void ProtXMLFile::load(const String& filename, ProteinIdentification& protein_ids, PeptideIdentification& peptide_ids)
{
//Filename for error messages in XMLHandler
file_ = filename;
resetMembers_();
// reset incoming data
protein_ids = ProteinIdentification();
peptide_ids = PeptideIdentification();
// remember data link while parsing
prot_id_ = &protein_ids;
pep_id_ = &peptide_ids;
parse_(filename, this);
}
void MascotXMLFile::load(const String& filename,
ProteinIdentification& protein_identification,
vector<PeptideIdentification>& id_data,
map<String, vector<AASequence> >& peptides,
const SpectrumMetaDataLookup& lookup)
{
//clear
protein_identification = ProteinIdentification();
id_data.clear();
Internal::MascotXMLHandler handler(protein_identification, id_data,
filename, peptides, lookup);
parse_(filename, &handler);
// since the Mascot XML can contain "peptides" without sequences,
// the identifications without any real peptide hit are removed
vector<PeptideIdentification> filtered_hits;
filtered_hits.reserve(id_data.size());
Size missing_sequence = 0; // counter
for (vector<PeptideIdentification>::iterator id_it = id_data.begin();
id_it != id_data.end(); ++id_it)
{
const vector<PeptideHit>& peptide_hits = id_it->getHits();
if (!peptide_hits.empty() &&
(peptide_hits.size() > 1 || !peptide_hits[0].getSequence().empty()))
{
filtered_hits.push_back(*id_it);
}
else if (!id_it->empty()) ++missing_sequence;
}
if (missing_sequence)
{
LOG_WARN << "Warning: Removed " << missing_sequence
<< " peptide identifications without sequence." << endl;
}
id_data.swap(filtered_hits);
// check if we have (some) RT information:
Size no_rt_count = 0;
for (vector<PeptideIdentification>::iterator id_it = id_data.begin();
id_it != id_data.end(); ++id_it)
{
if (!id_it->hasRT()) ++no_rt_count;
}
if (no_rt_count)
{
LOG_WARN << "Warning: " << no_rt_count << " (of " << id_data.size()
<< ") peptide identifications have no retention time value."
<< endl;
}
// if we have a mapping, but couldn't find any RT values, that's an error:
if (!lookup.empty() && (no_rt_count == id_data.size()))
{
throw Exception::MissingInformation(
__FILE__, __LINE__, __PRETTY_FUNCTION__,
"No retention time information for peptide identifications found");
}
// argh! Mascot 2.2 tends to repeat the first hit (yes it appears twice),
// so we delete one of them
for (vector<PeptideIdentification>::iterator it = id_data.begin();
it != id_data.end(); ++it)
{
vector<PeptideHit> peptide_hits = it->getHits();
// check if equal, except for rank
if (peptide_hits.size() > 1 &&
peptide_hits[0].getScore() == peptide_hits[1].getScore() &&
peptide_hits[0].getSequence() == peptide_hits[1].getSequence() &&
peptide_hits[0].getCharge() == peptide_hits[1].getCharge())
{
// erase first hit
peptide_hits.erase(peptide_hits.begin() + 1);
it->setHits(peptide_hits);
}
}
}
void
PepNovoOutfile::load(
const std::string & result_filename,
vector<PeptideIdentification> & peptide_identifications,
ProteinIdentification & protein_identification,
const double & score_threshold,
const IndexPosMappingType & index_to_precursor,
const map<String, String> & pnovo_modkey_to_mod_id
)
{
// generally used variables
StringList substrings;
map<String, Int> columns;
PeptideHit peptide_hit;
String
line,
score_type = "PepNovo",
version = "unknown",
identifier,
filename,
sequence,
sequence_with_mods;
DateTime datetime = DateTime::now(); // there's no date given from PepNovo
protein_identification.setDateTime(datetime);
peptide_identifications.clear();
PeptideIdentification peptide_identification;
protein_identification = ProteinIdentification();
// open the result
ifstream result_file(result_filename.c_str());
if (!result_file)
{
throw Exception::FileNotFound(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, result_filename);
}
Size line_number(0); // used to report in which line an error occurred
Size id_count(0); // number of IDs seen (not necessarily the ones finally returned)
getSearchEngineAndVersion(result_filename, protein_identification);
//if information could not be retrieved from the outfile use defaults
if (protein_identification.getSearchEngineVersion().empty())
{
protein_identification.setSearchEngine("PepNovo");
protein_identification.setSearchEngineVersion(version);
}
identifier = protein_identification.getSearchEngine() + "_" + datetime.getDate();
protein_identification.setIdentifier(identifier);
map<String, String> mod_mask_map;
const vector<String> & mods = protein_identification.getSearchParameters().variable_modifications;
for (vector<String>::const_iterator mod_it = mods.begin(); mod_it != mods.end(); ++mod_it)
{
if (mod_it->empty())
continue;
//cout<<*mod_it<<endl;
if (pnovo_modkey_to_mod_id.find(*mod_it) != pnovo_modkey_to_mod_id.end())
{
//cout<<keys_to_id.find(*mod_it)->second<<endl;
ResidueModification tmp_mod = ModificationsDB::getInstance()->getModification(pnovo_modkey_to_mod_id.find(*mod_it)->second);
if (mod_it->prefix(1) == "^" || mod_it->prefix(1) == "$")
{
mod_mask_map[*mod_it] = "(" + tmp_mod.getId() + ")";
}
else
{
mod_mask_map[*mod_it] = String(tmp_mod.getOrigin()) + "(" + tmp_mod.getId() + ")";
}
}
else
{
if (mod_it->prefix(1) != "^" && mod_it->prefix(1) != "$")
{
mod_mask_map[*mod_it] = mod_it->prefix(1) + "[" + mod_it->substr(1) + "]";
//cout<<mod_mask_map[*mod_it]<<endl;
}
else
{
mod_mask_map[*mod_it] = "[" + *mod_it + "]";
//cout<<mod_mask_map[*mod_it]<<endl;
}
}
}
Size index;
while (getline(result_file, line))
{
if (!line.empty() && (line[line.length() - 1] < 33)) line.resize(line.length() - 1); // remove weird EOL character
line.trim();
++line_number;
if (line.hasPrefix(">> ")) // >> 1 /home/shared/pepnovo/4611_raw_ms2_picked.mzXML.1001.2.dta
{
++id_count;
if (!peptide_identification.empty() && !peptide_identification.getHits().empty())
{
peptide_identifications.push_back(peptide_identification);
}
line.split(' ', substrings);
//String index = File::basename(line.substr(line.find(' ', strlen(">> ")) + 1));
if (substrings.size() < 3)
{
throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "Not enough columns (spectrum Id) in file in line " + String(line_number) + String(" (should be 2 or more)!"), result_filename);
}
try
{
index = substrings[2].trim().toInt();
}
catch (...)
{
throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "Expected an index number in line " + String(line_number) + String(" at position 2 (line was: '" + line + "')!"), result_filename);
}
//cout<<"INDEX: "<<index<<endl;
peptide_identification = PeptideIdentification();
bool success = false;
if (index_to_precursor.size()>0)
{
if (index_to_precursor.find(index) != index_to_precursor.end())
{
peptide_identification.setRT(index_to_precursor.find(index)->second.first);
peptide_identification.setMZ(index_to_precursor.find(index)->second.second);
success = true;
}
else throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "Index '" + String(index) + String("' in line '" + line + "' not found in index table (line was: '" + line + "')!"), result_filename);
}
if (!success)
{ // try to reconstruct from title entry (usually sensible when MGF is supplied to PepNovo)
try
{
if (substrings.size() >= 4)
{
StringList parts = ListUtils::create<String>(substrings[3], '_');
if (parts.size() >= 2)
{
peptide_identification.setRT(parts[1].toDouble());
peptide_identification.setMZ(parts[0].toDouble());
success = true;
}
}
}
catch (...)
{
}
if (!success) throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "Precursor could not be reconstructed from title '" + substrings[3] + String("' in line '" + line + "' (line was: '" + line + "')!"), result_filename);
}
peptide_identification.setSignificanceThreshold(score_threshold);
peptide_identification.setScoreType(score_type);
peptide_identification.setIdentifier(identifier);
}
else if (line.hasPrefix("#Index")) // #Index Prob Score N-mass C-Mass [M+H] Charge Sequence
{
if (columns.empty()) // map the column names to their column number
{
line.split('\t', substrings);
for (vector<String>::const_iterator s_i = substrings.begin(); s_i != substrings.end(); ++s_i)
{
if ((*s_i) == "#Index")
columns["Index"] = s_i - substrings.begin();
else if ((*s_i) == "RnkScr")
columns["RnkScr"] = s_i - substrings.begin();
else if ((*s_i) == "PnvScr")
columns["PnvScr"] = s_i - substrings.begin();
else if ((*s_i) == "N-Gap")
columns["N-Gap"] = s_i - substrings.begin();
else if ((*s_i) == "C-Gap")
columns["C-Gap"] = s_i - substrings.begin();
else if ((*s_i) == "[M+H]")
columns["[M+H]"] = s_i - substrings.begin();
else if ((*s_i) == "Charge")
columns["Charge"] = s_i - substrings.begin();
else if ((*s_i) == "Sequence")
columns["Sequence"] = s_i - substrings.begin();
}
if (columns.size() != 8)
{
result_file.close();
result_file.clear();
throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "Not enough columns in file in line " + String(line_number) + String(" (should be 8)!"), result_filename);
}
}
while (getline(result_file, line))
{
++line_number;
if (!line.empty() && (line[line.length() - 1] < 33))
line.resize(line.length() - 1);
line.trim();
if (line.empty())
break;
line.split('\t', substrings);
if (!substrings.empty())
{
if (substrings.size() != 8)
{
result_file.close();
result_file.clear();
throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, "Not enough columns in file in line " + String(line_number) + String(" (should be 8)!"), result_filename);
}
if (substrings[columns["RnkScr"]].toFloat() >= score_threshold)
{
peptide_hit = PeptideHit();
peptide_hit.setCharge(substrings[columns["Charge"]].toInt());
peptide_hit.setRank(substrings[columns["Index"]].toInt() + 1);
peptide_hit.setScore(substrings[columns["RnkScr"]].toFloat());
peptide_hit.setMetaValue("PnvScr", substrings[columns["PnvScr"]].toFloat());
peptide_hit.setMetaValue("N-Gap", substrings[columns["N-Gap"]].toFloat());
peptide_hit.setMetaValue("C-Gap", substrings[columns["C-Gap"]].toFloat());
peptide_hit.setMetaValue("MZ", substrings[columns["[M+H]"]].toFloat());
sequence = substrings[columns["Sequence"]];
for (map<String, String>::iterator mask_it = mod_mask_map.begin(); mask_it != mod_mask_map.end(); ++mask_it)
{
if (mask_it->first.hasPrefix("^") && sequence.hasSubstring(mask_it->first))
{
sequence.substitute(mask_it->first, "");
sequence = mask_it->second + sequence;
}
//cout<<mask_it->first<<" "<<mask_it->second<<endl;
sequence.substitute(mask_it->first, mask_it->second);
}
peptide_hit.setSequence(AASequence::fromString(sequence));
peptide_identification.insertHit(peptide_hit);
}
}
}
}
}
if (!peptide_identifications.empty() || !peptide_identification.getHits().empty())
{
peptide_identifications.push_back(peptide_identification);
}
result_file.close();
result_file.clear();
LOG_INFO << "Parsed " << id_count << " ids, retained " << peptide_identifications.size() << "." << std::endl;
}
