public void TestAbpSumMs1Spectra()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
var specFilePath = Path.Combine(Utils.DEFAULT_TEST_FILE_FOLDER, @"TestYufengData\QC_ShewIntact_2ug_3k_CID_4Apr14_Bane_PL011402.raw");
if (!File.Exists(specFilePath))
{
Assert.Ignore(@"Skipping test " + methodName + @" since file not found: " + specFilePath);
}
const int minScanNum = 5657;
const int maxScanNum = 5699;
const int MAX_POINTS = 50;
var run = PbfLcMsRun.GetLcMsRun(specFilePath);
if (run == null)
{
return;
}
var summedSpec = run.GetSummedMs1Spectrum(minScanNum, maxScanNum);
var peakList = summedSpec.GetPeakListWithin(1180.0, 1192.0);
var filteredPeakList = new List <Peak>();
PeakListUtils.FilterNoise(peakList, ref filteredPeakList);
new Spectrum(filteredPeakList, 0).Display(MAX_POINTS);
}
private void SetLcMsMatches(int ms2ScanNumber)
{
var productSpec = _run.GetSpectrum(ms2ScanNumber) as ProductSpectrum;
if (productSpec == null)
{
return;
}
var isolationWindow = productSpec.IsolationWindow;
var minMz = isolationWindow.MinMz;
var maxMz = isolationWindow.MaxMz;
List <Peak> precursorPeakList = null;
List <Peak> precursorSpecWindow = null;
var precursorSpec = _run.GetSpectrum(_run.GetPrecursorScanNum(ms2ScanNumber));
if (precursorSpec != null)
{
precursorPeakList = precursorSpec.GetPeakListWithin(minMz, maxMz);
precursorSpecWindow = precursorSpec.GetPeakListWithin(minMz - 1.0, maxMz + 1.0);
}
List <Peak> nextMs1PeakList = null;
List <Peak> nextMs1SpecWindow = null;
var nextScanNum = _run.GetNextScanNum(ms2ScanNumber, 1);
var nextSpec = _run.GetSpectrum(nextScanNum);
if (nextSpec != null)
{
nextMs1PeakList = nextSpec.GetPeakListWithin(minMz, maxMz);
nextMs1SpecWindow = nextSpec.GetPeakListWithin(minMz - 1.0, maxMz + 1.0);
}
List <Peak> peakList;
if (precursorPeakList != null && nextMs1PeakList != null)
{
peakList = PeakListUtils.Sum(precursorPeakList, nextMs1PeakList, _comparer);
}
else if (precursorPeakList != null)
{
peakList = precursorPeakList;
}
else if (nextMs1PeakList != null)
{
peakList = nextMs1PeakList;
}
else
{
return;
}
// Sort by intensity
peakList.Sort(new IntensityComparer());
var remainingPeakList = new LinkedList <Peak>(peakList);
SetLcMsMatches(remainingPeakList, ms2ScanNumber, precursorSpecWindow, nextMs1SpecWindow, MaxNumPeaksToConsider);
}
private IEnumerable <DeisotopedPeak> GetDeisotopedPeaks(List <Peak> specWindow, IEnumerable <Peak> peakList, int numDeisotopedPeaksToGet)
{
var peakListSortedByIntensity = new List <Peak>(peakList);
peakListSortedByIntensity.Sort(new IntensityComparer());
var remainingPeakList = new LinkedList <Peak>(peakListSortedByIntensity);
var deisotopedPeakSet = new SortedSet <DeisotopedPeak>();
while (remainingPeakList.Any())
{
var peakWithHighestIntensity = remainingPeakList.First.Value;
var peakMz = peakWithHighestIntensity.Mz;
var score = new double[_maxCharge + 1];
for (var charge = _maxCharge; charge >= _minCharge; charge--)
{
// check whether next isotope peak exists
var nextIsotopeMz = peakMz + Constants.C13MinusC12 / charge;
var nextIsotopePeak = PeakListUtils.FindPeak(specWindow, nextIsotopeMz, _tolerance);
if (nextIsotopePeak == null)
{
continue;
}
var mostAbundantIsotopeMass = (peakMz - Constants.Proton) * charge;
var averagineIsoEnv = Averagine.GetIsotopomerEnvelope(mostAbundantIsotopeMass);
var approxMostAbundantIsotopeIndex = averagineIsoEnv.MostAbundantIsotopeIndex;
var monoIsotopicMass = mostAbundantIsotopeMass - approxMostAbundantIsotopeIndex * Constants.C13MinusC12;
var averagineIsotopeProfile = Averagine.GetTheoreticalIsotopeProfile(monoIsotopicMass, charge);
var corr = PeakListUtils.GetPearsonCorrelation(specWindow, averagineIsotopeProfile, _comparer);
score[charge] = corr;
var isValid = true;
for (var mult = 2; mult <= _maxCharge / charge; mult++)
{
var multiple = charge * mult;
if (score[multiple] > 0.8 * corr)
{
isValid = false;
break;
}
}
if (!isValid)
{
continue;
}
deisotopedPeakSet.Add(new DeisotopedPeak(monoIsotopicMass, charge, corr));
if (deisotopedPeakSet.Count > numDeisotopedPeaksToGet)
{
deisotopedPeakSet.Remove(deisotopedPeakSet.Min);
}
}
remainingPeakList.RemoveFirst();
}
return(deisotopedPeakSet);
}
private void ApplyDeconvolution(List <Peak> specWindow, ref LinkedList <Peak> remainingPeakList, ref List <double> deisotopedMassList)
{
if (!remainingPeakList.Any())
{
return;
}
var peakWithHighestIntensity = remainingPeakList.First.Value;
var peakMz = peakWithHighestIntensity.Mz;
var score = new double[_maxCharge + 1];
for (var charge = _maxCharge; charge >= _minCharge; charge--)
{
// check whether next isotope peak exists
var nextIsotopeMz = peakMz + Constants.C13MinusC12 / charge;
var nextIsotopePeak = PeakListUtils.FindPeak(specWindow, nextIsotopeMz, _tolerance);
if (nextIsotopePeak == null)
{
continue;
}
var mostAbundantIsotopeMass = (peakMz - Constants.Proton) * charge;
var averagineIsoEnv = Averagine.GetIsotopomerEnvelope(mostAbundantIsotopeMass);
var approxMostAbundantIsotopeIndex = averagineIsoEnv.MostAbundantIsotopeIndex;
var monoIsotopicMass = mostAbundantIsotopeMass - approxMostAbundantIsotopeIndex * Constants.C13MinusC12;
var averagineIsotopeProfile = Averagine.GetTheoreticalIsotopeProfile(monoIsotopicMass, charge);
var corr = PeakListUtils.GetPearsonCorrelation(specWindow, averagineIsotopeProfile, _comparer);
score[charge] = corr;
var isValid = true;
for (var mult = 2; mult <= _maxCharge / charge; mult++)
{
var multiple = charge * mult;
if (score[multiple] > 0.8 * corr)
{
isValid = false;
break;
}
}
if (!isValid)
{
continue;
}
if (corr > _corrThreshold)
{
deisotopedMassList.Add(monoIsotopicMass);
}
}
remainingPeakList.RemoveFirst();
}
private void FindMatchedIons(double peakMz, List <Peak> peakList, ref List <Ms2Match> matchList)
{
for (var charge = _maxCharge; charge >= _minCharge; charge--)
{
// check whether next isotope peak exists
var nextIsotopeMz = peakMz + Constants.C13MinusC12 / charge;
var nextIsotopePeak = PeakListUtils.FindPeak(peakList, nextIsotopeMz, _tolerance);
if (nextIsotopePeak == null)
{
continue;
}
matchList.Add(new Ms2Match(peakMz, charge));
}
}
public void TestIcrTools()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
const string icrToolsPath = @"H:\Research\Yufeng\TopDownYufeng\ICRTools\yufeng_column_test2_Isos.csv";
if (!File.Exists(icrToolsPath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, icrToolsPath);
}
const double fitScoreThreshold = 0.15;
var icrToolsparser = new TsvFileParser(icrToolsPath, ',');
var monoMassArr = icrToolsparser.GetData("monoisotopic_mw").Select(Convert.ToDouble).ToArray();
var scanArray = icrToolsparser.GetData("scan_num").Select(s => Convert.ToInt32(s)).ToArray();
var fitArray = icrToolsparser.GetData("fit").Select(Convert.ToDouble).ToArray();
var peakList = fitArray.Where(fit => fit < fitScoreThreshold).Select((fit, i) => new Peak(monoMassArr[i], scanArray[i])).OrderBy(p => p.Mz).ToList();
Console.WriteLine("FitScoreThreshold: {0}", fitScoreThreshold);
Console.WriteLine("NumFeatures: {0}", peakList.Count);
//var peakList = monoMassArr.Select((mass, i) => new Peak(mass, scanArray[i])).OrderBy(p => p.Mz).ToList();
var massTolerance = new Tolerance(10);
const double scanTolerance = 10.0;
const string resultFilePath = @"H:\Research\Yufeng\TopDownYufeng\M1_V31\yufeng_column_test2_IcTda.tsv";
if (!File.Exists(resultFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, resultFilePath);
}
var parser = new TsvFileParser(resultFilePath);
var qValues = parser.GetData("QValue").Select(Convert.ToDouble).ToArray();
var masses = parser.GetData("Mass").Select(Convert.ToDouble).ToArray();
var scans = parser.GetData("Scan").Select(s => Convert.ToInt32(s)).ToArray();
var compositions = parser.GetData("Composition").ToArray();
var numTotal = 0;
var numMatch = 0;
var compSet = new HashSet <string>();
var matchedCompSet = new HashSet <string>();
for (var i = 0; i < qValues.Length; i++)
{
if (qValues[i] > 0.01)
{
break;
}
var mass = masses[i];
var scan = scans[i];
var matches = PeakListUtils.FindAllPeaks(peakList, mass, massTolerance)
.Where(p => p.Intensity >= scan - scanTolerance && p.Intensity <= scan + scanTolerance);
if (matches.Any())
{
++numMatch;
matchedCompSet.Add(compositions[i]);
}
++numTotal;
compSet.Add(compositions[i]);
}
Console.WriteLine("PrSMs: {0} / {1} = {2}", numMatch, numTotal, numMatch / (float)numTotal);
Console.WriteLine("Compositions: {0} / {1} = {2}", matchedCompSet.Count, compSet.Count, matchedCompSet.Count / (float)compSet.Count);
}
private void SetLcMsMatches(double peakMz, int scanNum, IList <Peak> precursorSpecWindow, IList <Peak> nextMs1SpecWindow)
{
var xicThisPeak = _run.GetPrecursorExtractedIonChromatogram(peakMz, _tolerance, scanNum);
if (xicThisPeak.Count < 2)
{
return;
}
for (var charge = _maxCharge; charge >= _minCharge; charge--)
{
// check whether next isotope peak exists
var nextIsotopeMz = peakMz + Constants.C13MinusC12 / charge;
var xicNextIsotope = _run.GetPrecursorExtractedIonChromatogram(nextIsotopeMz, _tolerance, scanNum);
if (!xicNextIsotope.Any())
{
continue;
}
if (xicThisPeak.GetCorrelation(xicNextIsotope) < _mostAbundantPlusOneIsotopeCorrThreshold)
{
continue;
}
var mostAbundantIsotopeMass = (peakMz - Constants.Proton) * charge;
var averagineIsoEnv = Averagine.GetIsotopomerEnvelope(mostAbundantIsotopeMass);
var approxMostAbundantIsotopeIndex = averagineIsoEnv.MostAbundantIsotopeIndex;
var monoIsotopicMass = mostAbundantIsotopeMass - approxMostAbundantIsotopeIndex * Constants.C13MinusC12;
// Isotope correlation
var averagineIsotopeProfile = Averagine.GetTheoreticalIsotopeProfile(monoIsotopicMass, charge);
var precursorIsotopeCorr = precursorSpecWindow != null?PeakListUtils.GetPearsonCorrelation(precursorSpecWindow, averagineIsotopeProfile, _comparer) : 0;
var nextMs1IsotopeCorr = nextMs1SpecWindow != null?PeakListUtils.GetPearsonCorrelation(nextMs1SpecWindow, averagineIsotopeProfile, _comparer) : 0;
var isotopeCorr = Math.Max(precursorIsotopeCorr, nextMs1IsotopeCorr);
if (isotopeCorr < _isotopeCorrThresholdThreshold)
{
continue;
}
if (_chargeCorrThresholdThreshold > 0.0)
{
var mzChargePlusOne = Ion.GetIsotopeMz(monoIsotopicMass, charge + 1, approxMostAbundantIsotopeIndex);
var xicPlusOneCharge = _run.GetPrecursorExtractedIonChromatogram(mzChargePlusOne, _tolerance, scanNum);
var corrPlusOneCharge = xicPlusOneCharge.Count >= 3 ? xicThisPeak.GetCorrelation(xicPlusOneCharge) : 0;
double corrMinusOneCharge;
if (charge > 1)
{
var mzChargeMinusOne = Ion.GetIsotopeMz(monoIsotopicMass, charge - 1, approxMostAbundantIsotopeIndex);
var xicMinusOneCharge = _run.GetPrecursorExtractedIonChromatogram(mzChargeMinusOne, _tolerance, scanNum);
corrMinusOneCharge = xicMinusOneCharge.Count >= 3 ? xicThisPeak.GetCorrelation(xicMinusOneCharge) : 0;
}
else
{
corrMinusOneCharge = 0.0;
}
var chargeCorr = Math.Max(corrPlusOneCharge, corrMinusOneCharge);
if (chargeCorr < _chargeCorrThresholdThreshold)
{
continue;
}
}
_lcMsMatchMap.SetMatches(monoIsotopicMass, xicThisPeak[0].ScanNum, xicThisPeak[xicThisPeak.Count - 1].ScanNum);
}
}