private void CalibrateSpectra(RegressionForestModel ms1predictor, RegressionForestModel ms2predictor)
{
Parallel.ForEach(Partitioner.Create(1, myMsDataFile.NumSpectra + 1), fff =>
{
for (int i = fff.Item1; i < fff.Item2; i++)
{
var scan = myMsDataFile.GetOneBasedScan(i);
if (scan is IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> > ms2Scan)
{
var precursorScan = myMsDataFile.GetOneBasedScan(ms2Scan.OneBasedPrecursorScanNumber.Value);
if (!ms2Scan.SelectedIonMonoisotopicGuessIntensity.HasValue && ms2Scan.SelectedIonMonoisotopicGuessMz.HasValue)
{
ms2Scan.ComputeMonoisotopicPeakIntensity(precursorScan.MassSpectrum);
}
double theFunc(IPeak x) => x.X - ms2predictor.Predict(new double[] { x.X, scan.RetentionTime, Math.Log(scan.TotalIonCurrent), scan.InjectionTime.HasValue ? Math.Log(scan.InjectionTime.Value) : double.NaN, Math.Log(x.Y) });
double theFuncForPrecursor(IPeak x) => x.X - ms1predictor.Predict(new double[] { x.X, precursorScan.RetentionTime, Math.Log(precursorScan.TotalIonCurrent), precursorScan.InjectionTime.HasValue ? Math.Log(precursorScan.InjectionTime.Value) : double.NaN, Math.Log(x.Y) });
ms2Scan.TransformMzs(theFunc, theFuncForPrecursor);
}
else
{
Func <IPeak, double> theFunc = x => x.X - ms1predictor.Predict(new double[] { x.X, scan.RetentionTime, Math.Log(scan.TotalIonCurrent), scan.InjectionTime.HasValue ? Math.Log(scan.InjectionTime.Value) : double.NaN, Math.Log(x.Y) });
scan.MassSpectrum.ReplaceXbyApplyingFunction(theFunc);
}
}
}
);
}
protected override MetaMorpheusEngineResults RunSpecific()
{
TerminusType terminusType = ProductTypeMethod.IdentifyTerminusType(lp);
foreach (var ok in allResultingIdentifications)
{
ok.MatchedIonDictOnlyMatches = new Dictionary <ProductType, double[]>();
ok.ProductMassErrorDa = new Dictionary <ProductType, double[]>();
ok.ProductMassErrorPpm = new Dictionary <ProductType, double[]>();
var theScan = myMsDataFile.GetOneBasedScan(ok.ScanNumber);
double thePrecursorMass = ok.ScanPrecursorMass;
foreach (var huh in lp)
{
var ionMasses = ok.CompactPeptides.First().Key.ProductMassesMightHaveDuplicatesAndNaNs(new List <ProductType> {
huh
});
Array.Sort(ionMasses);
List <double> matchedIonMassesList = new List <double>();
List <double> productMassErrorDaList = new List <double>();
List <double> productMassErrorPpmList = new List <double>();
MatchIons(theScan, fragmentTolerance, ionMasses, matchedIonMassesList, productMassErrorDaList, productMassErrorPpmList, thePrecursorMass, dissociationTypes, addCompIons);
double[] matchedIonMassesOnlyMatches = matchedIonMassesList.ToArray();
ok.MatchedIonDictOnlyMatches.Add(huh, matchedIonMassesOnlyMatches);
ok.ProductMassErrorDa.Add(huh, productMassErrorDaList.ToArray());
ok.ProductMassErrorPpm.Add(huh, productMassErrorPpmList.ToArray());
}
}
foreach (var ok in allResultingIdentifications.Where(b => b.NumDifferentCompactPeptides == 1))
{
var theScan = myMsDataFile.GetOneBasedScan(ok.ScanNumber);
double thePrecursorMass = ok.ScanPrecursorMass;
if (ok.FullSequence == null)
{
continue;
}
var representative = ok.CompactPeptides.First().Value.Item2.First();
var localizedScores = new List <double>();
for (int indexToLocalize = 0; indexToLocalize < representative.Length; indexToLocalize++)
{
PeptideWithSetModifications localizedPeptide = representative.Localize(indexToLocalize, ok.ScanPrecursorMass - representative.MonoisotopicMass);
var gg = localizedPeptide.CompactPeptide(terminusType).ProductMassesMightHaveDuplicatesAndNaNs(lp);
Array.Sort(gg);
var score = CalculatePeptideScore(theScan, fragmentTolerance, gg, thePrecursorMass, dissociationTypes, addCompIons, 0);
localizedScores.Add(score);
}
ok.LocalizedScores = localizedScores;
}
return(new LocalizationEngineResults(this));
}
public override IMsDataScan <IMzSpectrum <IMzPeak> > GetOneBasedScan(int oneBasedScanNumber)
{
if (Scans[oneBasedScanNumber - 1] == null)
{
var representativeScanNumber = oneBasedScanNumber + (numScansToAverage - 1) / 2;
var representative = raw.GetOneBasedScan(representativeScanNumber);
if (representative.MsnOrder != 1)
{
throw new MzLibException("Scan " + representativeScanNumber + " is not MS1 scan");
}
int msnOrder = 1;
Polarity polarity = representative.Polarity;
if (!representative.IsCentroid)
{
throw new MzLibException("Scan " + representativeScanNumber + " is not centroid scan");
}
bool isCentroid = true;
double retentionTime = representative.RetentionTime;
MZAnalyzerType mzAnalyzer = representative.MzAnalyzer;
IMzSpectrum <IMzPeak> peaks = CombinePeaks(raw.Where(b => b.OneBasedScanNumber >= oneBasedScanNumber && b.OneBasedScanNumber <= oneBasedScanNumber + numScansToAverage - 1).Select(b => b.MassSpectrum).ToList(), ppmToleranceForPeakCombination);
MzRange scanWindowRange = representative.ScanWindowRange;
double totalIonCurrent = peaks.SumOfAllY;
double injectionTime = double.NaN;
double[,] noiseData = null;
Scans[oneBasedScanNumber - 1] = new MsDataScan <IMzSpectrum <IMzPeak> >(peaks, oneBasedScanNumber, msnOrder, isCentroid, polarity, retentionTime, scanWindowRange, null, mzAnalyzer, totalIonCurrent, injectionTime, noiseData, "scan=" + oneBasedScanNumber);
}
return(Scans[oneBasedScanNumber - 1]);
}
public static IEnumerable <Ms2ScanWithSpecificMass> GetMs2Scans(
IMsDataFile <IMsDataScan <IMzSpectrum <IMzPeak> > > myMSDataFile,
string fullFilePath,
bool doPrecursorDeconvolution,
bool useProvidedPrecursorInfo,
double deconvolutionIntensityRatio,
int deconvolutionMaxAssumedChargeState,
Tolerance deconvolutionMassTolerance)
{
foreach (var ms2scan in myMSDataFile.OfType <IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> > >())
{
List <(double, int)> isolatedStuff = new List <(double, int)>();
if (ms2scan.OneBasedPrecursorScanNumber.HasValue)
{
var precursorSpectrum = myMSDataFile.GetOneBasedScan(ms2scan.OneBasedPrecursorScanNumber.Value);
ms2scan.RefineSelectedMzAndIntensity(precursorSpectrum.MassSpectrum);
if (ms2scan.SelectedIonMonoisotopicGuessMz.HasValue)
{
ms2scan.ComputeMonoisotopicPeakIntensity(precursorSpectrum.MassSpectrum);
}
if (doPrecursorDeconvolution)
{
foreach (var envelope in ms2scan.GetIsolatedMassesAndCharges(precursorSpectrum.MassSpectrum, 1, deconvolutionMaxAssumedChargeState, deconvolutionMassTolerance.Value, deconvolutionIntensityRatio))
{
var monoPeakMz = envelope.monoisotopicMass.ToMz(envelope.charge);
isolatedStuff.Add((monoPeakMz, envelope.charge));
}
}
}
if (useProvidedPrecursorInfo && ms2scan.SelectedIonChargeStateGuess.HasValue)
{
var precursorCharge = ms2scan.SelectedIonChargeStateGuess.Value;
if (ms2scan.SelectedIonMonoisotopicGuessMz.HasValue)
{
var precursorMZ = ms2scan.SelectedIonMonoisotopicGuessMz.Value;
if (!isolatedStuff.Any(b => deconvolutionMassTolerance.Within(precursorMZ.ToMass(precursorCharge), b.Item1.ToMass(b.Item2))))
{
isolatedStuff.Add((precursorMZ, precursorCharge));
}
}
else
{
var precursorMZ = ms2scan.SelectedIonMZ;
if (!isolatedStuff.Any(b => deconvolutionMassTolerance.Within(precursorMZ.ToMass(precursorCharge), b.Item1.ToMass(b.Item2))))
{
isolatedStuff.Add((precursorMZ, precursorCharge));
}
}
}
foreach (var heh in isolatedStuff)
{
yield return(new Ms2ScanWithSpecificMass(ms2scan, heh.Item1, heh.Item2, fullFilePath));
}
}
}
public static void CreateAndWriteMyMzmlWithCalibratedSpectra(IMsDataFile <IMsDataScan <IMzSpectrum <IMzPeak> > > myMsDataFile, string outputFile, bool writeIndexed)
{
var mzML = new Generated.mzMLType()
{
version = "1",
cvList = new Generated.CVListType()
};
mzML.cvList.count = "1";
mzML.cvList.cv = new Generated.CVType[1];
mzML.cvList.cv[0] = new Generated.CVType()
{
URI = @"https://raw.githubusercontent.com/HUPO-PSI/psi-ms-CV/master/psi-ms.obo",
fullName = "Proteomics Standards Initiative Mass Spectrometry Ontology",
id = "MS"
};
mzML.fileDescription = new Generated.FileDescriptionType()
{
fileContent = new Generated.ParamGroupType()
};
mzML.fileDescription.fileContent.cvParam = new Generated.CVParamType[2];
mzML.fileDescription.fileContent.cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000579" // MS1 Data
};
mzML.fileDescription.fileContent.cvParam[1] = new Generated.CVParamType()
{
accession = "MS:1000580" // MSn Data
};
mzML.softwareList = new Generated.SoftwareListType()
{
count = "1",
software = new Generated.SoftwareType[1]
};
// TODO: add the raw file fields
mzML.softwareList.software[0] = new Generated.SoftwareType()
{
id = "mzLib",
version = "1",
cvParam = new Generated.CVParamType[1]
};
mzML.softwareList.software[0].cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000799",
value = "mzLib"
};
// Leaving empty. Can't figure out the configurations.
// ToDo: read instrumentConfigurationList from mzML file
mzML.instrumentConfigurationList = new Generated.InstrumentConfigurationListType();
mzML.dataProcessingList = new Generated.DataProcessingListType()
{
count = "1",
dataProcessing = new Generated.DataProcessingType[1]
};
// Only writing mine! Might have had some other data processing (but not if it is a raw file)
// ToDo: read dataProcessingList from mzML file
mzML.dataProcessingList.dataProcessing[0] = new Generated.DataProcessingType()
{
id = "mzLibProcessing"
};
mzML.run = new Generated.RunType()
{
chromatogramList = new Generated.ChromatogramListType()
{
count = "1",
chromatogram = new Generated.ChromatogramType[1]
}
};
// ToDo: Finish the chromatogram writing!
mzML.run.chromatogramList.chromatogram[0] = new Generated.ChromatogramType();
mzML.run.spectrumList = new Generated.SpectrumListType()
{
count = (myMsDataFile.NumSpectra).ToString(CultureInfo.InvariantCulture),
defaultDataProcessingRef = "mzLibProcessing",
spectrum = new Generated.SpectrumType[myMsDataFile.NumSpectra]
};
// Loop over all spectra
for (int i = 1; i <= myMsDataFile.NumSpectra; i++)
{
mzML.run.spectrumList.spectrum[i - 1] = new Generated.SpectrumType()
{
defaultArrayLength = myMsDataFile.GetOneBasedScan(i).MassSpectrum.Size,
index = i.ToString(CultureInfo.InvariantCulture),
id = myMsDataFile.GetOneBasedScan(i).OneBasedScanNumber.ToString(),
cvParam = new Generated.CVParamType[8]
};
mzML.run.spectrumList.spectrum[i - 1].cvParam[0] = new Generated.CVParamType();
if (myMsDataFile.GetOneBasedScan(i).MsnOrder == 1)
{
mzML.run.spectrumList.spectrum[i - 1].cvParam[0].accession = "MS:1000579";
}
else if (myMsDataFile.GetOneBasedScan(i) is IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> > )
{
var scanWithPrecursor = myMsDataFile.GetOneBasedScan(i) as IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> >;
mzML.run.spectrumList.spectrum[i - 1].cvParam[0].accession = "MS:1000580";
// So needs a precursor!
mzML.run.spectrumList.spectrum[i - 1].precursorList = new Generated.PrecursorListType()
{
count = 1.ToString(),
precursor = new Generated.PrecursorType[1]
};
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0] = new Generated.PrecursorType();
string precursorID = scanWithPrecursor.OneBasedPrecursorScanNumber.ToString();
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].spectrumRef = precursorID;
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].selectedIonList = new Generated.SelectedIonListType()
{
count = 1.ToString(),
selectedIon = new Generated.ParamGroupType[1]
};
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].selectedIonList.selectedIon[0] = new Generated.ParamGroupType()
{
cvParam = new Generated.CVParamType[3]
};
// Selected ion MZ
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].selectedIonList.selectedIon[0].cvParam[0] = new Generated.CVParamType()
{
name = "selected ion m/z",
value = scanWithPrecursor.SelectedIonMZ.ToString(CultureInfo.InvariantCulture),
accession = "MS:1000744"
};
// Charge State
if (scanWithPrecursor.SelectedIonChargeStateGuess.HasValue)
{
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].selectedIonList.selectedIon[0].cvParam[1] = new Generated.CVParamType()
{
name = "charge state",
value = scanWithPrecursor.SelectedIonChargeStateGuess.Value.ToString(CultureInfo.InvariantCulture),
accession = "MS:1000041"
};
}
// Selected ion intensity
if (scanWithPrecursor.SelectedIonIntensity.HasValue)
{
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].selectedIonList.selectedIon[0].cvParam[2] = new Generated.CVParamType()
{
name = "peak intensity",
value = scanWithPrecursor.SelectedIonIntensity.Value.ToString(CultureInfo.InvariantCulture),
accession = "MS:1000042"
};
}
MzRange isolationRange = scanWithPrecursor.IsolationRange;
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].isolationWindow = new Generated.ParamGroupType()
{
cvParam = new Generated.CVParamType[3]
};
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].isolationWindow.cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000827",
name = "isolation window target m/z",
value = isolationRange.Mean.ToString(CultureInfo.InvariantCulture)
};
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].isolationWindow.cvParam[1] = new Generated.CVParamType()
{
accession = "MS:1000828",
name = "isolation window lower offset",
value = (isolationRange.Width / 2).ToString(CultureInfo.InvariantCulture)
};
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].isolationWindow.cvParam[2] = new Generated.CVParamType()
{
accession = "MS:1000829",
name = "isolation window upper offset",
value = (isolationRange.Width / 2).ToString(CultureInfo.InvariantCulture)
};
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].activation = new Generated.ParamGroupType()
{
cvParam = new Generated.CVParamType[1]
};
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].activation.cvParam[0] = new Generated.CVParamType();
DissociationType dissociationType = scanWithPrecursor.DissociationType;
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].activation.cvParam[0].accession = DissociationTypeAccessions[dissociationType];
mzML.run.spectrumList.spectrum[i - 1].precursorList.precursor[0].activation.cvParam[0].name = DissociationTypeNames[dissociationType];
}
mzML.run.spectrumList.spectrum[i - 1].cvParam[1] = new Generated.CVParamType()
{
name = "ms level",
accession = "MS:1000511",
value = myMsDataFile.GetOneBasedScan(i).MsnOrder.ToString(CultureInfo.InvariantCulture)
};
mzML.run.spectrumList.spectrum[i - 1].cvParam[2] = new Generated.CVParamType()
{
name = CentroidNames[myMsDataFile.GetOneBasedScan(i).IsCentroid],
accession = CentroidAccessions[myMsDataFile.GetOneBasedScan(i).IsCentroid]
};
if (PolarityNames.TryGetValue(myMsDataFile.GetOneBasedScan(i).Polarity, out string polarityName) && PolarityAccessions.TryGetValue(myMsDataFile.GetOneBasedScan(i).Polarity, out string polarityAccession))
{
mzML.run.spectrumList.spectrum[i - 1].cvParam[3] = new Generated.CVParamType()
{
name = polarityName,
accession = polarityAccession
};
}
// Spectrum title
mzML.run.spectrumList.spectrum[i - 1].cvParam[4] = new Generated.CVParamType()
{
name = "spectrum title",
accession = "MS:1000796",
value = myMsDataFile.GetOneBasedScan(i).OneBasedScanNumber.ToString()
};
if ((myMsDataFile.GetOneBasedScan(i).MassSpectrum.Size) > 0)
{
// Lowest observed mz
mzML.run.spectrumList.spectrum[i - 1].cvParam[5] = new Generated.CVParamType()
{
name = "lowest observed m/z",
accession = "MS:1000528",
value = myMsDataFile.GetOneBasedScan(i).MassSpectrum.FirstX.ToString(CultureInfo.InvariantCulture)
};
// Highest observed mz
mzML.run.spectrumList.spectrum[i - 1].cvParam[6] = new Generated.CVParamType()
{
name = "highest observed m/z",
accession = "MS:1000527",
value = myMsDataFile.GetOneBasedScan(i).MassSpectrum.LastX.ToString(CultureInfo.InvariantCulture)
};
}
// Total ion current
mzML.run.spectrumList.spectrum[i - 1].cvParam[7] = new Generated.CVParamType()
{
name = "total ion current",
accession = "MS:1000285",
value = myMsDataFile.GetOneBasedScan(i).TotalIonCurrent.ToString(CultureInfo.InvariantCulture)
};
// Retention time
mzML.run.spectrumList.spectrum[i - 1].scanList = new Generated.ScanListType()
{
count = "1",
scan = new Generated.ScanType[1]
};
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0] = new Generated.ScanType()
{
cvParam = new Generated.CVParamType[3]
};
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].cvParam[0] = new Generated.CVParamType()
{
name = "scan start time",
accession = "MS:1000016",
value = myMsDataFile.GetOneBasedScan(i).RetentionTime.ToString(CultureInfo.InvariantCulture),
unitCvRef = "UO",
unitAccession = "UO:0000031",
unitName = "minute"
};
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].cvParam[1] = new Generated.CVParamType()
{
name = "filter string",
accession = "MS:1000512",
value = myMsDataFile.GetOneBasedScan(i).ScanFilter
};
if (myMsDataFile.GetOneBasedScan(i).InjectionTime.HasValue)
{
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].cvParam[2] = new Generated.CVParamType()
{
name = "ion injection time",
accession = "MS:1000927",
value = myMsDataFile.GetOneBasedScan(i).InjectionTime.Value.ToString(CultureInfo.InvariantCulture)
};
}
if (myMsDataFile.GetOneBasedScan(i) is IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> > )
{
var scanWithPrecursor = myMsDataFile.GetOneBasedScan(i) as IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> >;
if (scanWithPrecursor.SelectedIonMonoisotopicGuessMz.HasValue)
{
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].userParam = new Generated.UserParamType[1];
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].userParam[0] = new Generated.UserParamType()
{
name = "[mzLib]Monoisotopic M/Z:",
value = scanWithPrecursor.SelectedIonMonoisotopicGuessMz.Value.ToString(CultureInfo.InvariantCulture)
};
}
}
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].scanWindowList = new Generated.ScanWindowListType()
{
count = 1,
scanWindow = new Generated.ParamGroupType[1]
};
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].scanWindowList.scanWindow[0] = new Generated.ParamGroupType()
{
cvParam = new Generated.CVParamType[2]
};
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].scanWindowList.scanWindow[0].cvParam[0] = new Generated.CVParamType()
{
name = "scan window lower limit",
accession = "MS:1000501",
value = myMsDataFile.GetOneBasedScan(i).ScanWindowRange.Minimum.ToString(CultureInfo.InvariantCulture)
};
mzML.run.spectrumList.spectrum[i - 1].scanList.scan[0].scanWindowList.scanWindow[0].cvParam[1] = new Generated.CVParamType()
{
name = "scan window upper limit",
accession = "MS:1000500",
value = myMsDataFile.GetOneBasedScan(i).ScanWindowRange.Maximum.ToString(CultureInfo.InvariantCulture)
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList = new Generated.BinaryDataArrayListType()
{
// ONLY WRITING M/Z AND INTENSITY DATA, NOT THE CHARGE! (but can add charge info later)
// CHARGE (and other stuff) CAN BE IMPORTANT IN ML APPLICATIONS!!!!!
count = 2.ToString(),
binaryDataArray = new Generated.BinaryDataArrayType[5]
};
// M/Z Data
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[0] = new Generated.BinaryDataArrayType()
{
binary = myMsDataFile.GetOneBasedScan(i).MassSpectrum.Get64BitXarray()
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[0].encodedLength = (4 * Math.Ceiling(((double)mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[0].binary.Length / 3))).ToString(CultureInfo.InvariantCulture);
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[0].cvParam = new Generated.CVParamType[3];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[0].cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000514",
name = "m/z array"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[0].cvParam[1] = new Generated.CVParamType()
{
accession = "MS:1000523",
name = "64-bit float"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[0].cvParam[2] = new Generated.CVParamType()
{
accession = "MS:1000576",
name = "no compression"
};
// Intensity Data
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[1] = new Generated.BinaryDataArrayType()
{
binary = myMsDataFile.GetOneBasedScan(i).MassSpectrum.Get64BitYarray()
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[1].encodedLength = (4 * Math.Ceiling(((double)mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[1].binary.Length / 3))).ToString(CultureInfo.InvariantCulture);
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[1].cvParam = new Generated.CVParamType[3];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[1].cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000515",
name = "intensity array"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[1].cvParam[1] = new Generated.CVParamType()
{
accession = "MS:1000523",
name = "64-bit float"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[1].cvParam[2] = new Generated.CVParamType()
{
accession = "MS:1000576",
name = "no compression"
};
if (myMsDataFile.GetOneBasedScan(i).NoiseData != null)
{
// mass
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2] = new Generated.BinaryDataArrayType()
{
binary = myMsDataFile.GetOneBasedScan(i).Get64BitNoiseDataMass()
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].encodedLength = (4 * Math.Ceiling(((double)mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].binary.Length / 3))).ToString(CultureInfo.InvariantCulture);
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].cvParam = new Generated.CVParamType[3];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000786",
name = "non-standard data array"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].cvParam[1] = new Generated.CVParamType()
{
accession = "MS:1000523",
name = "64-bit float"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].cvParam[2] = new Generated.CVParamType()
{
accession = "MS:1000576",
name = "no compression"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].userParam = new Generated.UserParamType[1];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[2].userParam[0] = new Generated.UserParamType()
{
name = "kelleherCustomType",
value = "noise m/z"
};
// noise
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3] = new Generated.BinaryDataArrayType()
{
binary = myMsDataFile.GetOneBasedScan(i).Get64BitNoiseDataNoise()
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].encodedLength = (4 * Math.Ceiling(((double)mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].binary.Length / 3))).ToString(CultureInfo.InvariantCulture);
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].cvParam = new Generated.CVParamType[3];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000786",
name = "non-standard data array"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].cvParam[1] = new Generated.CVParamType()
{
accession = "MS:1000523",
name = "64-bit float"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].cvParam[2] = new Generated.CVParamType()
{
accession = "MS:1000576",
name = "no compression"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].userParam = new Generated.UserParamType[1];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[3].userParam[0] = new Generated.UserParamType()
{
name = "kelleherCustomType",
value = "noise baseline"
};
// baseline
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4] = new Generated.BinaryDataArrayType()
{
binary = myMsDataFile.GetOneBasedScan(i).Get64BitNoiseDataBaseline()
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].encodedLength = (4 * Math.Ceiling(((double)mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].binary.Length / 3))).ToString(CultureInfo.InvariantCulture);
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].cvParam = new Generated.CVParamType[3];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].cvParam[0] = new Generated.CVParamType()
{
accession = "MS:1000786",
name = "non-standard data array"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].cvParam[1] = new Generated.CVParamType()
{
accession = "MS:1000523",
name = "64-bit float"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].cvParam[2] = new Generated.CVParamType()
{
accession = "MS:1000576",
name = "no compression"
};
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].userParam = new Generated.UserParamType[1];
mzML.run.spectrumList.spectrum[i - 1].binaryDataArrayList.binaryDataArray[4].userParam[0] = new Generated.UserParamType()
{
name = "kelleherCustomType",
value = "noise intensity"
};
}
}
if (writeIndexed)
{
throw new NotImplementedException("Writing indexed mzMLs not yet supported");
}
else
{
using (TextWriter writer = new StreamWriter(outputFile))
{
mzmlSerializer.Serialize(writer, mzML);
}
}
}
protected override MetaMorpheusEngineResults RunSpecific()
{
Status("Extracting data points:");
// The final training point list
int numMs1MassChargeCombinationsConsidered = 0;
int numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks = 0;
int numMs2MassChargeCombinationsConsidered = 0;
int numMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks = 0;
List <LabeledMs1DataPoint> Ms1List = new List <LabeledMs1DataPoint>();
List <LabeledMs2DataPoint> Ms2List = new List <LabeledMs2DataPoint>();
int numIdentifications = goodIdentifications.Count;
// Loop over identifications
HashSet <string> sequences = new HashSet <string>();
object lockObj = new object();
object lockObj2 = new object();
Parallel.ForEach(Partitioner.Create(0, numIdentifications), fff =>
{
for (int matchIndex = fff.Item1; matchIndex < fff.Item2; matchIndex++)
{
PeptideSpectralMatch identification = goodIdentifications[matchIndex];
// Each identification has an MS2 spectrum attached to it.
int ms2scanNumber = identification.ScanNumber;
int peptideCharge = identification.ScanPrecursorCharge;
if (identification.FullSequence == null)
{
continue;
}
var representativeSinglePeptide = identification.CompactPeptides.First().Value.Item2.First();
// Get the peptide, don't forget to add the modifications!!!!
var SequenceWithChemicalFormulas = representativeSinglePeptide.SequenceWithChemicalFormulas;
if (SequenceWithChemicalFormulas == null || representativeSinglePeptide.allModsOneIsNterminus.Any(b => b.Value.neutralLosses.Count != 1 || b.Value.neutralLosses.First() != 0))
{
continue;
}
Proteomics.Peptide coolPeptide = new Proteomics.Peptide(SequenceWithChemicalFormulas);
var ms2tuple = SearchMS2Spectrum(myMsDataFile.GetOneBasedScan(ms2scanNumber) as IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> >, coolPeptide, peptideCharge, identification);
// If MS2 has low evidence for peptide, skip and go to next one
if (ms2tuple.Item4 < numFragmentsNeededForEveryIdentification)
{
continue;
}
lock (lockObj2)
{
Ms2List.AddRange(ms2tuple.Item1);
numMs2MassChargeCombinationsConsidered += ms2tuple.Item2;
numMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks += ms2tuple.Item3;
if (sequences.Contains(identification.FullSequence))
{
continue; // Do not search same sequence multiple times in MS1 scans
}
sequences.Add(identification.FullSequence);
}
// Calculate isotopic distribution of the full peptide
var dist = IsotopicDistribution.GetDistribution(coolPeptide.GetChemicalFormula(), fineResolutionForIsotopeDistCalculation, 0.001);
double[] theoreticalMasses = dist.Masses.ToArray();
double[] theoreticalIntensities = dist.Intensities.ToArray();
Array.Sort(theoreticalIntensities, theoreticalMasses, Comparer <double> .Create((x, y) => y.CompareTo(x)));
var ms1tupleBack = SearchMS1Spectra(theoreticalMasses, theoreticalIntensities, ms2scanNumber, -1, peptideCharge, identification);
var ms1tupleForward = SearchMS1Spectra(theoreticalMasses, theoreticalIntensities, ms2scanNumber, 1, peptideCharge, identification);
lock (lockObj)
{
Ms1List.AddRange(ms1tupleBack.Item1);
numMs1MassChargeCombinationsConsidered += ms1tupleBack.Item2;
numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks += ms1tupleBack.Item3;
Ms1List.AddRange(ms1tupleForward.Item1);
numMs1MassChargeCombinationsConsidered += ms1tupleForward.Item2;
numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks += ms1tupleForward.Item3;
}
}
});
return(new DataPointAquisitionResults(this,
Ms1List,
Ms2List,
numMs1MassChargeCombinationsConsidered,
numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks,
numMs2MassChargeCombinationsConsidered,
numMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks
));
}
private (List <PeptideSpectralMatch>, DataPointAquisitionResults) GetDataAcquisitionResults(IMsDataFile <IMsDataScan <IMzSpectrum <IMzPeak> > > myMsDataFile, string currentDataFile, List <ModificationWithMass> variableModifications, List <ModificationWithMass> fixedModifications, List <Protein> proteinList, string taskId, ICommonParameters combinedParameters, Tolerance initPrecTol, Tolerance initProdTol)
{
var fileNameWithoutExtension = Path.GetFileNameWithoutExtension(currentDataFile);
MassDiffAcceptor searchMode;
if (initPrecTol is PpmTolerance)
{
searchMode = new SinglePpmAroundZeroSearchMode(initPrecTol.Value);
}
else
{
searchMode = new SingleAbsoluteAroundZeroSearchMode(initPrecTol.Value);
}
FragmentTypes fragmentTypesForCalibration = FragmentTypes.None;
if (combinedParameters.BIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.b;
}
if (combinedParameters.YIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.y;
}
if (combinedParameters.CIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.c;
}
if (combinedParameters.ZdotIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.zdot;
}
var listOfSortedms2Scans = GetMs2Scans(myMsDataFile, currentDataFile, combinedParameters.DoPrecursorDeconvolution, combinedParameters.UseProvidedPrecursorInfo, combinedParameters.DeconvolutionIntensityRatio, combinedParameters.DeconvolutionMaxAssumedChargeState, combinedParameters.DeconvolutionMassTolerance).OrderBy(b => b.PrecursorMass).ToArray();
PeptideSpectralMatch[] allPsmsArray = new PeptideSpectralMatch[listOfSortedms2Scans.Length];
List <ProductType> lp = new List <ProductType>();
if (combinedParameters.BIons)
{
lp.Add(ProductType.B);
}
if (combinedParameters.YIons)
{
lp.Add(ProductType.Y);
}
if (combinedParameters.CIons)
{
lp.Add(ProductType.C);
}
if (combinedParameters.ZdotIons)
{
lp.Add(ProductType.Zdot);
}
Log("Searching with searchMode: " + searchMode, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
});
Log("Searching with productMassTolerance: " + initProdTol, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
});
new ClassicSearchEngine(allPsmsArray, listOfSortedms2Scans, variableModifications, fixedModifications, proteinList, lp, searchMode, false, combinedParameters, initProdTol, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
List <PeptideSpectralMatch> allPsms = allPsmsArray.ToList();
Dictionary <CompactPeptideBase, HashSet <PeptideWithSetModifications> > compactPeptideToProteinPeptideMatching = ((SequencesToActualProteinPeptidesEngineResults) new SequencesToActualProteinPeptidesEngine(allPsms, proteinList, fixedModifications, variableModifications, lp, new List <IDigestionParams> {
combinedParameters.DigestionParams
}, combinedParameters.ReportAllAmbiguity, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run()).CompactPeptideToProteinPeptideMatching;
foreach (var huh in allPsms)
{
if (huh != null)
{
huh.MatchToProteinLinkedPeptides(compactPeptideToProteinPeptideMatching);
}
}
allPsms = allPsms.Where(b => b != null).OrderByDescending(b => b.Score).ThenBy(b => b.PeptideMonisotopicMass.HasValue ? Math.Abs(b.ScanPrecursorMass - b.PeptideMonisotopicMass.Value) : double.MaxValue).GroupBy(b => (b.FullFilePath, b.ScanNumber, b.PeptideMonisotopicMass)).Select(b => b.First()).ToList();
new FdrAnalysisEngine(allPsms, searchMode.NumNotches, false, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
List <PeptideSpectralMatch> goodIdentifications = allPsms.Where(b => b.FdrInfo.QValueNotch < 0.01 && !b.IsDecoy && b.FullSequence != null).ToList();
if (!goodIdentifications.Any())
{
Warn("No PSMs below 1% FDR observed!");
return(new List <PeptideSpectralMatch>(), null);
}
var dissociationTypes = MetaMorpheusEngine.DetermineDissociationType(lp);
foreach (var psm in allPsms)
{
var theScan = myMsDataFile.GetOneBasedScan(psm.ScanNumber);
double thePrecursorMass = psm.ScanPrecursorMass;
foreach (var huh in lp)
{
var ionMasses = psm.CompactPeptides.First().Key.ProductMassesMightHaveDuplicatesAndNaNs(new List <ProductType> {
huh
});
Array.Sort(ionMasses);
List <double> matchedIonMassesList = new List <double>();
List <double> productMassErrorDaList = new List <double>();
List <double> productMassErrorPpmList = new List <double>();
LocalizationEngine.MatchIons(theScan, initProdTol, ionMasses, matchedIonMassesList, productMassErrorDaList, productMassErrorPpmList, thePrecursorMass, dissociationTypes, false);
double[] matchedIonMassesOnlyMatches = matchedIonMassesList.ToArray();
psm.MatchedIonDictOnlyMatches.Add(huh, matchedIonMassesOnlyMatches);
psm.ProductMassErrorDa.Add(huh, productMassErrorDaList.ToArray());
psm.ProductMassErrorPpm.Add(huh, productMassErrorPpmList.ToArray());
}
}
DataPointAquisitionResults currentResult = (DataPointAquisitionResults) new DataPointAcquisitionEngine(
goodIdentifications,
myMsDataFile,
initPrecTol,
initProdTol,
CalibrationParameters.NumFragmentsNeededForEveryIdentification,
CalibrationParameters.MinMS1IsotopicPeaksNeededForConfirmedIdentification,
CalibrationParameters.MinMS2IsotopicPeaksNeededForConfirmedIdentification,
fragmentTypesForCalibration,
new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
return(goodIdentifications, currentResult);
}