void Normalizer::filterPeakMap(PeakMap & exp)
{
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
filterSpectrum(*it);
}
}
void ParentPeakMower::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;
}
// 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 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**)
{
//-------------------------------------------------------------
// 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;
}
ExitCodes main_(int, const char**)
{
// parsing parameters
String in(getStringOption_("in"));
String feature_in(getStringOption_("feature_in"));
String out(getStringOption_("out"));
double precursor_mass_tolerance(getDoubleOption_("precursor_mass_tolerance"));
// reading input
FileHandler fh;
FileTypes::Type in_type = fh.getType(in);
PeakMap exp;
fh.loadExperiment(in, exp, in_type, log_type_, false, false);
exp.sortSpectra();
FeatureMap feature_map;
if (feature_in != "")
{
FeatureXMLFile().load(feature_in, feature_map);
}
// calculations
FeatureFinderAlgorithmIsotopeWavelet iso_ff;
Param ff_param(iso_ff.getParameters());
ff_param.setValue("max_charge", getIntOption_("max_charge"));
ff_param.setValue("intensity_threshold", getDoubleOption_("intensity_threshold"));
iso_ff.setParameters(ff_param);
FeatureFinder ff;
ff.setLogType(ProgressLogger::NONE);
PeakMap exp2 = exp;
exp2.clear(false);
for (PeakMap::ConstIterator it = exp.begin(); it != exp.end(); ++it)
{
if (it->size() != 0)
{
exp2.addSpectrum(*it);
}
}
exp = exp2;
exp.updateRanges();
// TODO check MS2 and MS1 counts
ProgressLogger progresslogger;
progresslogger.setLogType(log_type_);
progresslogger.startProgress(0, exp.size(), "Correcting precursor masses");
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
progresslogger.setProgress(exp.end() - it);
if (it->getMSLevel() != 2)
{
continue;
}
// find first MS1 scan of the MS/MS scan
PeakMap::Iterator ms1_it = it;
while (ms1_it != exp.begin() && ms1_it->getMSLevel() != 1)
{
--ms1_it;
}
if (ms1_it == exp.begin() && ms1_it->getMSLevel() != 1)
{
writeLog_("Did not find a MS1 scan to the MS/MS scan at RT=" + String(it->getRT()));
continue;
}
if (ms1_it->size() == 0)
{
writeDebug_("No peaks in scan at RT=" + String(ms1_it->getRT()) + String(", skipping"), 1);
continue;
}
PeakMap::Iterator ms2_it = ms1_it;
++ms2_it;
while (ms2_it != exp.end() && ms2_it->getMSLevel() == 2)
{
// first: error checks
if (ms2_it->getPrecursors().empty())
{
writeDebug_("Warning: found no precursors of spectrum RT=" + String(ms2_it->getRT()) + ", skipping it.", 1);
++ms2_it;
continue;
}
else if (ms2_it->getPrecursors().size() > 1)
{
writeLog_("Warning: found more than one precursor of spectrum RT=" + String(ms2_it->getRT()) + ", using first one.");
}
Precursor prec = *ms2_it->getPrecursors().begin();
double prec_pos = prec.getMZ();
PeakMap new_exp;
// now excise small region from the MS1 spec for the feature finder (isotope pattern must be covered...)
PeakSpectrum zoom_spec;
for (PeakSpectrum::ConstIterator pit = ms1_it->begin(); pit != ms1_it->end(); ++pit)
{
if (pit->getMZ() > prec_pos - 3 && pit->getMZ() < prec_pos + 3)
{
zoom_spec.push_back(*pit);
}
}
new_exp.addSpectrum(zoom_spec);
new_exp.updateRanges();
FeatureMap features, seeds;
ff.run("isotope_wavelet", new_exp, features, ff_param, seeds);
if (features.empty())
{
writeDebug_("No features found for scan RT=" + String(ms1_it->getRT()), 1);
++ms2_it;
continue;
}
double max_int(numeric_limits<double>::min());
double min_dist(numeric_limits<double>::max());
Size max_int_feat_idx(0);
for (Size i = 0; i != features.size(); ++i)
{
if (fabs(features[i].getMZ() - prec_pos) < precursor_mass_tolerance &&
features[i].getIntensity() > max_int)
{
max_int_feat_idx = i;
max_int = features[i].getIntensity();
min_dist = fabs(features[i].getMZ() - prec_pos);
}
}
writeDebug_(" max_int=" + String(max_int) + " mz=" + String(features[max_int_feat_idx].getMZ()) + " charge=" + String(features[max_int_feat_idx].getCharge()), 5);
if (min_dist < precursor_mass_tolerance)
{
prec.setMZ(features[max_int_feat_idx].getMZ());
prec.setCharge(features[max_int_feat_idx].getCharge());
vector<Precursor> precs;
precs.push_back(prec);
ms2_it->setPrecursors(precs);
writeDebug_("Correcting precursor mass of spectrum RT=" + String(ms2_it->getRT()) + " from " + String(prec_pos) + " to " + String(prec.getMZ()) + " (z=" + String(prec.getCharge()) + ")", 1);
}
++ms2_it;
}
it = --ms2_it;
}
progresslogger.endProgress();
// writing output
fh.storeExperiment(out, exp, log_type_);
return EXECUTION_OK;
}
ExitCodes main_(int, const char**)
{
// 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;
}
void MassTraceDetection::run(const PeakMap& input_exp, std::vector<MassTrace>& found_masstraces)
{
// make sure the output vector is empty
found_masstraces.clear();
// gather all peaks that are potential chromatographic peak apices
// - use work_exp for actual work (remove peaks below noise threshold)
// - store potential apices in chrom_apices
PeakMap work_exp;
MapIdxSortedByInt chrom_apices;
Size total_peak_count(0);
std::vector<Size> spec_offsets;
spec_offsets.push_back(0);
Size spectra_count(0);
// *********************************************************** //
// Step 1: Detecting potential chromatographic apices
// *********************************************************** //
for (PeakMap::ConstIterator it = input_exp.begin(); it != input_exp.end(); ++it)
{
// check if this is a MS1 survey scan
if (it->getMSLevel() != 1) continue;
std::vector<Size> indices_passing;
for (Size peak_idx = 0; peak_idx < it->size(); ++peak_idx)
{
double tmp_peak_int((*it)[peak_idx].getIntensity());
if (tmp_peak_int > noise_threshold_int_)
{
// Assume that noise_threshold_int_ contains the noise level of the
// data and we want to be chrom_peak_snr times above the noise level
// --> add this peak as possible chromatographic apex
if (tmp_peak_int > chrom_peak_snr_ * noise_threshold_int_)
{
chrom_apices.insert(std::make_pair(tmp_peak_int, std::make_pair(spectra_count, indices_passing.size())));
}
indices_passing.push_back(peak_idx);
++total_peak_count;
}
}
PeakMap::SpectrumType tmp_spec(*it);
tmp_spec.select(indices_passing);
work_exp.addSpectrum(tmp_spec);
spec_offsets.push_back(spec_offsets.back() + tmp_spec.size());
++spectra_count;
}
if (spectra_count < 3)
{
throw Exception::InvalidValue(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION,
"Input map consists of too few MS1 spectra (less than 3!). Aborting...", String(spectra_count));
}
// discard last spectrum's offset
spec_offsets.pop_back();
// *********************************************************************
// Step 2: start extending mass traces beginning with the apex peak (go
// through all peaks in order of decreasing intensity)
// *********************************************************************
run_(chrom_apices, total_peak_count, work_exp, spec_offsets, found_masstraces);
return;
} // end of MassTraceDetection::run
ExitCodes main_(int, const char **) override
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
String in = getStringOption_("in");
String out = getStringOption_("out");
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
PeakMap exp;
MzMLFile f;
f.load(in, exp);
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
//determine maximum peak
exp.updateRanges();
double max = exp.getMaxInt() / 100.0;
for (PeakMap::Iterator it = exp.begin(); it != exp.end(); ++it)
{
if (it->getMSLevel() < 2)
{
for (PeakMap::SpectrumType::Iterator it2 = it->begin(); it2 != it->end(); ++it2)
{
it2->setIntensity(it2->getIntensity() / max);
}
}
}
/// @todo add chromatogram support for normalization, e.g. for MRM stuff (Andreas)
/*
vector<MSChromatogram > chroms = exp.getChromatograms();
double sum(0);
for (vector<MSChromatogram >::iterator it = chroms.begin(); it != chroms.end(); ++it)
{
for (MSChromatogram::Iterator it2 = it->begin(); it2 != it->end(); ++it2)
{
sum += it2->getIntensity();
}
}
for (vector<MSChromatogram >::iterator it = chroms.begin(); it != chroms.end(); ++it)
{
for (MSChromatogram::Iterator it2 = it->begin(); it2 != it->end(); ++it2)
{
it2->setIntensity(it2->getIntensity() / sum * 1000000.0);
}
}
exp.setChromatograms(chroms);
*/
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(exp, getProcessingInfo_(DataProcessing::NORMALIZATION));
f.store(out, exp);
return EXECUTION_OK;
}
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;
}