//Visualizing ProteinIdentification object
void MetaDataBrowser::visualize_(ProteinIdentification & meta, QTreeWidgetItem * parent)
{
ProteinIdentificationVisualizer * visualizer = new ProteinIdentificationVisualizer(isEditable(), this, this);
QStringList labels;
int id = ws_->addWidget(visualizer);
labels << QString("ProteinIdentification %1").arg(meta.getSearchEngine().c_str()) << QString::number(id);
visualizer->load(meta, id);
QTreeWidgetItem * item;
if (parent == nullptr)
{
item = new QTreeWidgetItem(treeview_, labels);
}
else
{
item = new QTreeWidgetItem(parent, labels);
}
//check for proteinhits objects
meta.assignRanks();
for (Size i = 0; i < meta.getHits().size(); ++i)
{
visualize_(const_cast<ProteinHit &>(meta.getHits()[i]), item);
}
visualize_(dynamic_cast<MetaInfoInterface &>(meta), item);
connectVisualizer_(visualizer);
}
void IDFilter::filterIdentificationsByProteins(const ProteinIdentification& identification,
const vector<FASTAFile::FASTAEntry>& proteins,
ProteinIdentification& filtered_identification)
{
String protein_sequences;
String accession_sequences;
vector<ProteinHit> filtered_protein_hits;
ProteinHit temp_protein_hit;
filtered_identification = identification;
filtered_identification.setHits(vector<ProteinHit>());
for (Size i = 0; i < proteins.size(); i++)
{
accession_sequences.append("*" + proteins[i].identifier);
}
accession_sequences.append("*");
for (Size i = 0; i < identification.getHits().size(); i++)
{
if (accession_sequences.find("*" + identification.getHits()[i].getAccession()) != String::npos)
{
filtered_protein_hits.push_back(identification.getHits()[i]);
}
}
filtered_identification.setHits(filtered_protein_hits);
filtered_identification.assignRanks();
}
void IDFilter::removeUnreferencedProteinHits(const ProteinIdentification& identification, const vector<PeptideIdentification> peptide_identifications, ProteinIdentification& filtered_identification)
{
const String& run_identifier = identification.getIdentifier();
// build set of protein accessions that are referenced by peptides
set<String> proteinaccessions_with_peptides;
for (Size i = 0; i != peptide_identifications.size(); ++i)
{
// run id of protein and peptide identification must match
if (run_identifier == peptide_identifications[i].getIdentifier())
{
const vector<PeptideHit>& tmp_pep_hits = peptide_identifications[i].getHits();
// extract protein accessions of each peptide hit
for (Size j = 0; j != tmp_pep_hits.size(); ++j)
{
const std::vector<String>& protein_accessions = tmp_pep_hits[j].getProteinAccessions();
for (Size k = 0; k != protein_accessions.size(); ++k)
{
String key = protein_accessions[k];
proteinaccessions_with_peptides.insert(key);
}
}
}
}
// add all protein hits referenced by a peptide
const vector<ProteinHit>& temp_protein_hits = identification.getHits();
vector<ProteinHit> filtered_protein_hits;
for (Size j = 0; j != temp_protein_hits.size(); ++j)
{
const String& protein_accession = temp_protein_hits[j].getAccession();
if (proteinaccessions_with_peptides.find(protein_accession) != proteinaccessions_with_peptides.end())
{
filtered_protein_hits.push_back(temp_protein_hits[j]);
}
}
// copy identification
filtered_identification = identification;
// assign filtered hits to protein identification
filtered_identification.setHits(filtered_protein_hits);
}
PeptideHit peptide_hit;
START_SECTION((XTandemXMLFile()))
ptr = new XTandemXMLFile();
TEST_NOT_EQUAL(ptr, nullPointer)
END_SECTION
START_SECTION(~XTandemXMLFile())
delete ptr;
END_SECTION
ptr = new XTandemXMLFile();
START_SECTION(void setModificationDefinitionsSet(const ModificationDefinitionsSet &rhs))
ModificationDefinitionsSet mod_set(ListUtils::create<String>(""), ListUtils::create<String>("Carbamidomethyl (C),Oxidation (M),Carboxymethyl (C)"));
ptr->setModificationDefinitionsSet(mod_set);
NOT_TESTABLE
END_SECTION
START_SECTION(void load(const String& filename, ProteinIdentification& protein_identification, std::vector<PeptideIdentification>& id_data))
ptr->load(OPENMS_GET_TEST_DATA_PATH("XTandemXMLFile_test.xml"), protein_identification, peptide_identifications);
TEST_EQUAL(peptide_identifications.size(), 303)
TEST_EQUAL(protein_identification.getHits().size(), 497)
ptr->load(OPENMS_GET_TEST_DATA_PATH("XTandemXMLFile_test_2.xml"), protein_identification, peptide_identifications);
END_SECTION
/////////////////////////////////////////////////////////////
/////////////////////////////////////////////////////////////
END_TEST
END_SECTION
START_SECTION(void load(const String &filename, ProteinIdentification &protein_ids, PeptideIdentification &peptide_ids))
{
ProtXMLFile f;
ProteinIdentification proteins;
PeptideIdentification peptides;
String prot_file;
StringList ids = ListUtils::create<String>("16627578304933075941,13229490167902618598");
// we do this twice, just to check that members are correctly reset etc..
for (Int i=0;i<2;++i)
{
prot_file = OPENMS_GET_TEST_DATA_PATH("ProtXMLFile_input_1.protXML");
f.load(prot_file, proteins, peptides);
TEST_EQUAL(proteins.getIdentifier(), ids[i]);
TEST_EQUAL(peptides.getIdentifier(), ids[i]);
// groups
TEST_EQUAL(proteins.getProteinGroups().size(), 7);
TEST_EQUAL(proteins.getProteinGroups()[0].probability, 0.9990);
TEST_EQUAL(proteins.getProteinGroups()[0].accessions.size(), 1);
TEST_EQUAL(proteins.getProteinGroups()[3].accessions.size(), 2);
TEST_EQUAL(proteins.getProteinGroups()[3].accessions[0],
"P01876|IGHA1_HUMAN");
TEST_EQUAL(proteins.getProteinGroups()[3].accessions[1],
"P01877|IGHA2_HUMAN");
TEST_EQUAL(proteins.getProteinGroups()[6].probability, 0.2026);
TEST_EQUAL(proteins.getProteinGroups()[6].accessions.size(), 1);
TEST_EQUAL(proteins.getIndistinguishableProteins().size(), 7);
TEST_EQUAL(proteins.getIndistinguishableProteins()[0].accessions.size(), 1);
TEST_EQUAL(proteins.getIndistinguishableProteins()[3].accessions.size(), 2);
TEST_EQUAL(proteins.getIndistinguishableProteins()[3].accessions[0],
"P01876|IGHA1_HUMAN");
TEST_EQUAL(proteins.getIndistinguishableProteins()[3].accessions[1],
"P01877|IGHA2_HUMAN");
TEST_EQUAL(proteins.getIndistinguishableProteins()[6].accessions.size(), 1);
// proteins
TEST_EQUAL(proteins.getHits().size(), 9);
TEST_EQUAL(proteins.getHits()[0].getAccession(), "P02787|TRFE_HUMAN");
TEST_EQUAL(proteins.getHits()[0].getCoverage(), 8.6);
TEST_EQUAL(proteins.getHits()[0].getScore(), 0.9990);
// this one is indistinguishable... therefore it should have minimal infos
TEST_EQUAL(proteins.getHits()[6].getAccession(), "P00739|HPTR_HUMAN");
TEST_EQUAL(proteins.getHits()[6].getCoverage(), -1);
TEST_EQUAL(proteins.getHits()[6].getScore(), -1);
TEST_EQUAL(proteins.getHits()[8].getAccession(), "P04217|A1BG_HUMAN");
TEST_EQUAL(proteins.getHits()[8].getCoverage(), 2.0);
TEST_EQUAL(proteins.getHits()[8].getScore(), 0.2026);
// peptides
TEST_EQUAL(peptides.getHits().size(), 16);
AASequence aa_seq("MYLGYEYVTAIR");
TEST_EQUAL(peptides.getHits()[0].getSequence(), aa_seq);
TEST_EQUAL(peptides.getHits()[0].getCharge(), 2);
TEST_EQUAL(peptides.getHits()[0].getScore(), 0.8633);
TEST_EQUAL(peptides.getHits()[0].getProteinAccessions().size(), 1);
TEST_EQUAL(peptides.getHits()[0].getProteinAccessions()[0], "P02787|TRFE_HUMAN");
TEST_EQUAL(peptides.getHits()[0].getMetaValue("is_unique"), true);
TEST_EQUAL(peptides.getHits()[0].getMetaValue("is_contributing"), true);
// load 2 nd file and
prot_file = OPENMS_GET_TEST_DATA_PATH("ProtXMLFile_input_2.protXML");
}
}
START_SECTION((IDFilter()))
ptr = new IDFilter();
TEST_NOT_EQUAL(ptr, nullPointer);
END_SECTION
START_SECTION((~IDFilter()))
delete ptr;
END_SECTION
START_SECTION((void filterIdentificationsByProteins(const ProteinIdentification& identification, const vector<FASTAFile::FASTAEntry>& proteins, ProteinIdentification& filtered_identification)))
ProteinIdentification protein_identification2;
IDFilter::filterIdentificationsByProteins(protein_identification, proteins, protein_identification2);
TEST_EQUAL(protein_identification2.getScoreType(), "Mascot")
TEST_EQUAL(protein_identification2.getHits().size(), 2)
TEST_EQUAL(protein_identification2.getHits()[0].getAccession(), "Q824A5")
TEST_EQUAL(protein_identification2.getHits()[1].getAccession(), "Q872T5")
END_SECTION
START_SECTION((void filterIdentificationsByProteins(const PeptideIdentification& identification, const vector<FASTAFile::FASTAEntry>& proteins, PeptideIdentification& filtered_identification, bool no_protein_identifiers = false)))
PeptideIdentification identification2;
IDFilter::filterIdentificationsByProteins(identification, proteins, identification2);
TEST_EQUAL(identification2.getScoreType(), "Mascot")
TEST_EQUAL(identification2.getHits().size(), 2)
TEST_EQUAL(identification2.getHits()[0].getSequence(), AASequence::fromString("LHASGITVTEIPVTATNFK"))
TEST_EQUAL(identification2.getHits()[1].getSequence(), AASequence::fromString("MRSLGYVAVISAVATDTDK"))
END_SECTION
void IDFilter::removeUnreferencedPeptideHits(const ProteinIdentification& identification,
vector<PeptideIdentification>& peptide_identifications,
bool delete_unreferenced_peptide_hits /* = false */)
{
const String& run_identifier = identification.getIdentifier();
// build set of protein accessions
set<String> all_prots;
const vector<ProteinHit>& temp_protein_hits = identification.getHits();
for (Size j = 0; j != temp_protein_hits.size(); ++j)
{
all_prots.insert(temp_protein_hits[j].getAccession());
}
vector<PeptideIdentification> filtered_peptide_identifications;
// remove peptides which are not referenced
for (Size i = 0; i != peptide_identifications.size(); ++i)
{
// run id of protein and peptide identification must match
if (run_identifier == peptide_identifications[i].getIdentifier())
{
const vector<PeptideHit>& tmp_pep_hits = peptide_identifications[i].getHits();
vector<PeptideHit> filtered_pep_hits;
// check protein accessions of each peptide hit
for (Size j = 0; j != tmp_pep_hits.size(); ++j)
{
vector<PeptideEvidence> hit_peptide_evidences = tmp_pep_hits[j].getPeptideEvidences();
vector<PeptideEvidence> valid_peptide_evidence;
for (vector<PeptideEvidence>::const_iterator pe_it = hit_peptide_evidences.begin(); pe_it != hit_peptide_evidences.end(); ++pe_it)
{
// find valid proteins
if (all_prots.find(pe_it->getProteinAccession()) != all_prots.end())
{
valid_peptide_evidence.push_back(*pe_it);
}
}
if (!valid_peptide_evidence.empty() || !delete_unreferenced_peptide_hits)
{
// if present, copy the hit
filtered_pep_hits.push_back(tmp_pep_hits[j]);
filtered_pep_hits.back().setPeptideEvidences(valid_peptide_evidence);
}
}
// if the peptide has hits, we use it
if (!filtered_pep_hits.empty())
{
filtered_peptide_identifications.push_back(peptide_identifications[i]);
filtered_peptide_identifications.back().setHits(filtered_pep_hits);
}
}
else // peptide is from another run, let it pass the filter
{
filtered_peptide_identifications.push_back(peptide_identifications[i]);
}
}
// exchange with new hits
filtered_peptide_identifications.swap(peptide_identifications);
}
void IBSpectraFile::store(const String& filename, const ConsensusMap& cm)
{
// typdefs for shorter code
typedef std::vector<ProteinHit>::iterator ProtHitIt;
// general settings .. do we need to expose these?
// ----------------------------------------------------------------------
/// Allow also non-unique peptides to be exported
bool allow_non_unique = true;
/// Intensities below this value will be set to 0.0 to avoid numerical problems when quantifying
double intensity_threshold = 0.00001;
// ----------------------------------------------------------------------
// guess experiment type
boost::shared_ptr<IsobaricQuantitationMethod> quantMethod = guessExperimentType_(cm);
// we need the protein identifications to reference the protein names
ProteinIdentification protIdent;
bool has_proteinIdentifications = false;
if (cm.getProteinIdentifications().size() > 0)
{
protIdent = cm.getProteinIdentifications()[0];
has_proteinIdentifications = true;
}
// start the file by adding the tsv header
TextFile textFile;
textFile.addLine(ListUtils::concatenate(constructHeader_(*quantMethod), "\t"));
for (ConsensusMap::ConstIterator cm_iter = cm.begin();
cm_iter != cm.end();
++cm_iter)
{
const ConsensusFeature& cFeature = *cm_iter;
std::vector<IdCSV> entries;
/// 1st we extract the identification information from the consensus feature
if (cFeature.getPeptideIdentifications().size() == 0 || !has_proteinIdentifications)
{
// we store unidentified hits anyway, because the iTRAQ quant is still helpful for normalization
entries.push_back(IdCSV());
}
else
{
// protein name:
const PeptideHit& peptide_hit = cFeature.getPeptideIdentifications()[0].getHits()[0];
std::set<String> protein_accessions = peptide_hit.extractProteinAccessions();
if (protein_accessions.size() != 1)
{
if (!allow_non_unique) continue; // we only want unique peptides
}
for (std::set<String>::const_iterator prot_ac = protein_accessions.begin(); prot_ac != protein_accessions.end(); ++prot_ac)
{
IdCSV entry;
entry.charge = cFeature.getPeptideIdentifications()[0].getHits()[0].getCharge();
entry.peptide = cFeature.getPeptideIdentifications()[0].getHits()[0].getSequence().toUnmodifiedString();
entry.theo_mass = cFeature.getPeptideIdentifications()[0].getHits()[0].getSequence().getMonoWeight(Residue::Full, cFeature.getPeptideIdentifications()[0].getHits()[0].getCharge());
// write modif
entry.modif = getModifString_(cFeature.getPeptideIdentifications()[0].getHits()[0].getSequence());
ProtHitIt proteinHit = protIdent.findHit(*prot_ac);
if (proteinHit == protIdent.getHits().end())
{
std::cerr << "Protein referenced in peptide not found...\n";
continue; // protein not found
}
entry.accession = proteinHit->getAccession();
entries.push_back(entry);
}
}
// 2nd we add the quantitative information of the channels
// .. skip features with 0 intensity
if (cFeature.getIntensity() == 0)
{
continue;
}
for (std::vector<IdCSV>::iterator entry = entries.begin();
entry != entries.end();
++entry)
{
// set parent intensity
entry->parent_intens = cFeature.getIntensity();
entry->retention_time = cFeature.getRT();
entry->spectrum = cFeature.getUniqueId();
entry->exp_mass = cFeature.getMZ();
// create output line
StringList currentLine;
// add entry to currentLine
entry->toStringList(currentLine);
// extract channel intensities and positions
std::map<Int, double> intensityMap;
ConsensusFeature::HandleSetType features = cFeature.getFeatures();
for (ConsensusFeature::HandleSetType::const_iterator fIt = features.begin();
fIt != features.end();
++fIt)
{
intensityMap[Int(fIt->getMZ())] = (fIt->getIntensity() > intensity_threshold ? fIt->getIntensity() : 0.0);
}
for (IsobaricQuantitationMethod::IsobaricChannelList::const_iterator it = quantMethod->getChannelInformation().begin();
it != quantMethod->getChannelInformation().end();
++it)
{
currentLine.push_back(String(it->center));
}
for (IsobaricQuantitationMethod::IsobaricChannelList::const_iterator it = quantMethod->getChannelInformation().begin();
it != quantMethod->getChannelInformation().end();
++it)
{
currentLine.push_back(String(intensityMap[int(it->center)]));
}
textFile.addLine(ListUtils::concatenate(currentLine, "\t"));
}
}
// write to file
textFile.store(filename);
}