cTheoreticalSpectrum.h

Go to the documentation of this file.

 
#ifndef _CTHEORETICALSPECTRUM_H
#define _CTHEORETICALSPECTRUM_H
 
#include <vector>
#include <set>
#include <unordered_set>
#include <map>
#include <string>
#include <QMetaType>
 
#include "core/cParameters.h"
#include "core/cPeakListSeries.h"
#include "core/cCandidateSet.h"
 
class cMainThread;
class cDeNovoGraph;
 
 
using namespace std;
 
 
int getNumberOfScoreHits(vector<double>& scores, double value);
 
 
struct splitSite {
    
    int first;
 
 
    int second;
 
 
    splitSite() {
        first = 0;
        second = 0;
    }
};
 
 
struct matchedSeries {
 
    vector<int> series;
 
 
    string name;
 
 
    matchedSeries() {
        series.clear();
        name = "";
    }
 
 
    void store(ofstream& os);
 
 
    void load(ifstream& is);
 
};
 
 
class cTheoreticalSpectrum {
 
    cParameters* parameters;
    cPeaksList theoreticalpeaks;
    cPeaksList experimentalpeaks;
    cCandidate candidate;
    int experimentalpeaksmatched;
    vector<set<int> > experimentalmatches;
    int scrambledpeaksmatched;
    map<eFragmentIonType, map<int, int> > matchedions;
    int peakstested;
    double experimentalpeaksmatchedratio;
    double weightedpeaksratio;
    double cosinesimilarity;
    int unmatchedexperimentalpeakscount;
    string coveragebyseries;
    bool valid;
    double sumofrelativeintensities;
    vector<string> rotationslabels;
    int validposition;
    int reversevalidposition;
    int seriescompleted;
    int pathid;
    map<int, double> targetscores;
    map<int, double> decoyscores;
    map<int, double> fdrscores;
 
    // remove hits of peaks without hits of parent peaks
    void clearFalseHits(map<eFragmentIonType, map<int, vector<int> > >& series, vector<eFragmentIonType>& ions, vector<int>& losses);
 
    // search for matches of experimental and theoretical peaks
    void searchForPeakPairs(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks, double fragmentmasserrortolerance);
 
    // compute additional scores
    void computeStatistics(bool writedescription);
 
    // generate charged forms of fragments
    void generateChargedFragments(cPeak& peak, map<string, int>& atoms, int& peaklistrealsize, int maxcharge, bool writedescription, bool disablesummary);
 
    // generate precursor ion and its variants
    void generatePrecursorIon(vector<int>& intcomposition, cBricksDatabase& bricksdatabasewithcombinations, int& theoreticalpeaksrealsize, bool writedescription);
 
    // generate precursor ion and its variants for other type
    void generatePrecursorIonForOther(int& theoreticalpeaksrealsize, bool writedescription);
 
    // generate scrambled sequences
    void generateScrambledIons(cBricksDatabase& bricksdatabase, bool writedescription, int& theoreticalpeaksrealsize);
 
    // update the list of losses
    void updateListOfNeutralLosses(cBricksDatabase& bricksdatabase, cBrick& block, vector<int>& currentlosses, vector<double>& fragmentlossmass, vector<string>& fragmentlosssummary, vector< map<string, int> >& fragmentlossmap, bool writedescription, bool disablesummary);
 
    // generate internal fragments
    void generateInternalFragments(cBricksDatabase& bricksdatabase, cPeak& peak, int maxcharge, int& peaklistrealsize, vector<int>& intcomposition, int pos, map<string, int>& atoms, vector<int>& currentlosses, eFragmentIonType fragmentiontype, ePeptideType peptidetype, TRotationInfo* trotation, bool writedescription, bool disablesummary);
 
    // select and normalize scrambled sequences
    void selectAndNormalizeScrambledSequences(unordered_set<string>& scrambledsequences);
 
    // select a proper fragment ion type for an experimental peak when masses of more theoretical fragment ions collide
    void selectHigherPriorityIonTypeCID(cPeak& experimentalpeak, cPeak& theoreticalpeak);
 
    // add a peak to the list
    void addPeakToList(cPeak& peak, int& theoreticalpeaksrealsize);
 
    // add an adduct to a peak description
    void addAdductToDescription(string& description, vector<string>& metals);
 
    // add theoretical peaks containing metals
    void addMetalPeaks(cPeak& peak, vector<string>& metals, int& peaklistrealsize, int peakcharge, bool writedescription);
 
    // remove unmatched biometal isotopes
    void removeUnmatchedMetalIsotopes(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // remove unmatched isotope patterns
    void removeUnmatchedIsotopePatterns(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks, cPeaksList& outputtheoreticalpeaks, bool storeunmatchedpeaks);
 
    // remove unmatched features
    int removeUnmatchedFeatures(bool lcms, int peaklistseriesvectorid, cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks, vector< vector<int> >& hintsindex/*, int id*/);
 
    // remove unmatched compounds
    int removeUnmatchedCompounds(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks, int minimumiontypes);
 
    // remove compounds outside RT range
    void removeCompoundsOutsideRTRange(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // remove peaks under annotation threshold
    void removePeaksUnderAnnotationThreshold(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // remove isotopic patterns under annotation threshold
    void removePatternsUnderAnnotationThreshold(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // remove unmatched patterns in fine isotopic patterns
    void removeUnmatchedPatternsFineSpectra(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // remove unmatched pattern by intensity ratio
    void removeUnmatchedPatternsByIntensityRatio(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // remove unmatched pattern by m/z difference
    void removeUnmatchedPatternsByMZDifference(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // calculate envelope scores
    void calculateEnvelopeScores(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // remove decoy peak matches
    void removeDecoyPeakMatches(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks);
 
    // normalize theoretical intensities
    void normalizeTheoreticalIntensities(cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize);
 
    // calibrate peaklist to the most intense theoretical peak
    double recalibrateToMostIntenseTheoreticalPeak(cPeaksList& peaklist);
 
    // calculate standard PPM deviation
    double getStandardPPMDeviation(cPeaksList& theoreticalpeaks, int start, int stop);
 
    // calculate standard intensity deviation
    double getStandardIntensityDeviation(cPeaksList& theoreticalpeaks, int start, int stop, double maximumexperimentalintensity);
 
    // calculate standard m/z difference deviation
    double getStandardMZDifferenceDeviation(cPeaksList& theoreticalpeaks, int start, int stop);
 
    // calculate the angle distance
    double getAngleDistance(cPeaksList& theoreticalpeaks, int start, int stop);
 
    // generate isotope patterns of fragment ions
    void generateFragmentIsotopePatterns(int& theoreticalpeaksrealsize, bool writedescription, unordered_map<string, int>* isotopeformuladesctoid);
 
    // fill annotations of experimental peaks and remove unmatched theoretical peaks
    void fillExperimentalAnnotationsAndRemoveUnmatchedTheoreticalPeaks(int& theoreticalpeaksrealsize, ePeptideType peptidetype, cPeaksList& unmatchedpeaksinmatchedpatterns, bool reportunmatchedtheoreticalpeaks, bool writedescription);
 
    // check the existence of an isotope
    bool checkIsotopeHighRes(string& elementstring, string& isotopestring, double isotopemass1, double isotopemass2, double minintensityratio, double maxintensityratio, cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks, int start, int stop, int maximumintensityid, double maximumexperimentalintensity);
 
    // check the existence of an isotope
    bool checkIsotopeLowRes(string& elementstring, string& isotopestring, double isotopemass1, double isotopemass2, double minintensityratio, double maxintensityratio, cPeaksList& theoreticalpeaks, int theoreticalpeaksrealsize, cPeaksList& experimentalpeaks, int start, int stop, int maximumintensityid, double maximumexperimentalintensity);
 
    // get intensity multiplier
    double getIntensityMultiplier(cPeaksList& theoreticalpeaks, int peakid, string& elementstring, size_t startpos);
 
 
public:
 
 
    cTheoreticalSpectrum();
 
 
    cTheoreticalSpectrum(cParameters* parameters, cCandidate& candidate);
 
 
    void clear(bool clearpeaklist = true);
    
 
    cCandidate& getCandidate();
    
 
    void setCandidate(cCandidate& candidate);
 
 
    int getNumberOfPeaks();
 
 
    int compareBranched(cPeaksList& sortedpeaklist, cBricksDatabase& bricksdatabasewithcombinations, bool writedescription, regex& sequencetag, regex& searchedsequence, cPeaksList& unmatchedpeaksinmatchedpatterns, unordered_map<string, int>* isotopeformuladesctoid);
 
 
    int compareLinear(cPeaksList& sortedpeaklist, cBricksDatabase& bricksdatabasewithcombinations, bool writedescription, regex& sequencetag, regex& searchedsequence, cPeaksList& unmatchedpeaksinmatchedpatterns, unordered_map<string, int>* isotopeformuladesctoid);
 
 
    int compareCyclic(cPeaksList& sortedpeaklist, cBricksDatabase& bricksdatabasewithcombinations, bool writedescription, regex& sequencetag, regex& searchedsequence, cPeaksList& unmatchedpeaksinmatchedpatterns, unordered_map<string, int>* isotopeformuladesctoid);
 
 
    int compareBranchCyclic(cPeaksList& sortedpeaklist, cBricksDatabase& bricksdatabasewithcombinations, bool writedescription, regex& sequencetag, regex& searchedsequence, cPeaksList& unmatchedpeaksinmatchedpatterns, unordered_map<string, int>* isotopeformuladesctoid);
 
 
    int compareLinearPolyketide(cPeaksList& sortedpeaklist, cBricksDatabase& bricksdatabasewithcombinations, bool writedescription, regex& sequencetag, regex& searchedsequence, cPeaksList& unmatchedpeaksinmatchedpatterns, unordered_map<string, int>* isotopeformuladesctoid);
 
 
    int compareCyclicPolyketide(cPeaksList& sortedpeaklist, cBricksDatabase& bricksdatabasewithcombinations, bool writedescription, regex& sequencetag, regex& searchedsequence, cPeaksList& unmatchedpeaksinmatchedpatterns, unordered_map<string, int>* isotopeformuladesctoid);
 
 
    int compareOther(cPeaksList& sortedpeaklist, bool writedescription, cPeaksList& unmatchedpeaksinmatchedpatterns, unordered_map<string, int>* isotopeformuladesctoid);
 
 
    void generateMSSpectrum(int sequencestart, int sequencestop, bool& terminatecomputation, bool writedescription);
 
 
    void generateFineMSSpectrum(int sequencestart, int sequencestop, bool& terminatecomputation);
 
 
    void getHintsIndex(int id, int peaklistseriesvectorid, cTheoreticalSpectrum& tsfull, cPeaksList& unmatchedpeaksinmatchedpatterns, vector< vector<int> >& hintsindex, bool lcmsrt, bool skipcomparison);
 
 
    void compareMSSpectrum(int id, int peaklistseriesvectorid, cTheoreticalSpectrum& tsfull, cPeaksList& unmatchedpeaksinmatchedpatterns, vector< vector<int> >& hintsindex, bool lcmsrt, bool skipcomparison);
 
 
    void compareAverageMSSpectrum(cPeaksList& averagespectrum, cTheoreticalSpectrum& tsfull, cPeaksList& unmatchedpeaksinmatchedpatterns);
 
 
 
    void finalizeMSSpectrum(cPeaksList& unmatchedpeaksinmatchedpatterns, bool writedescription);
 
 
    int getNumberOfMatchedPeaks() const;
 
 
    int getNumberOfScrambledPeaks();
 
 
    int getNumberOfMatchedPeaks(eFragmentIonType iontype, int neutrallosstype) const;
 
 
    int getNumberOfMatchedPeaksB() const;
 
 
    int getNumberOfMatchedPeaksY() const;
 
 
    int getNumberOfMatchedPeaksYB() const;
 
 
    double getRatioOfMatchedPeaks();
 
 
    double getWeightedRatioOfMatchedPeaks() const;
 
 
    double getCosineSimilarity() const;
 
 
    void generateNTerminalFragmentIons(int maxcharge, int& peaklistrealsize, vector<int>& intcomposition, eFragmentIonType fragmentiontype, int neutrallosstype, cBricksDatabase& bricksdatabase, bool writedescription, int rotationid, vector<splitSite>& splittingsites, vector<cFragmentIonType>& searchedmodifications, ePeptideType peptidetype, bool regularblocksorder, TRotationInfo* trotation = 0, eResidueLossType leftresiduelosstype = h2o_loss, bool hasfirstblockartificial = false);
    
 
    void generateCTerminalFragmentIons(int maxcharge, int& peaklistrealsize, vector<int>& intcomposition, eFragmentIonType fragmentiontype, int neutrallosstype, cBricksDatabase& bricksdatabase, bool writedescription, int rotationid, vector<splitSite>& splittingsites, vector<cFragmentIonType>& searchedmodifications, ePeptideType peptidetype, bool regularblocksorder, TRotationInfo* trotation = 0, eResidueLossType rightresiduelosstype = h2o_loss, bool haslastblockartificial = false);
 
 
    void sortByMass(int limit = -1);
 
 
    void resizePeakList(int size);
 
 
    double getSumOfRelativeIntensities() const;
 
 
    cPeak& operator[](int position);
 
 
    cPeaksList& getExperimentalSpectrum();
 
 
    void setExperimentalSpectrum(cPeaksList& experimentalspectrum);
 
 
    set<int>& getExperimentalMatches(int peakid);
 
 
    string getCoverageBySeries();
 
 
    int getUnmatchedExperimentalPeaksCount();
 
 
    cPeaksList* getTheoreticalPeaks();
 
 
    bool isValid();
 
 
    int getValidPosition();
 
 
    int getReverseValidPosition();
 
 
    void setValidSequence(regex& searchedsequence);
 
 
    vector<string>& getLabelsOfRotations();
 
 
    int getNumberOfCompletedSeries();
 
 
    void setNumberOfCompletedSeries(int numberofcompletedseries);
 
 
    void setPathId(int pathid);
 
 
    int getPathId() const;
 
 
    void setParameters(cParameters* parameters);
 
 
    double getTargetPatternScore(int groupid);
 
 
    double getTargetPatternFDR(int groupid);
 
 
    map<int, double>& getTargetScores();
 
 
    map<int, double>& getDecoyScores();
 
 
    void setFDRs(vector<double>& targetscoresvector, vector<double>& fdrs, cPeaksList& unmatchedpeaksinmatchedpatterns);
 
 
    void store(ofstream& os);
 
 
    void load(ifstream& is, int fileversionpart1, int fileversionpart2, int fileversionpart3);
 
 
    bool equals(cTheoreticalSpectrum& secondtheoreticalspectrum);
 
 
    void attach(cTheoreticalSpectrum& secondtheoreticalspectrum);
 
};
 
 
#endif