ExitCodes main_(int, const char **) override
{
//-------------------------------------------------------------
// parameter handling
//-------------------------------------------------------------
//input/output files
String in(getStringOption_("in"));
String out(getStringOption_("out"));
//-------------------------------------------------------------
// loading input
//-------------------------------------------------------------
PeakMap exp;
MzMLFile f;
f.setLogType(log_type_);
PeakFileOptions options;
options.clearMSLevels();
options.addMSLevel(2);
f.getOptions() = options;
f.load(in, exp);
writeDebug_("Data set contains " + String(exp.size()) + " spectra", 1);
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
vector<PeptideIdentification> pep_ids;
CompNovoIdentificationCID comp_novo_id;
// set the options
Param algorithm_param = getParam_().copy("algorithm:", true);
comp_novo_id.setParameters(algorithm_param);
comp_novo_id.getIdentifications(pep_ids, exp);
algorithm_param = comp_novo_id.getParameters();
//-------------------------------------------------------------
// writing output
//-------------------------------------------------------------
DateTime now = DateTime::now();
String date_string = now.get();
String identifier("CompNovoCID_" + date_string);
for (vector<PeptideIdentification>::iterator it = pep_ids.begin(); it != pep_ids.end(); ++it)
{
it->assignRanks();
it->setIdentifier(identifier);
}
vector<ProteinIdentification> prot_ids;
ProteinIdentification prot_id;
prot_id.setIdentifier(identifier);
prot_id.setDateTime(now);
StringList ms_runs;
exp.getPrimaryMSRunPath(ms_runs);
prot_id.setPrimaryMSRunPath(ms_runs);
ProteinIdentification::SearchParameters search_parameters;
search_parameters.charges = "+2-+3";
if (algorithm_param.getValue("tryptic_only").toBool())
{
search_parameters.digestion_enzyme = *(ProteaseDB::getInstance()->getEnzyme("Trypsin"));
}
else
{
search_parameters.digestion_enzyme = *(ProteaseDB::getInstance()->getEnzyme("no cleavage"));
}
search_parameters.mass_type = ProteinIdentification::MONOISOTOPIC;
search_parameters.fixed_modifications = algorithm_param.getValue("fixed_modifications");
search_parameters.variable_modifications = algorithm_param.getValue("variable_modifications");
search_parameters.missed_cleavages = (UInt)algorithm_param.getValue("missed_cleavages");
search_parameters.fragment_mass_tolerance = (double)algorithm_param.getValue("fragment_mass_tolerance");
search_parameters.precursor_mass_tolerance = (double)algorithm_param.getValue("precursor_mass_tolerance");
search_parameters.fragment_mass_tolerance_ppm = false;
search_parameters.precursor_mass_tolerance_ppm = false;
prot_id.setSearchParameters(search_parameters);
prot_id.setSearchEngineVersion("0.9beta");
prot_id.setSearchEngine("CompNovo");
prot_ids.push_back(prot_id);
IdXMLFile().store(out, prot_ids, pep_ids);
return EXECUTION_OK;
}
ExitCodes main_(int, const char**) override
{
String tmp_dir = QDir::toNativeSeparators((File::getTempDirectory() + "/" + File::getUniqueName() + "/").toQString()); // body for the tmp files
{
QDir d;
d.mkpath(tmp_dir.toQString());
}
String logfile(getStringOption_("log"));
String myrimatch_executable(getStringOption_("myrimatch_executable"));
//-------------------------------------------------------------
// get version of MyriMatch
//-------------------------------------------------------------
QProcess qp;
String myrimatch_version;
MyriMatchVersion myrimatch_version_i;
// we invoke myrimatch w/o arguments. that yields a return code != 0. but
// there is no other way for version 2.1 to get the version number
qp.start(myrimatch_executable.toQString(), QStringList(), QIODevice::ReadOnly); // does automatic escaping etc...
qp.waitForFinished();
String output(QString(qp.readAllStandardOutput()));
vector<String> lines;
vector<String> version_split;
output.split('\n', lines);
// the version number is expected to be in the second line
if (lines.size() < 2)
{
writeLog_("Warning: MyriMatch version output (" + output + ") not formatted as expected!");
return EXTERNAL_PROGRAM_ERROR;
}
// the version is expected to be something like:
// MyriMatch 2.1.111 (2011-12-27)
lines[1].split(' ', version_split);
if (version_split.size() == 3 && getVersion_(version_split[1], myrimatch_version_i))
{
myrimatch_version = version_split[1].removeWhitespaces();
writeDebug_("Setting MyriMatch version to " + myrimatch_version, 1);
}
else
{
writeLog_("Warning: MyriMatch version output (" + output + ") not formatted as expected!");
return EXTERNAL_PROGRAM_ERROR;
}
if (! ( (myrimatch_version_i.myrimatch_major == 2) && // major must be 2
(myrimatch_version_i.myrimatch_minor == 1 || myrimatch_version_i.myrimatch_minor == 2) // minor .1 or .2
))
{
writeLog_("Warning: unsupported MyriMatch version (" + myrimatch_version + "). Tested only for MyriMatch 2.1.x and 2.2.x."
"\nIf you encounter parameter errors, you can try the flag 'ignoreConfigErrors', but be aware that MyriMatch might be misconfigured.");
}
//-------------------------------------------------------------
// parsing parameters
//-------------------------------------------------------------
String inputfile_name = File::absolutePath(getStringOption_("in"));
String outputfile_name = getStringOption_("out");
String db_name = File::absolutePath(String(getStringOption_("database")));
// building parameter String
StringList parameters;
if (getFlag_("ignoreConfigErrors")) parameters << "-ignoreConfigErrors";
// Common Identification engine options
StringList static_mod_list;
StringList dynamic_mod_list;
translateModifications(static_mod_list, dynamic_mod_list);
if (!static_mod_list.empty())
parameters << "-StaticMods" << ListUtils::concatenate(static_mod_list, " ");
if (!dynamic_mod_list.empty())
parameters << "-DynamicMods" << ListUtils::concatenate(dynamic_mod_list, " ");
parameters << "-ProteinDatabase" << File::absolutePath(db_name);
if (getFlag_("precursor_mass_tolerance_avg"))
{
parameters << "-AvgPrecursorMzTolerance";
}
else
{
parameters << "-MonoPrecursorMzTolerance";
}
String precursor_mass_tolerance_unit = getStringOption_("precursor_mass_tolerance_unit") == "Da" ? " m/z" : " ppm";
parameters << String(getDoubleOption_("precursor_mass_tolerance")) + precursor_mass_tolerance_unit;
String fragment_mass_tolerance_unit = getStringOption_("fragment_mass_tolerance_unit");
if (fragment_mass_tolerance_unit == "Da")
{
fragment_mass_tolerance_unit = "m/z";
}
parameters << "-FragmentMzTolerance" << String(getDoubleOption_("fragment_mass_tolerance")) + " " + fragment_mass_tolerance_unit;
StringList slf = getStringList_("SpectrumListFilters");
if (slf.size() > 0)
{
if (myrimatch_version_i.myrimatch_minor <= 1)
{ // use quotes around the slf arguments (will be added automatically by Qt during call), i.e. "-SpectrumListFilters" "peakPicking false 2-"
parameters << "-SpectrumListFilters" << ListUtils::concatenate(slf, ";") << "";
}
else
{ // no quotes -- pass a single argument, i.e. "-SpectrumListFilters peakPicking false 2-"
parameters << "-SpectrumListFilters " + ListUtils::concatenate(slf, ";") << "";
}
}
//parameters << "-ThreadCountMultiplier" << String(getIntOption_("threads")); // MyriMatch does not recognise this, even though it's in the manual.
// MyriMatch specific parameters
parameters << "-NumChargeStates" << getIntOption_("NumChargeStates");
parameters << "-TicCutoffPercentage" << String(getDoubleOption_("TicCutoffPercentage"));
parameters << "-MaxDynamicMods" << getIntOption_("MaxDynamicMods");
parameters << "-MaxResultRank" << getIntOption_("MaxResultRank");
parameters << "-MinTerminiCleavages" << getIntOption_("MinTerminiCleavages");
parameters << "-MaxMissedCleavages" << getIntOption_("MaxMissedCleavages");
String cleavage_rule = getStringOption_("CleavageRules");
if (cleavage_rule.empty())
{
cleavage_rule = "Trypsin/P";
}
parameters << "-CleavageRules" << cleavage_rule;
// advanced parameters
parameters << "-MinPeptideMass" << getDoubleOption_("MinPeptideMass");
parameters << "-MaxPeptideMass" << getDoubleOption_("MaxPeptideMass");
parameters << "-MinPeptideLength" << getIntOption_("MinPeptideLength");
parameters << "-MaxPeptideLength" << getIntOption_("MaxPeptideLength");
parameters << "-NumIntensityClasses" << getIntOption_("NumIntensityClasses");
parameters << "-ClassSizeMultiplier" << getDoubleOption_("ClassSizeMultiplier");
parameters << "-MonoisotopeAdjustmentSet" << getStringOption_("MonoisotopeAdjustmentSet");
parameters << "-cpus" << getIntOption_("threads");
// Constant parameters
// DecoyPrefix worked only when set through the config file
String cfg_file = tmp_dir + "myrimatch.cfg";
ofstream f(cfg_file.c_str());
f << "DecoyPrefix=\"\"\n";
f.close();
parameters << "-cfg" << cfg_file;
// path to input file must be the last parameter
parameters << inputfile_name;
//-------------------------------------------------------------
// calculations
//-------------------------------------------------------------
QStringList qparam;
writeDebug_("MyriMatch arguments:", 1);
writeDebug_(String("\"") + ListUtils::concatenate(parameters, "\" \"") + "\"", 1);
for (Size i = 0; i < parameters.size(); ++i)
{
qparam << parameters[i].toQString();
}
QProcess process;
// Bad style, because it breaks relative paths?
process.setWorkingDirectory(tmp_dir.toQString());
process.start(myrimatch_executable.toQString(), qparam, QIODevice::ReadOnly);
bool success = process.waitForFinished(-1);
String myri_msg(QString(process.readAllStandardOutput()));
String myri_err(QString(process.readAllStandardError()));
writeDebug_(myri_msg, 1);
writeDebug_(myri_err, 0);
if (!success || process.exitStatus() != 0 || process.exitCode() != 0)
{
writeLog_("Error: MyriMatch problem! (Details can be seen in the logfile: \"" + logfile + "\")");
writeLog_("Note: This message can also be triggered if you run out of space in your tmp directory");
return EXTERNAL_PROGRAM_ERROR;
}
//-------------------------------------------------------------
// reading MyriMatch output
//-------------------------------------------------------------
writeDebug_("Reading output of MyriMatch", 5);
String exp_name = File::basename(inputfile_name);
String pep_file = tmp_dir + File::removeExtension(exp_name) + ".pepXML";
vector<ProteinIdentification> protein_identifications;
vector<PeptideIdentification> peptide_identifications;
PeakMap exp;
if (File::exists(pep_file))
{
MzMLFile fh;
fh.load(inputfile_name, exp);
SpectrumMetaDataLookup lookup;
lookup.readSpectra(exp.getSpectra());
PepXMLFile().load(pep_file, protein_identifications,
peptide_identifications, exp_name, lookup);
}
else
{
writeLog_("Error: MyriMatch problem! No pepXML output file (expected as '" + pep_file + "') was generated by MyriMatch.");
writeLog_("Note: This message can be triggered if no MS2 spectra were found or no identifications were made.");
writeLog_(" Myrimatch expects MS2 spectra in mzML files to contain the MSn tag. MSSpectrum with MS level 2 is not sufficient. You can use FileConverter to create such an mzML file by converting from mzML --> mzXML --> mzML.");
return EXTERNAL_PROGRAM_ERROR;
}
if (debug_level_ == 0)
{
QFile(pep_file.toQString()).remove();
QFile(cfg_file.toQString()).remove();
}
else
{
writeDebug_(String("Not removing '") + pep_file + "' for debugging purposes. Please delete manually!", 1);
writeDebug_(String("Not removing '") + cfg_file + "' for debugging purposes. Please delete manually!", 1);
}
//-------------------------------------------------------------
// writing results
//-------------------------------------------------------------
ProteinIdentification::SearchParameters search_parameters;
search_parameters.db = getStringOption_("database");
ProteinIdentification::PeakMassType mass_type = getFlag_("precursor_mass_tolerance_avg") == true ? ProteinIdentification::AVERAGE : 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_("MaxMissedCleavages");
search_parameters.fragment_mass_tolerance = getDoubleOption_("fragment_mass_tolerance");
search_parameters.precursor_mass_tolerance = getDoubleOption_("precursor_mass_tolerance");
search_parameters.precursor_mass_tolerance_ppm = getStringOption_("precursor_mass_tolerance_unit") == "ppm" ? true : false;
search_parameters.fragment_mass_tolerance_ppm = getStringOption_("fragment_mass_tolerance_unit") == "ppm" ? true : false;
protein_identifications[0].setSearchParameters(search_parameters);
protein_identifications[0].setSearchEngineVersion(myrimatch_version);
protein_identifications[0].setSearchEngine("MyriMatch");
if (!protein_identifications.empty())
{
StringList ms_runs;
exp.getPrimaryMSRunPath(ms_runs);
protein_identifications[0].setPrimaryMSRunPath(ms_runs);
}
IdXMLFile().store(outputfile_name, protein_identifications, peptide_identifications);
return EXECUTION_OK;
}
ExitCodes main_(int, const char**) override
{
//-------------------------------------------------------------
// 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.getColumnHeaders()[0].label = "light";
results_map.getColumnHeaders()[0].filename = in;
results_map.getColumnHeaders()[1].label = "heavy";
results_map.getColumnHeaders()[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
//-------------------------------------------------------------
StringList ms_runs;
exp.getPrimaryMSRunPath(ms_runs);
if (feature_out != "")
{
all_features.setPrimaryMSRunPath(ms_runs);
FeatureXMLFile().store(feature_out, all_features);
}
writeDebug_("Writing output", 1);
results_map.setPrimaryMSRunPath(ms_runs);
ConsensusXMLFile().store(out, results_map);
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**) override
{
//input file names
String in = getStringOption_("in");
String out = getStringOption_("out");
String out_mzq = getStringOption_("out_mzq");
//prevent loading of fragment spectra
PeakFileOptions options;
options.setMSLevels(vector<Int>(1, 1));
//reading input data
MzMLFile f;
f.getOptions() = options;
f.setLogType(log_type_);
PeakMap exp;
f.load(in, exp);
exp.updateRanges();
if (exp.getSpectra().empty())
{
throw OpenMS::Exception::FileEmpty(__FILE__, __LINE__, __FUNCTION__, "Error: No MS1 spectra in input file.");
}
// determine type of spectral data (profile or centroided)
SpectrumSettings::SpectrumType spectrum_type = exp[0].getType();
if (spectrum_type == SpectrumSettings::PROFILE)
{
if (!getFlag_("force"))
{
throw OpenMS::Exception::IllegalArgument(__FILE__, __LINE__, __FUNCTION__, "Error: Profile data provided but centroided spectra expected. To enforce processing of the data set the -force flag.");
}
}
//load seeds
FeatureMap seeds;
if (getStringOption_("seeds") != "")
{
FeatureXMLFile().load(getStringOption_("seeds"), seeds);
}
//setup of FeatureFinder
FeatureFinder ff;
ff.setLogType(log_type_);
// A map for the resulting features
FeatureMap features;
StringList ms_runs;
exp.getPrimaryMSRunPath(ms_runs);
features.setPrimaryMSRunPath(ms_runs);
// get parameters specific for the feature finder
Param feafi_param = getParam_().copy("algorithm:", true);
writeDebug_("Parameters passed to FeatureFinder", feafi_param, 3);
// Apply the feature finder
ff.run(FeatureFinderAlgorithmPicked::getProductName(), exp, features, feafi_param, seeds);
features.applyMemberFunction(&UniqueIdInterface::setUniqueId);
// DEBUG
if (debug_level_ > 10)
{
FeatureMap::Iterator it;
for (it = features.begin(); it != features.end(); ++it)
{
if (!it->isMetaEmpty())
{
vector<String> keys;
it->getKeys(keys);
LOG_INFO << "Feature " << it->getUniqueId() << endl;
for (Size i = 0; i < keys.size(); i++)
{
LOG_INFO << " " << keys[i] << " = " << it->getMetaValue(keys[i]) << endl;
}
}
}
}
//-------------------------------------------------------------
// writing files
//-------------------------------------------------------------
//annotate output with data processing info
addDataProcessing_(features, getProcessingInfo_(DataProcessing::QUANTITATION));
// write features to user specified output file
FeatureXMLFile map_file;
// Remove detailed convex hull information and subordinate features
// (unless requested otherwise) to reduce file size of feature files
// unless debugging is turned on.
if (debug_level_ < 5)
{
FeatureMap::Iterator it;
for (it = features.begin(); it != features.end(); ++it)
{
it->getConvexHull().expandToBoundingBox();
for (Size i = 0; i < it->getConvexHulls().size(); ++i)
{
it->getConvexHulls()[i].expandToBoundingBox();
}
it->getSubordinates().clear();
}
}
map_file.store(out, features);
if (!out_mzq.trim().empty())
{
std::vector<DataProcessing> tmp;
for (Size i = 0; i < exp[0].getDataProcessing().size(); i++)
{
tmp.push_back(*exp[0].getDataProcessing()[i].get());
}
MSQuantifications msq(features, exp.getExperimentalSettings(), tmp );
msq.assignUIDs();
MzQuantMLFile file;
file.store(out_mzq, msq);
}
return EXECUTION_OK;
}
