identificationgroup.cpp

/*******************************************************************************
 * Copyright (c) 2017 Olivier Langella <olivier.langella@u-psud.fr>.
 *
 * This file is part of XTPcpp.
 *
 *     XTPcpp is free software: you can redistribute it and/or modify
 *     it under the terms of the GNU General Public License as published by
 *     the Free Software Foundation, either version 3 of the License, or
 *     (at your option) any later version.
 *
 *     XTPcpp is distributed in the hope that it will be useful,
 *     but WITHOUT ANY WARRANTY; without even the implied warranty of
 *     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *     GNU General Public License for more details.
 *
 *     You should have received a copy of the GNU General Public License
 *     along with XTPcpp.  If not, see <http://www.gnu.org/licenses/>.
 *
 * Contributors:
 *     Olivier Langella <olivier.langella@u-psud.fr> - initial API and
 *implementation
 ******************************************************************************/

#include "identificationgroup.h"
#include "project.h"
#include "../utils/groupstore.h"
#include <pappsomspp/pappsoexception.h>


IdentificationGroup::IdentificationGroup(Project *project)
{
  _p_project = project;
}

IdentificationGroup::~IdentificationGroup()
{
  auto it = _protein_match_list.begin();
  while(it != _protein_match_list.end())
    {
      delete(*it);
      it++;
    }
}

const PtmGroupingExperiment *
IdentificationGroup::getPtmGroupingExperiment() const
{
  return _p_grp_ptm_experiment;
}
const GroupStore &
IdentificationGroup::getGroupStore() const
{
  return _group_store;
}

pappso::pappso_double
IdentificationGroup::computeProtoNsafSum(const MsRun *p_msrun_id,
                                         const Label *p_label) const
{
  pappso::pappso_double nsaf_sum = 0;
  for(auto &p_protein_match : _protein_match_list)
    {
      if(p_protein_match->getValidationState() >= ValidationState::grouped)
        {
          nsaf_sum += p_protein_match->getProtoNsaf(p_msrun_id, p_label);
        }
    }
  return nsaf_sum;
}

unsigned int
IdentificationGroup::countPeptideMass(ValidationState state) const
{

  std::vector<pappso::GrpPeptide *> count_peptide_mass;
  for(auto &p_protein_match : _protein_match_list)
    {
      if(p_protein_match->getValidationState() >= state)
        {
          p_protein_match->countPeptideMass(count_peptide_mass, state);
        }
    }
  std::sort(count_peptide_mass.begin(), count_peptide_mass.end());
  auto last = std::unique(count_peptide_mass.begin(), count_peptide_mass.end());
  return std::distance(count_peptide_mass.begin(), last);
}

unsigned int
IdentificationGroup::countPeptideMassSample(ValidationState state) const
{

  std::vector<std::size_t> count_peptide_mass_sample;
  for(auto &p_protein_match : _protein_match_list)
    {
      // if(p_protein_match->getValidationState() >= state)
      // {
      p_protein_match->countPeptideMassSample(count_peptide_mass_sample, state);
      // }
    }
  std::sort(count_peptide_mass_sample.begin(), count_peptide_mass_sample.end());
  auto last = std::unique(count_peptide_mass_sample.begin(),
                          count_peptide_mass_sample.end());
  return std::distance(count_peptide_mass_sample.begin(), last);
}

unsigned int
IdentificationGroup::countDecoyPeptideMassSample(ValidationState state) const
{
  std::vector<std::size_t> count_peptide_mass_sample;
  for(auto &p_protein_match : _protein_match_list)
    {
      if(p_protein_match->getProteinXtpSp().get()->isDecoy())
        {
          // if(p_protein_match->getValidationState() >= state)
          //  {
          p_protein_match->countPeptideMassSample(count_peptide_mass_sample,
                                                  state);
          //}
        }
    }
  std::sort(count_peptide_mass_sample.begin(), count_peptide_mass_sample.end());
  auto last = std::unique(count_peptide_mass_sample.begin(),
                          count_peptide_mass_sample.end());
  return std::distance(count_peptide_mass_sample.begin(), last);
}


unsigned int
IdentificationGroup::countDecoyProteinMatch(ValidationState state) const
{
  return std::count_if(
    _protein_match_list.begin(),
    _protein_match_list.end(),
    [state](const ProteinMatch *p_protein_match) {
      if((p_protein_match->getProteinXtpSp().get()->isDecoy()) &&
         (p_protein_match->getValidationState() >= state))
        {
          return true;
        }
      else
        {
          return false;
        }
    });
}

unsigned int
IdentificationGroup::countPeptideEvidence(ValidationState state) const
{
  std::set<const PeptideEvidence *> peptide_evidence_set;
  for(auto &p_protein_match : _protein_match_list)
    {
      // if(p_protein_match->getValidationState() >= state)
      //  {
      p_protein_match->collectPeptideEvidences(peptide_evidence_set, state);
      //  }
    }
  return peptide_evidence_set.size();
}

unsigned int
IdentificationGroup::countDecoyPeptideEvidence(ValidationState state) const
{
  std::set<const PeptideEvidence *> peptide_evidence_set;
  for(auto &p_protein_match : _protein_match_list)
    {
      // if(p_protein_match->getValidationState() >= state)
      //  {
      if(p_protein_match->getProteinXtpSp().get()->isDecoy())
        {
          p_protein_match->collectPeptideEvidences(peptide_evidence_set, state);
        }
      //  }
    }
  return peptide_evidence_set.size();
}

unsigned int
IdentificationGroup::countPeptideMatch(ValidationState state) const
{
  unsigned int i = 0;
  for(auto &p_protein_match : _protein_match_list)
    {
      if(p_protein_match->getValidationState() >= state)
        {
          i += p_protein_match->countPeptideMatch(state);
        }
    }
  return i;
}

unsigned int
IdentificationGroup::countDecoyPeptideMatch(ValidationState state) const
{
  unsigned int i = 0;
  for(auto &p_protein_match : _protein_match_list)
    {
      if(p_protein_match->getValidationState() >= state)
        {
          if(p_protein_match->getProteinXtpSp().get()->isDecoy())
            {
              i += p_protein_match->countPeptideMatch(state);
            }
        }
    }
  return i;
}
unsigned int
IdentificationGroup::countProteinMatch(ValidationState state) const
{
  return std::count_if(_protein_match_list.begin(),
                       _protein_match_list.end(),
                       [state](const ProteinMatch *p_protein_match) {
                         if(p_protein_match->getValidationState() >= state)
                           {
                             return true;
                           }
                         else
                           {
                             return false;
                           }
                       });
}

void
IdentificationGroup::updateAutomaticFilters(
  const AutomaticFilterParameters &automatic_filter_parameters)
{
  qDebug() << "IdentificationGroup::updateAutomaticFilters begin";
  for(IdentificationDataSource *p_identification_source_list : _id_source_list)
    {
      p_identification_source_list->getPeptideEvidenceStore()
        .updateAutomaticFilters(automatic_filter_parameters);
    }

  qDebug()
    << "IdentificationGroup::updateAutomaticFilters begin p_protein_match";
  for(auto &p_protein_match : _protein_match_list)
    {
      p_protein_match->updateAutomaticFilters(automatic_filter_parameters);
    }

  if(_p_grp_experiment != nullptr)
    {
    }
  qDebug() << "IdentificationGroup::updateAutomaticFilters end";
}
ProteinMatch *
IdentificationGroup::getProteinMatchInstance(const QString accession)
{
  if(accession.isEmpty())
    {
      throw pappso::PappsoException(
        QObject::tr("Error protein match not found : accession is empty"));
    }
  auto it_cache = _cache_accession_protein_match.find(accession);
  if(it_cache == _cache_accession_protein_match.end())
    {
      // accession not found in cache
      ProteinMatch *p_protein_match = new ProteinMatch();
      _cache_accession_protein_match.insert(
        std::pair<QString, ProteinMatch *>(accession, p_protein_match));
      _protein_match_list.push_back(p_protein_match);
      return p_protein_match;
    }
  else
    {
      return it_cache->second;
    }
  return nullptr;
}
void
IdentificationGroup::addProteinMatch(ProteinMatch *protein_match)
{
  _protein_match_list.push_back(protein_match);
}

bool
IdentificationGroup::contains(const MsRun *p_msrun) const
{
  for(const MsRunSp &msrun : _ms_run_list)
    {
      if(msrun.get() == p_msrun)
        return true;
    }
  return false;
}

bool
IdentificationGroup::containSample(const QString &sample) const
{
  for(const MsRunSp &msrun : _ms_run_list)
    {
      if(msrun.get()->getSampleName() == sample)
        return true;
    }
  return false;
}

void
IdentificationGroup::addIdentificationDataSourceP(
  IdentificationDataSource *p_identification_source)
{
  addMsRunSp(p_identification_source->getMsRunSp());
  auto it = std::find(
    _id_source_list.begin(), _id_source_list.end(), p_identification_source);
  if(it == _id_source_list.end())
    {
      _id_source_list.push_back(p_identification_source);
    }
}
void
IdentificationGroup::addMsRunSp(MsRunSp ms_run_sp)
{
  auto it = std::find(_ms_run_list.begin(), _ms_run_list.end(), ms_run_sp);
  if(it == _ms_run_list.end())
    {
      _ms_run_list.push_back(ms_run_sp);
    }
}
const std::vector<MsRunSp> &
IdentificationGroup::getMsRunSpList() const
{
  return _ms_run_list;
}
const std::vector<IdentificationDataSource *> &
IdentificationGroup::getIdentificationDataSourceList() const
{
  return _id_source_list;
}
const std::vector<ProteinMatch *> &
IdentificationGroup::getProteinMatchList() const
{
  return _protein_match_list;
}
std::vector<ProteinMatch *> &
IdentificationGroup::getProteinMatchList()
{
  return _protein_match_list;
}

std::size_t
IdentificationGroup::countGroup() const
{
  return _group_store.countGroup();
}

std::size_t
IdentificationGroup::countSubGroup() const
{
  return _group_store.countSubGroup();
}

void
IdentificationGroup::collectMhDelta(
  std::vector<pappso::pappso_double> &delta_list,
  pappso::PrecisionUnit unit,
  ValidationState state) const
{
  std::set<const PeptideEvidence *> peptide_evidence_list;
  for(auto &p_protein_match : _protein_match_list)
    {
      if(p_protein_match->getValidationState() >= state)
        {
          if(!p_protein_match->getProteinXtpSp().get()->isDecoy())
            {
              // p_protein_match->collectMhDelta(already_counted, delta_list,
              // unit, state);
              p_protein_match->collectPeptideEvidences(peptide_evidence_list,
                                                       state);
            }
        }
    }
  for(const PeptideEvidence *p_peptide_evidence : peptide_evidence_list)
    {
      if(unit == pappso::PrecisionUnit::ppm)
        {
          delta_list.push_back(p_peptide_evidence->getPpmDeltaMass());
        }
      else
        {
          delta_list.push_back(p_peptide_evidence->getDeltaMass());
        }
    }
}
void
IdentificationGroup::startGrouping(
  ContaminantRemovalMode contaminantRemovalMode,
  const GroupingType &grouping_type,
  WorkMonitorInterface *p_work_monitor)
{
  qDebug() << "IdentificationGroup::startGrouping begin ";
  if(_p_grp_experiment != nullptr)
    {
      delete _p_grp_experiment;
    }
  _p_grp_experiment = GroupingExperiment::newInstance(
    contaminantRemovalMode, grouping_type, p_work_monitor);
  for(ProteinMatch *p_protein_match : _protein_match_list)
    {
      p_protein_match->setGroupingExperiment(_p_grp_experiment);
    }

  _p_grp_experiment->startGrouping();
  _group_store.clear();
  for(ProteinMatch *p_protein_match : _protein_match_list)
    {
      p_protein_match->setGroupInstance(_group_store);
    }

  if(_p_grp_ptm_experiment != nullptr)
    {
      delete _p_grp_ptm_experiment;
      _p_grp_ptm_experiment = nullptr;
    }
  qDebug() << "IdentificationGroup::startGrouping end ";
}

void
IdentificationGroup::startPtmGrouping()
{
  qDebug() << "IdentificationGroup::startPtmGrouping begin ";
  if(_p_grp_ptm_experiment != nullptr)
    {
      delete _p_grp_ptm_experiment;
      _p_grp_ptm_experiment = nullptr;
    }
  _p_grp_ptm_experiment = new PtmGroupingExperiment();
  _p_grp_ptm_experiment->setValidationState(ValidationState::grouped);
  for(ProteinMatch *p_protein_match : _protein_match_list)
    {
      _p_grp_ptm_experiment->addProteinMatch(p_protein_match);
    }

  _p_grp_ptm_experiment->startGrouping();

  qDebug() << "IdentificationGroup::startPtmGrouping end ";
}

const QString
IdentificationGroup::getTabName() const
{
  return _ms_run_list[0]->getSampleName();
}

pappso::pappso_double
IdentificationGroup::getPsmFdr(ValidationState state) const
{

  pappso::pappso_double total_peptide = countPeptideEvidence(state);
  pappso::pappso_double false_peptide = countDecoyPeptideEvidence(state);

  return (false_peptide / total_peptide);
}


pappso::pappso_double
IdentificationGroup::getPeptideMassFdr(ValidationState state) const
{
  // ValidationState state = ValidationState::valid;
  pappso::pappso_double total_peptide = countPeptideMassSample(state);
  pappso::pappso_double false_peptide = countDecoyPeptideMassSample(state);

  return (false_peptide / total_peptide);
}

pappso::pappso_double
IdentificationGroup::getProteinFdr(ValidationState state) const
{
  // ValidationState state = ValidationState::valid;
  pappso::pappso_double total_prot = countProteinMatch(state);
  pappso::pappso_double false_prot = countDecoyProteinMatch(state);


  return (false_prot / total_prot);
}


void
IdentificationGroup::getSameXicPeptideEvidenceList(
  std::vector<const PeptideEvidence *> &peptide_evidence_list,
  const MsRun *p_msrun,
  const PeptideXtp *p_peptide,
  unsigned int charge) const
{
  if(!contains(p_msrun))
    return;
  for(const IdentificationDataSource *p_identification_source : _id_source_list)
    {
      if(p_msrun == p_identification_source->getMsRunSp().get())
        {
          p_identification_source->getPeptideEvidenceStore()
            .getSameXicPeptideEvidenceList(
              peptide_evidence_list, p_msrun, p_peptide, charge);
        }
    }
}