bool EnzymaticDigestionLogModel::isCleavageSite_(
const AASequence& protein, const AASequence::ConstIterator& iterator) const
{
if (enzyme_.getName() != "Trypsin") // no cleavage
{
throw Exception::InvalidParameter(__FILE__, __LINE__, __PRETTY_FUNCTION__, String("EnzymaticDigestionLogModel: enzyme '") + enzyme_.getName() + " does not support logModel!");
}
else
{
if ((!enzyme_.getRegEx().hasSubstring(iterator->getOneLetterCode())) || *iterator == 'P') // wait for R or K
{
return false;
}
SignedSize pos = distance(AASequence::ConstIterator(protein.begin()),
iterator) - 4; // start position in sequence
double score_cleave = 0, score_missed = 0;
for (SignedSize i = 0; i < 9; ++i)
{
if ((pos + i >= 0) && (pos + i < (SignedSize)protein.size()))
{
BindingSite_ bs(i, protein[pos + i].getOneLetterCode());
Map<BindingSite_, CleavageModel_>::const_iterator pos_it =
model_data_.find(bs);
if (pos_it != model_data_.end()) // no data for non-std. amino acids
{
score_cleave += pos_it->second.p_cleave;
score_missed += pos_it->second.p_miss;
}
}
}
return score_missed - score_cleave > log_model_threshold_;
}
}
String CompNovoIdentificationBase::getModifiedStringFromAASequence_(const AASequence & sequence)
{
String seq;
for (AASequence::ConstIterator it = sequence.begin(); it != sequence.end(); ++it)
{
if (residue_to_name_.has(&*it))
{
seq += residue_to_name_[&*it];
}
else
{
seq += it->getOneLetterCode();
}
}
return seq;
}
String SILACLabeler::getUnmodifiedSequence_(const Feature& feature, const String& arginine_label, const String& lysine_label) const
{
String unmodified_sequence = "";
for (AASequence::ConstIterator residue = feature.getPeptideIdentifications()[0].getHits()[0].getSequence().begin();
residue != feature.getPeptideIdentifications()[0].getHits()[0].getSequence().end();
++residue)
{
if (*residue == 'R' && residue->getModification() == arginine_label)
{
unmodified_sequence.append("R");
}
else if (*residue == 'K' && residue->getModification() == lysine_label)
{
unmodified_sequence.append("K");
}
else
{
unmodified_sequence.append(residue->getOneLetterCode());
}
}
return unmodified_sequence;
}
void TheoreticalSpectrumGenerator::addLosses_(RichPeakSpectrum & spectrum, const AASequence & ion, double intensity, Residue::ResidueType res_type, int charge) const
{
RichPeak1D p;
set<String> losses;
for (AASequence::ConstIterator it = ion.begin(); it != ion.end(); ++it)
{
if (it->hasNeutralLoss())
{
vector<EmpiricalFormula> loss_formulas = it->getLossFormulas();
for (Size i = 0; i != loss_formulas.size(); ++i)
{
losses.insert(loss_formulas[i].toString());
}
}
}
if (!add_isotopes_)
{
p.setIntensity(intensity * rel_loss_intensity_);
}
for (set<String>::const_iterator it = losses.begin(); it != losses.end(); ++it)
{
EmpiricalFormula loss_ion = ion.getFormula(res_type, charge) - EmpiricalFormula(*it);
// thanks to Chris and Sandro
// check for negative element frequencies (might happen if losses are not allowed for specific ions)
bool negative_elements(false);
for (EmpiricalFormula::ConstIterator eit = loss_ion.begin(); eit != loss_ion.end(); ++eit)
{
if (eit->second < 0)
{
negative_elements = true;
break;
}
}
if (negative_elements)
{
continue;
}
double loss_pos = loss_ion.getMonoWeight() / (double)charge;
const String& loss_name = *it;
if (add_isotopes_)
{
IsotopeDistribution dist = loss_ion.getIsotopeDistribution(max_isotope_);
UInt j(0);
for (IsotopeDistribution::ConstIterator iso = dist.begin(); iso != dist.end(); ++iso)
{
p.setMZ((double)(loss_pos + j) / (double)charge);
p.setIntensity(intensity * rel_loss_intensity_ * iso->second);
if (add_metainfo_ && j == 0)
{
// note: important to construct a string from char. If omitted it will perform pointer arithmetics on the "-" string literal
String ion_name = String(residueTypeToIonLetter_(res_type)) + String(ion.size()) + "-" + loss_name + String(charge, '+');
p.setMetaValue("IonName", ion_name);
}
spectrum.push_back(p);
}
}
else
{
p.setMZ(loss_pos);
if (add_metainfo_)
{
// note: important to construct a string from char. If omitted it will perform pointer arithmetics on the "-" string literal
String ion_name = String(residueTypeToIonLetter_(res_type)) + String(ion.size()) + "-" + loss_name + String(charge, '+');
p.setMetaValue("IonName", ion_name);
}
spectrum.push_back(p);
}
}
}
void TheoreticalSpectrumGenerator::addPeaks(RichPeakSpectrum & spectrum, const AASequence & peptide, Residue::ResidueType res_type, Int charge)
{
if (peptide.empty())
{
return;
}
Map<DoubleReal, AASequence> ions;
Map<DoubleReal, String> names;
AASequence ion;
DoubleReal intensity(0);
bool add_first_prefix_ion(param_.getValue("add_first_prefix_ion").toBool());
// generate the ion peaks
switch (res_type)
{
case Residue::AIon:
{
Size i = 1;
if (!add_first_prefix_ion)
{
i = 2;
}
for (; i < peptide.size(); ++i)
{
ion = peptide.getPrefix(i);
DoubleReal pos = ion.getMonoWeight(Residue::AIon, charge) / (DoubleReal)charge;
ions[pos] = ion;
names[pos] = "a" + String(i) + String(charge, '+');
}
intensity = (DoubleReal)param_.getValue("a_intensity");
break;
}
case Residue::BIon:
{
Size i = 1;
if (!add_first_prefix_ion)
{
i = 2;
}
for (; i < peptide.size(); ++i)
{
ion = peptide.getPrefix(i);
DoubleReal pos = ion.getMonoWeight(Residue::BIon, charge) / (DoubleReal)charge;
ions[pos] = ion;
names[pos] = "b" + String(i) + String(charge, '+');
}
intensity = (DoubleReal)param_.getValue("b_intensity");
break;
}
case Residue::CIon:
{
Size i = 1;
if (!add_first_prefix_ion)
{
i = 2;
}
for (; i < peptide.size(); ++i)
{
ion = peptide.getPrefix(i);
DoubleReal pos = ion.getMonoWeight(Residue::CIon, charge) / (DoubleReal)charge;
ions[pos] = ion;
names[pos] = "c" + String(i) + String(charge, '+');
}
intensity = (DoubleReal)param_.getValue("c_intensity");
break;
}
case Residue::XIon:
{
for (Size i = 1; i < peptide.size(); ++i)
{
ion = peptide.getSuffix(i);
DoubleReal pos = ion.getMonoWeight(Residue::XIon, charge) / (DoubleReal)charge;
ions[pos] = ion;
names[pos] = "x" + String(i) + String(charge, '+');
}
intensity = (DoubleReal)param_.getValue("x_intensity");
break;
}
case Residue::YIon:
{
for (Size i = 1; i < peptide.size(); ++i)
{
ion = peptide.getSuffix(i);
DoubleReal pos = ion.getMonoWeight(Residue::YIon, charge) / (DoubleReal)charge;
ions[pos] = ion;
names[pos] = "y" + String(i) + String(charge, '+');
}
intensity = (DoubleReal)param_.getValue("y_intensity");
break;
}
case Residue::ZIon:
{
for (Size i = 1; i < peptide.size(); ++i)
{
ion = peptide.getSuffix(i);
DoubleReal pos = ion.getMonoWeight(Residue::ZIon, charge) / (DoubleReal)charge;
ions[pos] = ion;
names[pos] = "z" + String(i) + String(charge, '+');
}
intensity = (DoubleReal)param_.getValue("z_intensity");
break;
}
default:
cerr << "Cannot create peaks of that ion type" << endl;
}
// get the params
bool add_losses(param_.getValue("add_losses").toBool());
bool add_metainfo(param_.getValue("add_metainfo").toBool());
bool add_isotopes(param_.getValue("add_isotopes").toBool());
Int max_isotope((Int)param_.getValue("max_isotope"));
DoubleReal rel_loss_intensity((DoubleReal)param_.getValue("relative_loss_intensity"));
for (Map<DoubleReal, AASequence>::ConstIterator cit = ions.begin(); cit != ions.end(); ++cit)
{
ion = cit->second;
DoubleReal pos = cit->first;
String ion_name = names[pos];
if (add_isotopes)
{
IsotopeDistribution dist = ion.getFormula(res_type, charge).getIsotopeDistribution(max_isotope);
UInt j(0);
for (IsotopeDistribution::ConstIterator it = dist.begin(); it != dist.end(); ++it, ++j)
{
p_.setMZ((DoubleReal)(pos + (DoubleReal)j * Constants::NEUTRON_MASS_U) / (DoubleReal)charge);
p_.setIntensity(intensity * it->second);
if (add_metainfo && j == 0)
{
p_.setMetaValue("IonName", ion_name);
}
spectrum.push_back(p_);
}
}
else
{
p_.setMZ(pos);
p_.setIntensity(intensity);
if (add_metainfo)
{
p_.setMetaValue("IonName", ion_name);
}
spectrum.push_back(p_);
}
if (add_losses)
{
set<String> losses;
for (AASequence::ConstIterator it = cit->second.begin(); it != cit->second.end(); ++it)
{
if (it->hasNeutralLoss())
{
vector<EmpiricalFormula> loss_formulas = it->getLossFormulas();
for (Size i = 0; i != loss_formulas.size(); ++i)
{
losses.insert(loss_formulas[i].toString());
}
}
}
if (!add_isotopes)
{
p_.setIntensity(intensity * rel_loss_intensity);
}
for (set<String>::const_iterator it = losses.begin(); it != losses.end(); ++it)
{
EmpiricalFormula loss_ion = ion.getFormula(res_type, charge) - EmpiricalFormula(*it);
// thanks to Chris and Sandro
// check for negative element frequencies (might happen if losses are not allowed for specific ions)
bool negative_elements(false);
for (EmpiricalFormula::ConstIterator eit = loss_ion.begin(); eit != loss_ion.end(); ++eit)
{
if (eit->second < 0)
{
negative_elements = true;
break;
}
}
if (negative_elements)
{
continue;
}
DoubleReal loss_pos = loss_ion.getMonoWeight() / (DoubleReal)charge;
String loss_name = *it;
if (add_isotopes)
{
IsotopeDistribution dist = loss_ion.getIsotopeDistribution(max_isotope);
UInt j(0);
for (IsotopeDistribution::ConstIterator iso = dist.begin(); iso != dist.end(); ++iso)
{
p_.setMZ((DoubleReal)(loss_pos + j) / (DoubleReal)charge);
p_.setIntensity(intensity * rel_loss_intensity * iso->second);
if (add_metainfo && j == 0)
{
p_.setMetaValue("IonName", ion_name + "-" + loss_name);
}
spectrum.push_back(p_);
}
}
else
{
p_.setMZ(loss_pos);
if (add_metainfo)
{
p_.setMetaValue("IonName", ion_name + "-" + loss_name);
}
spectrum.push_back(p_);
}
}
}
}
if (add_metainfo)
{
p_.setMetaValue("IonName", String(""));
}
spectrum.sortByPosition();
return;
}
bool ModificationDefinitionsSet::isCompatible(const AASequence & peptide) const
{
set<String> var_names(getVariableModificationNames()), fixed_names(getFixedModificationNames());
// no modifications present and needed
if (fixed_names.empty() && !peptide.isModified())
{
return true;
}
// check whether the fixed modifications are fulfilled
if (!fixed_names.empty())
{
for (set<String>::const_iterator it1 = fixed_names.begin(); it1 != fixed_names.end(); ++it1)
{
String origin = ModificationsDB::getInstance()->getModification(*it1).getOrigin();
// only single 1lc amino acids are allowed
if (origin.size() != 1)
{
continue;
}
for (AASequence::ConstIterator it2 = peptide.begin(); it2 != peptide.end(); ++it2)
{
if (origin == it2->getOneLetterCode())
{
// check whether the residue is modified (has to be)
if (!it2->isModified())
{
return false;
}
// check whether the modification is the same
if (ModificationsDB::getInstance()->getModification(*it1).getId() != it2->getModification())
{
return false;
}
}
}
}
}
// check wether other modifications than the variable are present
for (AASequence::ConstIterator it = peptide.begin(); it != peptide.end(); ++it)
{
if (it->isModified())
{
String mod = ModificationsDB::getInstance()->getModification(it->getOneLetterCode(), it->getModification(), ResidueModification::ANYWHERE).getFullId();
if (var_names.find(mod) == var_names.end() &&
fixed_names.find(mod) == fixed_names.end())
{
return false;
}
}
}
if (peptide.hasNTerminalModification())
{
String mod = ModificationsDB::getInstance()->getTerminalModification(peptide.getNTerminalModification(), ResidueModification::N_TERM).getFullId();
if (var_names.find(mod) == var_names.end() &&
fixed_names.find(mod) == fixed_names.end())
{
return false;
}
}
if (peptide.hasCTerminalModification())
{
String mod = ModificationsDB::getInstance()->getTerminalModification(peptide.getCTerminalModification(), ResidueModification::C_TERM).getFullId();
if (var_names.find(mod) == var_names.end() &&
fixed_names.find(mod) == fixed_names.end())
{
return false;
}
}
return true;
}