Sequest", ignore_proteins_per_peptide);
in_files_it->split('.', in_file_vec);
for (Size j = 0; j < peptide_ids_seq.size(); ++j)
{
// We have to explicitly set the identifiers, because the normal set ones are composed of search engine name and date, which is the same for a bunch of sequest out-files.
peptide_ids_seq[j].setIdentifier(*in_files_it + "_" + i);
Int scan_number = 0;
if (!mz_file.empty())
{
try
{
scan_number = in_file_vec[2].toInt();
peptide_ids_seq[j].setRT(num_and_rt[scan_number]);
}
catch (Exception::ConversionError& e)
{
writeLog_(String("Error: Cannot read scan number as integer. '") + e.getMessage());
}
catch (exception& e)
{
writeLog_(String("Error: Cannot read scan number as integer. '") + e.what());
}
//double real_mz = ( peptide_ids_seq[j].getMZ() - hydrogen_mass )/ (double)peptide_ids_seq[j].getHits()[0].getCharge(); // ???? semantics of mz
const double real_mz = peptide_ids_seq[j].getMZ() / (double) peptide_ids_seq[j].getHits()[0].getCharge();
peptide_ids_seq[j].setMZ(real_mz);
}
writeDebug_(String("scan: ") + String(scan_number) + String(" RT: ") + String(peptide_ids_seq[j].getRT()) + " MZ: " + String(peptide_ids_seq[j].getMZ()) + " Ident: " + peptide_ids_seq[j].getIdentifier(), 4);
peptide_identifications.push_back(peptide_ids_seq[j]);
}
protein_id_seq.setIdentifier(*in_files_it + "_" + i);
protein_identifications.push_back(protein_id_seq);
++i;
}
catch (Exception::ParseError& pe)
{
writeLog_(pe.getMessage() + String("(file: ") + *in_files_it + ")");
throw;
}
catch (...)
{
writeLog_(String("Error reading file: ") + *in_files_it);
throw;
}
}
writeDebug_("All files processed.", 3);
} // ! directory
else
{
FileTypes::Type in_type = fh.getType(in);
if (in_type == FileTypes::PEPXML)
{
String exp_name = getStringOption_("mz_file");
String orig_name = getStringOption_("mz_name");
bool use_precursor_data = getFlag_("use_precursor_data");
if (exp_name.empty())
{
PepXMLFile().load(in, protein_identifications,
peptide_identifications, orig_name);
}
else
{
MSExperiment<> exp;
fh.loadExperiment(exp_name, exp);
if (!orig_name.empty())
{
exp_name = orig_name;
}
PepXMLFile().load(in, protein_identifications,
peptide_identifications, exp_name, exp,
use_precursor_data);
}
}
else if (in_type == FileTypes::IDXML)
{
IdXMLFile().load(in, protein_identifications, peptide_identifications);
}
else if (in_type == FileTypes::MZIDENTML)
{
LOG_WARN << "Converting from mzid: you might experience loss of information depending on the capabilities of the target format." << endl;
MzIdentMLFile().load(in, protein_identifications, peptide_identifications);
}
else if (in_type == FileTypes::PROTXML)
{
protein_identifications.resize(1);
peptide_identifications.resize(1);
ProtXMLFile().load(in, protein_identifications[0],
peptide_identifications[0]);
}
else if (in_type == FileTypes::OMSSAXML)
{
protein_identifications.resize(1);
OMSSAXMLFile().load(in, protein_identifications[0],
peptide_identifications, true);
}
else if (in_type == FileTypes::MASCOTXML)
{
String scan_regex = getStringOption_("scan_regex");
String exp_name = getStringOption_("mz_file");
MascotXMLFile::RTMapping rt_mapping;
if (!exp_name.empty())
{
PeakMap exp;
// load only MS2 spectra:
fh.getOptions().addMSLevel(2);
fh.loadExperiment(exp_name, exp, FileTypes::MZML, log_type_);
MascotXMLFile::generateRTMapping(exp.begin(), exp.end(), rt_mapping);
}
protein_identifications.resize(1);
MascotXMLFile().load(in, protein_identifications[0],
peptide_identifications, rt_mapping, scan_regex);
}
else if (in_type == FileTypes::XML)
{
ProteinIdentification protein_id;
XTandemXMLFile().load(in, protein_id, peptide_identifications);
protein_id.setSearchEngineVersion("");
protein_id.setSearchEngine("XTandem");
protein_identifications.push_back(protein_id);
String exp_name = getStringOption_("mz_file");
if (!exp_name.empty())
{
PeakMap exp;
fh.getOptions().addMSLevel(2);
fh.loadExperiment(exp_name, exp, FileTypes::MZML, log_type_);
for (vector<PeptideIdentification>::iterator it = peptide_identifications.begin(); it != peptide_identifications.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 << "XTandem xml: Error: id '" << id << "' not found in peak map!" << endl;
}
}
}
}
else
{
writeLog_("Unknown input file type given. Aborting!");
printUsage_();
return ILLEGAL_PARAMETERS;
}
}
logger.endProgress();
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
const String out = getStringOption_("out");
FileTypes::Type out_type = FileTypes::nameToType(getStringOption_("out_type"));
if (out_type == FileTypes::UNKNOWN)
{
out_type = fh.getTypeByFileName(out);
}
if (out_type == FileTypes::UNKNOWN)
{
writeLog_("Error: Could not determine output file type!");
return PARSE_ERROR;
}
logger.startProgress(0, 1, "Storing...");
if (out_type == FileTypes::PEPXML)
{
bool peptideprophet_analyzed = getFlag_("peptideprophet_analyzed");
String mz_file = getStringOption_("mz_file");
String mz_name = getStringOption_("mz_name");
PepXMLFile().store(out, protein_identifications, peptide_identifications, mz_file, mz_name, peptideprophet_analyzed);
}
else if (out_type == FileTypes::IDXML)
{
IdXMLFile().store(out, protein_identifications, peptide_identifications);
}
else if (out_type == FileTypes::MZIDENTML)
{
MzIdentMLFile().store(out, protein_identifications, peptide_identifications);
}
else if (out_type == FileTypes::FASTA)
{
Size count = 0;
ofstream fasta(out.c_str(), ios::out);
for (Size i = 0; i < peptide_identifications.size(); ++i)
{
for (Size l = 0; l < peptide_identifications[i].getHits().size(); ++l)
{
const PeptideHit& hit = peptide_identifications[i].getHits()[l];
fasta << ">" << hit.getSequence().toUnmodifiedString() << "|" << count++
<< "|" << hit.getSequence().toString() << endl;
String seq = hit.getSequence().toUnmodifiedString();
// FASTA files should have at most 60 characters of sequence info per line
for (Size j = 0; j < seq.size(); j += 60)
{
Size k = min(j + 60, seq.size());
fasta << string(seq[j], seq[k]) << endl;
}
}
}
}
else
{
writeLog_("Unsupported output file type given. Aborting!");
printUsage_();
return ILLEGAL_PARAMETERS;
}
logger.endProgress();
return EXECUTION_OK;
}
TOLERANCE_ABSOLUTE(0.01)
MzDataFile file;
PeakMap e;
// real test
file.load(OPENMS_GET_TEST_DATA_PATH("MzDataFile_1.mzData"), e);
//test DocumentIdentifier addition
TEST_STRING_EQUAL(e.getLoadedFilePath(), OPENMS_GET_TEST_DATA_PATH("MzDataFile_1.mzData"));
TEST_STRING_EQUAL(FileTypes::typeToName(e.getLoadedFileType()), "mzData");
//---------------------------------------------------------------------------
// ms-level, RT, native ID
//---------------------------------------------------------------------------
TEST_EQUAL(e.size(), 3)
TEST_EQUAL(e[0].getMSLevel(), 1)
TEST_EQUAL(e[1].getMSLevel(), 2)
TEST_EQUAL(e[2].getMSLevel(), 1)
TEST_REAL_SIMILAR(e[0].getRT(), 60)
TEST_REAL_SIMILAR(e[1].getRT(), 120)
TEST_REAL_SIMILAR(e[2].getRT(), 180)
TEST_STRING_EQUAL(e[0].getNativeID(), "spectrum=10")
TEST_STRING_EQUAL(e[1].getNativeID(), "spectrum=11")
TEST_STRING_EQUAL(e[2].getNativeID(), "spectrum=12")
TEST_EQUAL(e[0].getType(), SpectrumSettings::UNKNOWN)
//---------------------------------------------------------------------------
//meta data array meta data
//---------------------------------------------------------------------------
TEST_EQUAL(e[0].getFloatDataArrays()[0].getMetaValue("URL"), "www.open-ms.de")
TOLERANCE_ABSOLUTE(0.01)
PeakMap e;
DTA2DFile file;
//test exception
TEST_EXCEPTION( Exception::FileNotFound , file.load("dummy/dummy.dta2d",e) )
// real test
file.load(OPENMS_GET_TEST_DATA_PATH("DTA2DFile_test_1.dta2d"),e);
//test DocumentIdentifier addition
TEST_STRING_EQUAL(e.getLoadedFilePath(), OPENMS_GET_TEST_DATA_PATH("DTA2DFile_test_1.dta2d"));
TEST_STRING_EQUAL(FileTypes::typeToName(e.getLoadedFileType()),"dta2d");
TEST_EQUAL(e.size(), 9);
ABORT_IF(e.size() != 9)
TEST_STRING_EQUAL(e[0].getNativeID(),"index=0")
TEST_STRING_EQUAL(e[1].getNativeID(),"index=1")
TEST_STRING_EQUAL(e[2].getNativeID(),"index=2")
TEST_STRING_EQUAL(e[3].getNativeID(),"index=3")
TEST_STRING_EQUAL(e[4].getNativeID(),"index=4")
TEST_STRING_EQUAL(e[5].getNativeID(),"index=5")
TEST_STRING_EQUAL(e[6].getNativeID(),"index=6")
TEST_STRING_EQUAL(e[7].getNativeID(),"index=7")
TEST_STRING_EQUAL(e[8].getNativeID(),"index=8")
PeakMap::const_iterator it(e.begin());
TEST_REAL_SIMILAR((*it)[0].getPosition()[0], 230.02)
TEST_REAL_SIMILAR(it->getRT(), 4711.1)
chrom2.setChromatogramType(ChromatogramSettings::SELECTED_REACTION_MONITORING_CHROMATOGRAM);
ChromatogramPeak peak1, peak2, peak3;
peak1.setRT(0.1);
peak2.setRT(0.2);
peak3.setRT(0.3);
chrom1.push_back(peak1);
chrom1.push_back(peak2);
chrom2.push_back(peak2);
chrom2.push_back(peak2);
exp.addChromatogram(chrom1);
exp.addChromatogram(chrom2);
TEST_EQUAL(exp.size(), 0)
TEST_EQUAL(exp.getChromatograms().size(), 2)
ChromatogramTools().convertChromatogramsToSpectra(exp);
TEST_EQUAL(exp.size(), 4)
TEST_EQUAL(exp.getChromatograms().size(), 0)
TEST_REAL_SIMILAR(exp[0][0].getMZ(), 200.1)
TEST_EQUAL(exp[0].getPrecursors().size(), 1)
TEST_REAL_SIMILAR(exp[0].getPrecursors().begin()->getMZ(), 100.1)
}
END_SECTION
START_SECTION(template <typename ExperimentType> void convertSpectraToChromatograms(ExperimentType& exp, bool remove_spectra = false))
{
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;
}
START_SECTION((SpectraMerger(const SpectraMerger& source)))
SpectraMerger copy(*e_ptr);
TEST_EQUAL(copy.getParameters(), e_ptr->getParameters())
END_SECTION
START_SECTION((SpectraMerger& operator=(const SpectraMerger& source)))
SpectraMerger copy;
copy = *e_ptr;
TEST_EQUAL(copy.getParameters(), e_ptr->getParameters())
END_SECTION
START_SECTION((template < typename MapType > void mergeSpectraBlockWise(MapType &exp)))
PeakMap exp, exp2;
MzMLFile().load(OPENMS_GET_TEST_DATA_PATH("SpectraMerger_input_2.mzML"), exp);
TEST_EQUAL(exp.size(), 144)
exp2 = exp;
SpectraMerger merger;
Param p;
p.setValue("block_method:rt_block_size", 5);
p.setValue("block_method:ms_levels", IntList::create(StringList::create("1")));
merger.setParameters(p);
merger.mergeSpectraBlockWise(exp);
TEST_EQUAL(exp.size(), 130);
exp=exp2;
p.setValue("block_method:rt_block_size", 4);
p.setValue("block_method:ms_levels", IntList::create(StringList::create("2")));
merger.setParameters(p);
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;
}
ExitCodes main_(int , const char**) override
{
// path to the log file
String logfile(getStringOption_("log"));
String pepnovo_executable(getStringOption_("pepnovo_executable"));
PeakMap exp;
String inputfile_name = getStringOption_("in");
writeDebug_(String("Input file: ") + inputfile_name, 1);
String outputfile_name = getStringOption_("out");
writeDebug_(String("Output file: ") + outputfile_name, 1);
String model_directory = getStringOption_("model_directory");
writeDebug_(String("model directory: ") + model_directory, 1);
String model_name = getStringOption_("model");
writeDebug_(String("model directory: ") + model_name, 1);
double fragment_tolerance = getDoubleOption_("fragment_tolerance");
if (fragment_tolerance!=-1.0 && (fragment_tolerance<0 || fragment_tolerance>0.75))
{
writeLog_("Invalid fragment tolerance");
printUsage_();
return ILLEGAL_PARAMETERS;
}
double pm_tolerance = getDoubleOption_("pm_tolerance");
if (pm_tolerance!=-1.0 && (pm_tolerance<0.0 || pm_tolerance>5.0))
{
writeLog_("Invalid fragment tolerance");
printUsage_();
return ILLEGAL_PARAMETERS;
}
Int tag_length = getIntOption_("tag_length");
if ( tag_length!=-1 && (tag_length<3 || tag_length>6))
{
writeLog_("Invalid fragment tolerance");
printUsage_();
return ILLEGAL_PARAMETERS;
}
String digest = getStringOption_("digest");
Size num_solutions=getIntOption_("num_solutions");
//-------------------------------------------------------------
// reading input
//-------------------------------------------------------------
// only load msLevel 2
MzMLFile mzml_infile;
mzml_infile.getOptions().addMSLevel(2);
mzml_infile.setLogType(log_type_);
mzml_infile.load(inputfile_name, exp);
// we map the native id to the MZ and RT to be able to
// map the IDs back to the spectra (RT, and MZ Meta Information)
PepNovoOutfile::IndexPosMappingType index_to_precursor;
for (Size i = 0; i < exp.size(); ++i)
{
index_to_precursor[i]= make_pair(exp[i].getRT(), exp[i].getPrecursors()[0].getPosition()[0]); //set entry <RT, MZ>
}
logfile = getStringOption_("log");
QDir qdir_models_source(model_directory.c_str());
if (!qdir_models_source.exists())
{
writeLog_("The model directory does not exist");
return INPUT_FILE_NOT_FOUND;
}
// create temp directory
QDir qdir_temp(File::getTempDirectory().toQString());
String temp_data_directory = File::getUniqueName();
qdir_temp.mkdir(temp_data_directory.toQString());
qdir_temp.cd(temp_data_directory.toQString());
temp_data_directory = File::getTempDirectory() + "/" + temp_data_directory; // delete later
String mgf_file = temp_data_directory + "/" + File::getUniqueName() + ".mgf"; // the mzXML parser of PepNovo is somewhat broken.. don't use mzXML
MascotGenericFile().store(mgf_file, exp);
bool error(false);
try
{
//temporary File to store PepNovo output
String temp_pepnovo_outfile = qdir_temp.absoluteFilePath("tmp_pepnovo_out.txt");
String tmp_models_dir = qdir_temp.absoluteFilePath("Models");
std::map<String, String>mods_and_keys; //, key_to_id;
if (qdir_temp.cd("Models"))
{
writeLog_("The temporary directory already contains \"Model\" Folder. Please delete it and re-run. Aborting!");
return CANNOT_WRITE_OUTPUT_FILE;
}
else
{
qdir_temp.mkdir("Models");
qdir_temp.cd("Models");
}
//copy the Models folder of OpenMS into the temp_data_directory
QStringList pepnovo_files = qdir_models_source.entryList(QDir::Dirs | QDir::Files|QDir::NoDotAndDotDot);
if (pepnovo_files.empty())
{
writeLog_("The \"Model\" directory does not contain model files. Aborting!");
return INPUT_FILE_NOT_FOUND;
}
for (QStringList::ConstIterator file_it=pepnovo_files.begin(); file_it!=pepnovo_files.end(); ++file_it)
{
if (qdir_models_source.cd(*file_it))
{
qdir_temp.mkdir(*file_it);
qdir_temp.cd(*file_it);
QStringList subdir_files = qdir_models_source.entryList(QDir::Dirs | QDir::Files|QDir::NoDotAndDotDot);
for (QStringList::ConstIterator subdir_file_it=subdir_files.begin(); subdir_file_it!=subdir_files.end(); ++subdir_file_it)
{
QFile::copy(qdir_models_source.filePath(*subdir_file_it), qdir_temp.filePath(*subdir_file_it));
}
qdir_temp.cdUp();
qdir_models_source.cdUp();
}
else
{
QFile::copy(qdir_models_source.filePath(*file_it), qdir_temp.filePath(*file_it));
}
}
//generate PTM File and store in temp directory
PepNovoInfile p_novo_infile;
String ptm_command;
if (!getStringList_("fixed_modifications").empty() || !getStringList_("variable_modifications").empty())
{
p_novo_infile.setModifications(getStringList_("fixed_modifications"), getStringList_("variable_modifications"));
p_novo_infile.store(qdir_temp.filePath("PepNovo_PTMs.txt"));
pepnovo_files.append("PepNovo_PTMs.txt");
p_novo_infile.getModifications(mods_and_keys);
for (std::map<String, String>::const_iterator key_it=mods_and_keys.begin(); key_it!=mods_and_keys.end();++key_it)
{
if (ptm_command!="")
{
ptm_command+=":";
}
ptm_command+= key_it->first;
//key_to_id[key_it->second]=key_it->first;
}
}
//-------------------------------------------------------------
// (3) running program according to parameters
//-------------------------------------------------------------
QStringList arguments;
arguments << "-file" << mgf_file.toQString();
arguments << "-model" << model_name.toQString();
if (pm_tolerance != -1 ) arguments << "-pm_tolerance"<<String(pm_tolerance).toQString();
if (fragment_tolerance != -1 ) arguments << "-fragment_tolerance" <<String(fragment_tolerance).toQString();
if (!ptm_command.empty()) arguments <<"-PTMs" <<ptm_command.toQString();
if (getFlag_("correct_pm")) arguments << "-correct_pm";
if (getFlag_("use_spectrum_charge")) arguments << "-use_spectrum_charge";
if (getFlag_("use_spectrum_mz")) arguments << "-use_spectrum_mz";
if (getFlag_("no_quality_filter")) arguments << "-no_quality_filter";
arguments << "-digest" << digest.toQString();
arguments << "-num_solutions" << String(num_solutions).toQString();
if (tag_length!=-1) arguments<<"-tag_length" << String(tag_length).toQString();
arguments<<"-model_dir" << tmp_models_dir.toQString();
//arguments<<">" << temp_pepnovo_outfile.toQString();
writeDebug_("Use this line to call PepNovo: ", 1);
writeDebug_(pepnovo_executable + " " + String(arguments.join(" ")), 1);
QProcess process;
process.setStandardOutputFile(temp_pepnovo_outfile.toQString());
process.setStandardErrorFile(temp_pepnovo_outfile.toQString());
process.start(pepnovo_executable.toQString(), arguments); // does automatic escaping etc...
if (process.waitForFinished(-1))
{
//if PepNovo finished successfully use PepNovoOutfile to parse the results and generate idXML
std::vector< PeptideIdentification > peptide_identifications;
ProteinIdentification protein_identification;
StringList ms_runs;
exp.getPrimaryMSRunPath(ms_runs);
protein_identification.setPrimaryMSRunPath(ms_runs);
PepNovoOutfile p_novo_outfile;
//resolve PTMs (match them back to the OpenMs Identifier String)
std::vector<ProteinIdentification>prot_ids;
p_novo_outfile.load(temp_pepnovo_outfile, peptide_identifications, protein_identification, -1e5, index_to_precursor, mods_and_keys);
prot_ids.push_back(protein_identification);
IdXMLFile().store(outputfile_name, prot_ids, peptide_identifications);
}
if (process.exitStatus() != 0) error = true;
}
catch(Exception::BaseException &exc)
{
writeLog_(exc.what());
LOG_ERROR << "Error occurred: " << exc.what() << std::endl;
error = true;
}
if (!error)
{
File::removeDirRecursively(temp_data_directory);
return EXECUTION_OK;
}
else
{
writeLog_("PepNovo problem. Aborting! (Details can be seen in outfiles: '" + temp_data_directory + "')");
return EXTERNAL_PROGRAM_ERROR;
}
}
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;
}
void MassTraceDetection::run_(const MapIdxSortedByInt& chrom_apices,
const Size total_peak_count,
const PeakMap& work_exp,
const std::vector<Size>& spec_offsets,
std::vector<MassTrace>& found_masstraces)
{
boost::dynamic_bitset<> peak_visited(total_peak_count);
Size trace_number(1);
// check presence of FWHM meta data
int fwhm_meta_idx(-1);
Size fwhm_meta_count(0);
for (Size i = 0; i < work_exp.size(); ++i)
{
if (work_exp[i].getFloatDataArrays().size() > 0 &&
work_exp[i].getFloatDataArrays()[0].getName() == "FWHM_ppm")
{
if (work_exp[i].getFloatDataArrays()[0].size() != work_exp[i].size())
{ // float data should always have the same size as the corresponding array
throw Exception::InvalidSize(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, work_exp[i].size());
}
fwhm_meta_idx = 0;
++fwhm_meta_count;
}
}
if (fwhm_meta_count > 0 && fwhm_meta_count != work_exp.size())
{
throw Exception::Precondition(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION,
String("FWHM meta arrays are expected to be missing or present for all MS spectra [") + fwhm_meta_count + "/" + work_exp.size() + "].");
}
this->startProgress(0, total_peak_count, "mass trace detection");
Size peaks_detected(0);
for (MapIdxSortedByInt::const_reverse_iterator m_it = chrom_apices.rbegin(); m_it != chrom_apices.rend(); ++m_it)
{
Size apex_scan_idx(m_it->second.first);
Size apex_peak_idx(m_it->second.second);
if (peak_visited[spec_offsets[apex_scan_idx] + apex_peak_idx])
{
continue;
}
Peak2D apex_peak;
apex_peak.setRT(work_exp[apex_scan_idx].getRT());
apex_peak.setMZ(work_exp[apex_scan_idx][apex_peak_idx].getMZ());
apex_peak.setIntensity(work_exp[apex_scan_idx][apex_peak_idx].getIntensity());
Size trace_up_idx(apex_scan_idx);
Size trace_down_idx(apex_scan_idx);
std::list<PeakType> current_trace;
current_trace.push_back(apex_peak);
std::vector<double> fwhms_mz; // peak-FWHM meta values of collected peaks
// Initialization for the iterative version of weighted m/z mean calculation
double centroid_mz(apex_peak.getMZ());
double prev_counter(apex_peak.getIntensity() * apex_peak.getMZ());
double prev_denom(apex_peak.getIntensity());
updateIterativeWeightedMeanMZ(apex_peak.getMZ(), apex_peak.getIntensity(), centroid_mz, prev_counter, prev_denom);
std::vector<std::pair<Size, Size> > gathered_idx;
gathered_idx.push_back(std::make_pair(apex_scan_idx, apex_peak_idx));
if (fwhm_meta_idx != -1)
{
fwhms_mz.push_back(work_exp[apex_scan_idx].getFloatDataArrays()[fwhm_meta_idx][apex_peak_idx]);
}
Size up_hitting_peak(0), down_hitting_peak(0);
Size up_scan_counter(0), down_scan_counter(0);
bool toggle_up = true, toggle_down = true;
Size conseq_missed_peak_up(0), conseq_missed_peak_down(0);
Size max_consecutive_missing(trace_termination_outliers_);
double current_sample_rate(1.0);
// Size min_scans_to_consider(std::floor((min_sample_rate_ /2)*10));
Size min_scans_to_consider(5);
// double outlier_ratio(0.3);
// double ftl_mean(centroid_mz);
double ftl_sd((centroid_mz / 1e6) * mass_error_ppm_);
double intensity_so_far(apex_peak.getIntensity());
while (((trace_down_idx > 0) && toggle_down) ||
((trace_up_idx < work_exp.size() - 1) && toggle_up)
)
{
// *********************************************************** //
// Step 2.1 MOVE DOWN in RT dim
// *********************************************************** //
if ((trace_down_idx > 0) && toggle_down)
{
const MSSpectrum<>& spec_trace_down = work_exp[trace_down_idx - 1];
if (!spec_trace_down.empty())
{
Size next_down_peak_idx = spec_trace_down.findNearest(centroid_mz);
double next_down_peak_mz = spec_trace_down[next_down_peak_idx].getMZ();
double next_down_peak_int = spec_trace_down[next_down_peak_idx].getIntensity();
double right_bound = centroid_mz + 3 * ftl_sd;
double left_bound = centroid_mz - 3 * ftl_sd;
if ((next_down_peak_mz <= right_bound) &&
(next_down_peak_mz >= left_bound) &&
!peak_visited[spec_offsets[trace_down_idx - 1] + next_down_peak_idx]
)
{
Peak2D next_peak;
next_peak.setRT(spec_trace_down.getRT());
next_peak.setMZ(next_down_peak_mz);
next_peak.setIntensity(next_down_peak_int);
current_trace.push_front(next_peak);
// FWHM average
if (fwhm_meta_idx != -1)
{
fwhms_mz.push_back(spec_trace_down.getFloatDataArrays()[fwhm_meta_idx][next_down_peak_idx]);
}
// Update the m/z mean of the current trace as we added a new peak
updateIterativeWeightedMeanMZ(next_down_peak_mz, next_down_peak_int, centroid_mz, prev_counter, prev_denom);
gathered_idx.push_back(std::make_pair(trace_down_idx - 1, next_down_peak_idx));
// Update the m/z variance dynamically
if (reestimate_mt_sd_) // && (down_hitting_peak+1 > min_flank_scans))
{
// if (ftl_t > min_fwhm_scans)
{
updateWeightedSDEstimateRobust(next_peak, centroid_mz, ftl_sd, intensity_so_far);
}
}
++down_hitting_peak;
conseq_missed_peak_down = 0;
}
else
{
++conseq_missed_peak_down;
}
}
--trace_down_idx;
++down_scan_counter;
// trace termination criterion: max allowed number of
// consecutive outliers reached OR cancel extension if
// sampling_rate falls below min_sample_rate_
if (trace_termination_criterion_ == "outlier")
{
if (conseq_missed_peak_down > max_consecutive_missing)
{
toggle_down = false;
}
}
else if (trace_termination_criterion_ == "sample_rate")
{
current_sample_rate = (double)(down_hitting_peak + up_hitting_peak + 1) /
(double)(down_scan_counter + up_scan_counter + 1);
if (down_scan_counter > min_scans_to_consider && current_sample_rate < min_sample_rate_)
{
// std::cout << "stopping down..." << std::endl;
toggle_down = false;
}
}
}
// *********************************************************** //
// Step 2.2 MOVE UP in RT dim
// *********************************************************** //
if ((trace_up_idx < work_exp.size() - 1) && toggle_up)
{
const MSSpectrum<>& spec_trace_up = work_exp[trace_up_idx + 1];
if (!spec_trace_up.empty())
{
Size next_up_peak_idx = spec_trace_up.findNearest(centroid_mz);
double next_up_peak_mz = spec_trace_up[next_up_peak_idx].getMZ();
double next_up_peak_int = spec_trace_up[next_up_peak_idx].getIntensity();
double right_bound = centroid_mz + 3 * ftl_sd;
double left_bound = centroid_mz - 3 * ftl_sd;
if ((next_up_peak_mz <= right_bound) &&
(next_up_peak_mz >= left_bound) &&
!peak_visited[spec_offsets[trace_up_idx + 1] + next_up_peak_idx])
{
Peak2D next_peak;
next_peak.setRT(spec_trace_up.getRT());
next_peak.setMZ(next_up_peak_mz);
next_peak.setIntensity(next_up_peak_int);
current_trace.push_back(next_peak);
if (fwhm_meta_idx != -1)
{
fwhms_mz.push_back(spec_trace_up.getFloatDataArrays()[fwhm_meta_idx][next_up_peak_idx]);
}
// Update the m/z mean of the current trace as we added a new peak
updateIterativeWeightedMeanMZ(next_up_peak_mz, next_up_peak_int, centroid_mz, prev_counter, prev_denom);
gathered_idx.push_back(std::make_pair(trace_up_idx + 1, next_up_peak_idx));
// Update the m/z variance dynamically
if (reestimate_mt_sd_) // && (up_hitting_peak+1 > min_flank_scans))
{
// if (ftl_t > min_fwhm_scans)
{
updateWeightedSDEstimateRobust(next_peak, centroid_mz, ftl_sd, intensity_so_far);
}
}
++up_hitting_peak;
conseq_missed_peak_up = 0;
}
else
{
++conseq_missed_peak_up;
}
}
++trace_up_idx;
++up_scan_counter;
if (trace_termination_criterion_ == "outlier")
{
if (conseq_missed_peak_up > max_consecutive_missing)
{
toggle_up = false;
}
}
else if (trace_termination_criterion_ == "sample_rate")
{
current_sample_rate = (double)(down_hitting_peak + up_hitting_peak + 1) / (double)(down_scan_counter + up_scan_counter + 1);
if (up_scan_counter > min_scans_to_consider && current_sample_rate < min_sample_rate_)
{
// std::cout << "stopping up" << std::endl;
toggle_up = false;
}
}
}
}
// std::cout << "current sr: " << current_sample_rate << std::endl;
double num_scans(down_scan_counter + up_scan_counter + 1 - conseq_missed_peak_down - conseq_missed_peak_up);
double mt_quality((double)current_trace.size() / (double)num_scans);
// std::cout << "mt quality: " << mt_quality << std::endl;
double rt_range(std::fabs(current_trace.rbegin()->getRT() - current_trace.begin()->getRT()));
// *********************************************************** //
// Step 2.3 check if minimum length and quality of mass trace criteria are met
// *********************************************************** //
bool max_trace_criteria = (max_trace_length_ < 0.0 || rt_range < max_trace_length_);
if (rt_range >= min_trace_length_ && max_trace_criteria && mt_quality >= min_sample_rate_)
{
// std::cout << "T" << trace_number << "\t" << mt_quality << std::endl;
// mark all peaks as visited
for (Size i = 0; i < gathered_idx.size(); ++i)
{
peak_visited[spec_offsets[gathered_idx[i].first] + gathered_idx[i].second] = true;
}
// create new MassTrace object and store collected peaks from list current_trace
MassTrace new_trace(current_trace);
new_trace.updateWeightedMeanRT();
new_trace.updateWeightedMeanMZ();
if (!fwhms_mz.empty()) new_trace.fwhm_mz_avg = Math::median(fwhms_mz.begin(), fwhms_mz.end());
new_trace.setQuantMethod(quant_method_);
//new_trace.setCentroidSD(ftl_sd);
new_trace.updateWeightedMZsd();
new_trace.setLabel("T" + String(trace_number));
++trace_number;
found_masstraces.push_back(new_trace);
peaks_detected += new_trace.getSize();
this->setProgress(peaks_detected);
}
}
this->endProgress();
}
}
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)"));
params.setValue("variable_modifications", ListUtils::create<String>("Oxidation (M),Deamidated (N),Deamidated (Q)"));
ptr->setParameters(params);
ExitCodes main_(int, const char**) override
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
//input file names
String in = getStringOption_("in");
String read_method = getStringOption_("read_method");
bool load_data = getStringOption_("loadData") == "true";
if (read_method == "streaming")
{
std::cout << "Read method: streaming" << std::endl;
// Create the consumer, set output file name, transform
TICConsumer consumer;
MzMLFile mzml;
mzml.setLogType(log_type_);
PeakFileOptions opt = mzml.getOptions();
opt.setFillData(load_data); // whether to actually load any data
opt.setSkipXMLChecks(true); // save time by not checking base64 strings for whitespaces
opt.setMaxDataPoolSize(100);
opt.setAlwaysAppendData(false);
mzml.setOptions(opt);
mzml.transform(in, &consumer, true, true);
std::cout << "There are " << consumer.nr_spectra << " spectra and " << consumer.nr_peaks << " peaks in the input file." << std::endl;
std::cout << "The total ion current is " << consumer.TIC << std::endl;
size_t after;
SysInfo::getProcessMemoryConsumption(after);
std::cout << " Memory consumption after " << after << std::endl;
}
else if (read_method == "regular")
{
std::cout << "Read method: regular" << std::endl;
MzMLFile mzml;
mzml.setLogType(log_type_);
PeakFileOptions opt = mzml.getOptions();
opt.setFillData(load_data); // whether to actually load any data
opt.setSkipXMLChecks(true); // save time by not checking base64 strings for whitespaces
mzml.setOptions(opt);
PeakMap map;
mzml.load(in, map);
double TIC = 0.0;
long int nr_peaks = 0;
for (Size i =0; i < map.size(); i++)
{
nr_peaks += map[i].size();
for (Size j = 0; j < map[i].size(); j++)
{
TIC += map[i][j].getIntensity();
}
}
std::cout << "There are " << map.size() << " spectra and " << nr_peaks << " peaks in the input file." << std::endl;
std::cout << "The total ion current is " << TIC << std::endl;
size_t after;
SysInfo::getProcessMemoryConsumption(after);
std::cout << " Memory consumption after " << after << std::endl;
}
else if (read_method == "indexed")
{
std::cout << "Read method: indexed" << std::endl;
IndexedMzMLFileLoader imzml;
// load data from an indexed MzML file
OnDiscPeakMap map;
imzml.load(in, map);
double TIC = 0.0;
long int nr_peaks = 0;
if (load_data)
{
for (Size i =0; i < map.getNrSpectra(); i++)
{
OpenMS::Interfaces::SpectrumPtr sptr = map.getSpectrumById(i);
nr_peaks += sptr->getIntensityArray()->data.size();
TIC += std::accumulate(sptr->getIntensityArray()->data.begin(), sptr->getIntensityArray()->data.end(), 0.0);
}
}
std::cout << "There are " << map.getNrSpectra() << " spectra and " << nr_peaks << " peaks in the input file." << std::endl;
std::cout << "The total ion current is " << TIC << std::endl;
size_t after;
SysInfo::getProcessMemoryConsumption(after);
std::cout << " Memory consumption after " << after << std::endl;
}
else if (read_method == "indexed_parallel")
{
std::cout << "Read method: indexed (parallel)" << std::endl;
IndexedMzMLFileLoader imzml;
PeakFileOptions opt = imzml.getOptions();
opt.setFillData(load_data); // whether to actually load any data
imzml.setOptions(opt);
// load data from an indexed MzML file
OnDiscPeakMap map;
map.openFile(in, true);
map.setSkipXMLChecks(true);
double TIC = 0.0;
long int nr_peaks = 0;
if (load_data)
{
// firstprivate means that each thread has its own instance of the
// variable, each copy initialized with the initial value
#ifdef _OPENMP
#pragma omp parallel for firstprivate(map)
#endif
for (SignedSize i =0; i < (SignedSize)map.getNrSpectra(); i++)
{
OpenMS::Interfaces::SpectrumPtr sptr = map.getSpectrumById(i);
double nr_peaks_l = sptr->getIntensityArray()->data.size();
double TIC_l = std::accumulate(sptr->getIntensityArray()->data.begin(), sptr->getIntensityArray()->data.end(), 0.0);
#ifdef _OPENMP
#pragma omp critical (indexed)
#endif
{
TIC += TIC_l;
nr_peaks += nr_peaks_l;
}
}
}
std::cout << "There are " << map.getNrSpectra() << " spectra and " << nr_peaks << " peaks in the input file." << std::endl;
std::cout << "The total ion current is " << TIC << std::endl;
size_t after;
SysInfo::getProcessMemoryConsumption(after);
std::cout << " Memory consumption after " << after << std::endl;
}
else if (read_method == "cached")
{
std::cout << "Read method: cached" << std::endl;
// Special handling of cached mzML as input types:
// we expect two paired input files which we should read into exp
std::vector<String> split_out;
in.split(".cachedMzML", split_out);
if (split_out.size() != 2)
{
LOG_ERROR << "Cannot deduce base path from input '" << in <<
"' (note that '.cachedMzML' should only occur once as the final ending)" << std::endl;
return ILLEGAL_PARAMETERS;
}
String in_meta = split_out[0] + ".mzML";
MzMLFile f;
f.setLogType(log_type_);
CachedmzML cacher;
cacher.setLogType(log_type_);
CachedmzML cache;
cache.createMemdumpIndex(in);
const std::vector<std::streampos> spectra_index = cache.getSpectraIndex();
std::ifstream ifs_;
ifs_.open(in.c_str(), std::ios::binary);
double TIC = 0.0;
long int nr_peaks = 0;
for (Size i=0; i < spectra_index.size(); ++i)
{
BinaryDataArrayPtr mz_array(new BinaryDataArray);
BinaryDataArrayPtr intensity_array(new BinaryDataArray);
int ms_level = -1;
double rt = -1.0;
ifs_.seekg(spectra_index[i]);
CachedmzML::readSpectrumFast(mz_array, intensity_array, ifs_, ms_level, rt);
nr_peaks += intensity_array->data.size();
for (Size j = 0; j < intensity_array->data.size(); j++)
{
TIC += intensity_array->data[j];
}
}
std::cout << "There are " << spectra_index.size() << " spectra and " << nr_peaks << " peaks in the input file." << std::endl;
std::cout << "The total ion current is " << TIC << std::endl;
size_t after;
SysInfo::getProcessMemoryConsumption(after);
std::cout << " Memory consumption after " << after << std::endl;
}
else if (read_method == "cached_parallel")
{
std::cout << "Read method: cached parallel" << std::endl;
// Special handling of cached mzML as input types:
// we expect two paired input files which we should read into exp
std::vector<String> split_out;
in.split(".cachedMzML", split_out);
if (split_out.size() != 2)
{
LOG_ERROR << "Cannot deduce base path from input '" << in <<
"' (note that '.cachedMzML' should only occur once as the final ending)" << std::endl;
return ILLEGAL_PARAMETERS;
}
String in_meta = split_out[0] + ".mzML";
MzMLFile f;
f.setLogType(log_type_);
CachedmzML cacher;
cacher.setLogType(log_type_);
CachedmzML cache;
cache.createMemdumpIndex(in);
const std::vector<std::streampos> spectra_index = cache.getSpectraIndex();
FileAbstraction filestream(in);
double TIC = 0.0;
long int nr_peaks = 0;
#ifdef _OPENMP
#pragma omp parallel for firstprivate(filestream)
#endif
for (SignedSize i=0; i < (SignedSize)spectra_index.size(); ++i)
{
BinaryDataArrayPtr mz_array(new BinaryDataArray);
BinaryDataArrayPtr intensity_array(new BinaryDataArray);
int ms_level = -1;
double rt = -1.0;
// we only change the position of the thread-local filestream
filestream.getStream().seekg(spectra_index[i]);
CachedmzML::readSpectrumFast(mz_array, intensity_array, filestream.getStream(), ms_level, rt);
double nr_peaks_l = intensity_array->data.size();
double TIC_l = std::accumulate(intensity_array->data.begin(), intensity_array->data.end(), 0.0);
#ifdef _OPENMP
#pragma omp critical (indexed)
#endif
{
TIC += TIC_l;
nr_peaks += nr_peaks_l;
}
}
std::cout << "There are " << spectra_index.size() << " spectra and " << nr_peaks << " peaks in the input file." << std::endl;
std::cout << "The total ion current is " << TIC << std::endl;
size_t after;
SysInfo::getProcessMemoryConsumption(after);
std::cout << " Memory consumption after " << after << std::endl;
}
return EXECUTION_OK;
}
