public static void TestCompIons_MatchIonsScore()
{
TestDataFile t = new TestDataFile();
Tolerance productMassTolerance = new AbsoluteTolerance(0.01);
double precursorMass = 300;
//The below theoretical does not accurately represent B-Y ions
double[] sorted_theoretical_product_masses_for_this_peptide = new double[] { precursorMass + (2 * Constants.ProtonMass) - 275.1350, precursorMass + (2 * Constants.ProtonMass) - 258.127, precursorMass + (2 * Constants.ProtonMass) - 257.1244, 50, 60, 70, 147.0764, precursorMass + (2 * Constants.ProtonMass) - 147.0764, precursorMass + (2 * Constants.ProtonMass) - 70, precursorMass + (2 * Constants.ProtonMass) - 60, precursorMass + (2 * Constants.ProtonMass) - 50, 257.1244, 258.127, 275.1350 }; //{ 50, 60, 70, 147.0764, 257.1244, 258.127, 275.1350 }
List <Product> productsWithLocalizedMassDiff = new List <Product>();
foreach (double d in sorted_theoretical_product_masses_for_this_peptide)
{
productsWithLocalizedMassDiff.Add(new Product(ProductType.b, FragmentationTerminus.Both, d, 1, 1, 0));
}
CommonParameters commonParametersNoComp = new CommonParameters {
ProductMassTolerance = new AbsoluteTolerance(0.01)
};
CommonParameters commonParametersWithComp = new CommonParameters(productMassTolerance: new AbsoluteTolerance(0.01), addCompIons: true);
MsDataScan scan = t.GetOneBasedScan(2);
var scanWithMass = new Ms2ScanWithSpecificMass(scan, precursorMass.ToMz(1), 1, "", new CommonParameters());
List <MatchedFragmentIon> matchedIons = MetaMorpheusEngine.MatchFragmentIons(scanWithMass, productsWithLocalizedMassDiff, commonParametersNoComp);
List <MatchedFragmentIon> matchedCompIons = MetaMorpheusEngine.MatchFragmentIons(scanWithMass, productsWithLocalizedMassDiff, commonParametersWithComp);
matchedCompIons.AddRange(matchedIons);
// score when the mass-diff is on this residue
double localizedScore = MetaMorpheusEngine.CalculatePeptideScore(scan, matchedIons);
double scoreNormal = MetaMorpheusEngine.CalculatePeptideScore(scan, matchedIons);
double scoreComp = MetaMorpheusEngine.CalculatePeptideScore(scan, matchedCompIons);
Assert.IsTrue(scoreNormal * 2 == scoreComp && scoreComp > scoreNormal + 1);
}
public static void TestPsmHeader()
{
CommonParameters commonParameters = new CommonParameters();
PeptideWithSetModifications pepWithSetMods = new Protein(
"MQQQQQQQ",
"accession1",
"org",
new List <Tuple <string, string> > {
new Tuple <string, string>("geneNameType", "geneName")
},
new Dictionary <int, List <Modification> > {
{ 2, new List <Modification> {
new Modification("mod", "mod")
} }
},
name: "name",
fullName: "fullName",
sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(2, "P", "Q", "changed this sequence")
})
.Digest(commonParameters.DigestionParams, new List <Modification>(), new List <Modification>()).First();
MsDataFile myMsDataFile = new TestDataFile(pepWithSetMods, "quadratic");
MsDataScan scann = myMsDataFile.GetOneBasedScan(2);
Ms2ScanWithSpecificMass scan = new Ms2ScanWithSpecificMass(scann, 4, 1, null, new CommonParameters());
var theoreticalIons = new List <Product>();
pepWithSetMods.Fragment(DissociationType.HCD, FragmentationTerminus.Both, theoreticalIons);
var matchedIons = MetaMorpheusEngine.MatchFragmentIons(scan, theoreticalIons, new CommonParameters());
PeptideSpectralMatch psm = new PeptideSpectralMatch(pepWithSetMods, 1, 2, 3, scan, commonParameters, matchedIons);
psm.ResolveAllAmbiguities();
var t = psm.ToString();
var tabsepheader = PeptideSpectralMatch.GetTabSeparatedHeader();
Assert.AreEqual(psm.ToString().Count(f => f == '\t'), PeptideSpectralMatch.GetTabSeparatedHeader().Count(f => f == '\t'));
Assert.AreEqual(psm.ToString().Count(f => f == '\t'), PeptideSpectralMatch.GetTabSeparatedHeader().Count(f => f == '\t'));
Tolerance fragmentTolerance = new PpmTolerance(10);
new LocalizationEngine(new List <PeptideSpectralMatch> {
psm
}, myMsDataFile, new CommonParameters(productMassTolerance: fragmentTolerance), null, new List <string>()).Run();
Assert.AreEqual(psm.ToString().Count(f => f == '\t'), PeptideSpectralMatch.GetTabSeparatedHeader().Count(f => f == '\t'));
psm.SetFdrValues(6, 6, 6, 6, 6, 0, 0, 0);
Assert.AreEqual(psm.ToString().Count(f => f == '\t'), PeptideSpectralMatch.GetTabSeparatedHeader().Count(f => f == '\t'));
}
public static void GlyTest_GlyGetTheoreticalFragments()
{
Protein pep = new Protein("TKPREEQYNSTYR", "accession");
DigestionParams digestionParams = new DigestionParams(minPeptideLength: 7);
var aPeptideWithSetModifications = pep.Digest(digestionParams, new List <Modification>(), new List <Modification>());
string[] motifs = new string[] { "Nxs", "Nxt" };
var sites = GlycoSpectralMatch.GetPossibleModSites(aPeptideWithSetModifications.Last(), motifs);
Glycan glycan = Glycan.Struct2Glycan("(N(F)(N(H(H(N))(H(N)))))", 0);
//using (StreamWriter output = new StreamWriter(Path.Combine(TestContext.CurrentContext.TestDirectory, "GlycanFragmentions.txt")))
//{
// foreach (var product in fragmentIons)
// {
// foreach (var ion in product.Item2)
// {
// output.WriteLine(ion.Annotation + "\t" + ion.NeutralLoss.ToString() + "\t" + ion.NeutralMass.ToString());
// }
// }
//}
CommonParameters commonParameters = new CommonParameters(deconvolutionMassTolerance: new PpmTolerance(20), trimMsMsPeaks: false);
string filePath = Path.Combine(TestContext.CurrentContext.TestDirectory, @"GlycoTestData/Glyco_3383.mgf"); //"25170.mgf"
MyFileManager myFileManager = new MyFileManager(true);
var msDataFile = myFileManager.LoadFile(filePath, commonParameters);
var listOfSortedms2Scans = MetaMorpheusTask.GetMs2Scans(msDataFile, filePath, commonParameters).ToArray();
var glycanMod = Glycan.NGlycanToModification(glycan);
var glycopep = GlycoPeptides.GenerateGlycopeptide(sites[0], aPeptideWithSetModifications.Last(), glycan);
List <Product> fragmentIons = new List <Product>();
glycopep.Fragment(DissociationType.HCD, FragmentationTerminus.Both, fragmentIons);
var glycanYIons = GlycoPeptides.GetGlycanYIons(listOfSortedms2Scans[0].PrecursorMass, glycan);
var matchedGlycanYIons = MetaMorpheusEngine.MatchFragmentIons(listOfSortedms2Scans[0], glycanYIons, commonParameters);
Assert.AreEqual(matchedGlycanYIons.Count, 14);
//TO DO: The neutroloss is not annotated well.
var matchedFragmentIons = MetaMorpheusEngine.MatchFragmentIons(listOfSortedms2Scans[0], fragmentIons, commonParameters);
var coreIons = GlycoPeptides.ScanGetTrimannosylCore(matchedFragmentIons, glycan);
Assert.AreEqual(coreIons.Count, 6);
var filter = GlycoPeptides.ScanTrimannosylCoreFilter(matchedFragmentIons, glycan);
Assert.AreEqual(filter, true);
var NGlycans = GlycanDatabase.LoadGlycan(GlobalVariables.NGlycanLocations[0], true, false);
var bestGlycans = GlycoPeptides.MatchBestGlycan(listOfSortedms2Scans[0], NGlycans.ToArray(), commonParameters).Where(p => p != null && p.Item2 >= 2).OrderByDescending(p => p.Item2).Take(100).OrderBy(p => p.Item3).ToArray();;
}
public static void SingleEventArgsTest()
{
Protein parentProteinForMatch = new Protein("MEK", null);
CommonParameters commonParameters = new CommonParameters(digestionParams: new DigestionParams(minPeptideLength: 1));
var fsp = new List <(string fileName, CommonParameters fileSpecificParameters)>();
fsp.Add(("", commonParameters));
ModificationMotif.TryGetMotif("E", out ModificationMotif motif);
List <Modification> variableModifications = new List <Modification> {
new Modification(_originalId: "21", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 21.981943)
};
List <PeptideWithSetModifications> allPeptidesWithSetModifications = parentProteinForMatch.Digest(commonParameters.DigestionParams, new List <Modification>(), variableModifications).ToList();
Assert.AreEqual(4, allPeptidesWithSetModifications.Count());
PeptideWithSetModifications ps = allPeptidesWithSetModifications.First();
PeptideWithSetModifications pepWithSetModsForSpectrum = allPeptidesWithSetModifications[1];
MsDataFile myMsDataFile = new TestDataFile(new List <PeptideWithSetModifications> {
pepWithSetModsForSpectrum
});
Tolerance fragmentTolerance = new AbsoluteTolerance(0.01);
Ms2ScanWithSpecificMass scan = new Ms2ScanWithSpecificMass(myMsDataFile.GetAllScansList().Last(), pepWithSetModsForSpectrum.MonoisotopicMass.ToMz(1), 1, null, new CommonParameters());
var theoreticalProducts = new List <Product>();
ps.Fragment(DissociationType.HCD, FragmentationTerminus.Both, theoreticalProducts);
var matchedIons = MetaMorpheusEngine.MatchFragmentIons(scan, theoreticalProducts, new CommonParameters());
PeptideSpectralMatch newPsm = new PeptideSpectralMatch(ps, 0, 0, 2, scan, commonParameters, matchedIons);
LocalizationEngine f = new LocalizationEngine(new List <PeptideSpectralMatch> {
newPsm
}, myMsDataFile, new CommonParameters(), fsp, new List <string>());
var singleEngine = new SingleEngineEventArgs(f);
Assert.That(singleEngine.MyEngine.Equals(f));
var singleFile = new SingleFileEventArgs("", new List <string>());
Assert.That(singleFile.WrittenFile.Equals(""));
var stringList = new StringListEventArgs(new List <string>());
var rr = (stringList.StringList.DefaultIfEmpty().First());
Assert.That(stringList.StringList.DefaultIfEmpty().First() == null);
}
public static void TestLocalization()
{
Protein parentProteinForMatch = new Protein("MEK", null);
CommonParameters commonParameters = new CommonParameters(digestionParams: new DigestionParams(minPeptideLength: 1));
var fsp = new List <(string fileName, CommonParameters fileSpecificParameters)>();
fsp.Add(("", commonParameters));
ModificationMotif.TryGetMotif("E", out ModificationMotif motif);
List <Modification> variableModifications = new List <Modification> {
new Modification(_originalId: "21", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 21.981943)
};
List <PeptideWithSetModifications> allPeptidesWithSetModifications = parentProteinForMatch.Digest(commonParameters.DigestionParams, new List <Modification>(), variableModifications).ToList();
Assert.AreEqual(4, allPeptidesWithSetModifications.Count());
PeptideWithSetModifications ps = allPeptidesWithSetModifications.First();
PeptideWithSetModifications pepWithSetModsForSpectrum = allPeptidesWithSetModifications[1];
MsDataFile myMsDataFile = new TestDataFile(new List <PeptideWithSetModifications> {
pepWithSetModsForSpectrum
});
Tolerance fragmentTolerance = new AbsoluteTolerance(0.01);
Ms2ScanWithSpecificMass scan = new Ms2ScanWithSpecificMass(myMsDataFile.GetAllScansList().Last(), pepWithSetModsForSpectrum.MonoisotopicMass.ToMz(1), 1, null, new CommonParameters());
var theoreticalProducts = new List <Product>();
ps.Fragment(DissociationType.HCD, FragmentationTerminus.Both, theoreticalProducts);
var matchedIons = MetaMorpheusEngine.MatchFragmentIons(scan, theoreticalProducts, new CommonParameters());
PeptideSpectralMatch newPsm = new PeptideSpectralMatch(ps, 0, 0, 2, scan, commonParameters, matchedIons);
newPsm.ResolveAllAmbiguities();
LocalizationEngine f = new LocalizationEngine(new List <PeptideSpectralMatch> {
newPsm
}, myMsDataFile, commonParameters, fsp, new List <string>());
f.Run();
// single peak matches
Assert.AreEqual(1, newPsm.MatchedFragmentIons.Where(p => p.NeutralTheoreticalProduct.ProductType == ProductType.b).Count());//including b1 now
Assert.AreEqual(1, newPsm.MatchedFragmentIons.Where(p => p.NeutralTheoreticalProduct.ProductType == ProductType.y).Count());
// when localizing, three peaks match
Assert.IsTrue(newPsm.LocalizedScores[1] > 4 && newPsm.LocalizedScores[1] < 5);//we have another matched ion
}
public static void TestCompIons_MatchIons()
{
TestDataFile t = new TestDataFile(0.0001);
Tolerance productMassTolerance = new AbsoluteTolerance(0.01);
double precursorMass = 402.18629720155;
//The below theoretical does not accurately represent B-Y ions
double[] sorted_theoretical_product_masses_for_this_peptide = new double[] { 50, 60, 70, 147.0764 - Constants.ProtonMass, 200, 215, 230, 245, precursorMass + Constants.ProtonMass - 147.0764, 258.127, 275.1350, precursorMass + (2 * Constants.ProtonMass) - 70, precursorMass + (2 * Constants.ProtonMass) - 60, precursorMass + (2 * Constants.ProtonMass) - 50 }; //{ 50, 60, 70, 147.0764, 257.1244, 258.127, 275.1350 }
List <double> matchedIonMassesT = new List <double>();
List <double> matchedDaErrorT = new List <double>();
List <double> matchedPpmErrorT = new List <double>();
List <double> matchedIonIntensityT = new List <double>();
List <double> matchedIonMassesF = new List <double>();
List <double> matchedDaErrorF = new List <double>();
List <double> matchedPpmErrorF = new List <double>();
List <double> matchedIonIntensityF = new List <double>();
List <int> matchedIonSeriesT = new List <int>();
List <int> matchedIonSeriesF = new List <int>();
MetaMorpheusEngine.MatchIonsOld(t.GetOneBasedScan(2), productMassTolerance, sorted_theoretical_product_masses_for_this_peptide, matchedIonSeriesT, matchedIonMassesT, matchedDaErrorT, matchedPpmErrorT, matchedIonIntensityT, precursorMass, ProductType.BnoB1ions, true);
MetaMorpheusEngine.MatchIonsOld(t.GetOneBasedScan(2), productMassTolerance, sorted_theoretical_product_masses_for_this_peptide, matchedIonSeriesF, matchedIonMassesF, matchedDaErrorF, matchedPpmErrorF, matchedIonIntensityF, precursorMass, ProductType.BnoB1ions, false);
//Test the number of series is doubled
Assert.IsTrue(matchedIonSeriesT.Count == matchedIonSeriesF.Count * 2);
//Test the number of ions is doubled
Assert.IsTrue(matchedIonMassesT.Count == matchedIonMassesF.Count * 2);
//Test the number of da errors is doubled
Assert.IsTrue(matchedDaErrorT.Count == matchedDaErrorF.Count * 2);
//test the number of ppm errors is doubled
Assert.IsTrue(matchedPpmErrorT.Count == matchedPpmErrorF.Count * 2);
//test the number of the intensity values is doubled
Assert.IsTrue(matchedIonIntensityT.Count == matchedIonIntensityF.Count * 2);
foreach (double d in matchedDaErrorF)
{
Assert.IsTrue(d <= 0.01);
}
foreach (double d in matchedDaErrorT)
{
Assert.IsTrue(d <= 0.01);
}
}
public static void TestCompIons_MatchIonsScore()
{
TestDataFile t = new TestDataFile();
Tolerance productMassTolerance = new AbsoluteTolerance(0.01);
double precursorMass = 300;
//The below theoretical does not accurately represent B-Y ions
double[] sorted_theoretical_product_masses_for_this_peptide = new double[] { precursorMass + (2 * Constants.protonMass) - 275.1350, precursorMass + (2 * Constants.protonMass) - 258.127, precursorMass + (2 * Constants.protonMass) - 257.1244, 50, 60, 70, 147.0764, precursorMass + (2 * Constants.protonMass) - 147.0764, precursorMass + (2 * Constants.protonMass) - 70, precursorMass + (2 * Constants.protonMass) - 60, precursorMass + (2 * Constants.protonMass) - 50, 257.1244, 258.127, 275.1350 }; //{ 50, 60, 70, 147.0764, 257.1244, 258.127, 275.1350 }
List <ProductType> lp = new List <ProductType> {
ProductType.B, ProductType.Y
};
double scoreT = MetaMorpheusEngine.CalculatePeptideScore(t.GetOneBasedScan(2), productMassTolerance, sorted_theoretical_product_masses_for_this_peptide, precursorMass, new List <DissociationType> {
DissociationType.HCD
}, true, 0);
double scoreF = MetaMorpheusEngine.CalculatePeptideScore(t.GetOneBasedScan(2), productMassTolerance, sorted_theoretical_product_masses_for_this_peptide, precursorMass, new List <DissociationType> {
DissociationType.HCD
}, false, 0);
Assert.IsTrue(scoreT == scoreF * 2 && scoreT > scoreF + 1);
}
public static void AddCompIonsMzOutput()
{
PeptideWithSetModifications pwsm = new PeptideWithSetModifications("ASDFASDF", null);
Ms2ScanWithSpecificMass testScan = MetaMorpheusTask.GetMs2Scans(new TestDataFile(pwsm), null, new CommonParameters()).OrderBy(b => b.PrecursorMass).First();
CommonParameters cp = new CommonParameters(addCompIons: true);
List <Product> theoreticalIons = new List <Product>();
pwsm.Fragment(cp.DissociationType, FragmentationTerminus.Both, theoreticalIons);
List <MatchedFragmentIon> matchedIons = MetaMorpheusEngine.MatchFragmentIons(testScan, theoreticalIons, cp);
//check that the matchedIons have m/z values that are similar to their neutral mass.
//There was an "issue" where the saved m/z was the original experimental peak (which is the complementary of the added ion).
//A fix was introduced to save a "fake" m/z for the added ion
foreach (MatchedFragmentIon ion in matchedIons)
{
Assert.IsTrue(ion.NeutralTheoreticalProduct.NeutralMass < ion.Mz);
Assert.IsTrue(ion.NeutralTheoreticalProduct.NeutralMass + 2 > ion.Mz);
}
}
public static Tuple <int, double, double>[] MatchBestGlycan(Ms2ScanWithSpecificMass theScan, Glycan[] glycans, CommonParameters commonParameters)
{
Tuple <int, double, double>[] tuples = new Tuple <int, double, double> [glycans.Length]; //Tuple<id, Yion matched score, glycan mass>
//TO DO: Parallel this?
for (int i = 0; i < glycans.Length; i++)
{
if (theScan.PrecursorMass - (double)glycans[i].Mass / 1E5 < 350) //Filter large glycans
{
continue;
}
List <Product> YIons = GetGlycanYIons(theScan.PrecursorMass, glycans[i]);
List <MatchedFragmentIon> matchedFragmentIons = MetaMorpheusEngine.MatchFragmentIons(theScan, YIons, commonParameters);
if (ScanTrimannosylCoreFilter(matchedFragmentIons, glycans[i]))
{
var score = MetaMorpheusEngine.CalculatePeptideScore(theScan.TheScan, matchedFragmentIons);
tuples[i] = new Tuple <int, double, double>(i, score, (double)glycans[i].Mass / 1E5);
}
}
return(tuples);
}
public DataPointAquisitionResults(
MetaMorpheusEngine dataPointAcquisitionEngine,
List <PeptideSpectralMatch> psms,
List <LabeledDataPoint> ms1List,
List <LabeledDataPoint> ms2List,
int numMs1MassChargeCombinationsConsidered,
int numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks,
int numMs2MassChargeCombinationsConsidered,
int numMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks)
: base(dataPointAcquisitionEngine)
{
Psms = psms;
Ms1List = ms1List;
Ms2List = ms2List;
Ms1InfoTh = Ms1List.Select(b => b.ExperimentalMz - b.TheoreticalMz).MeanStandardDeviation();
Ms2InfoTh = Ms2List.Select(b => b.ExperimentalMz - b.TheoreticalMz).MeanStandardDeviation();
Ms1InfoPpm = Ms1List.Select(b => (b.ExperimentalMz - b.TheoreticalMz) / b.TheoreticalMz).MeanStandardDeviation();
Ms2InfoPpm = Ms2List.Select(b => (b.ExperimentalMz - b.TheoreticalMz) / b.TheoreticalMz).MeanStandardDeviation();
NumMs1MassChargeCombinationsConsidered = numMs1MassChargeCombinationsConsidered;
NumMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks = numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks;
NumMs2MassChargeCombinationsConsidered = numMs2MassChargeCombinationsConsidered;
NumMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks = numMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks;
var precursorErrors = psms.Select(p => (p.ScanPrecursorMass - p.PeptideMonisotopicMass.Value) / p.PeptideMonisotopicMass.Value * 1e6).ToList();
PsmPrecursorIqrPpmError = Statistics.InterquartileRange(precursorErrors);
PsmPrecursorMedianPpmError = Statistics.Median(precursorErrors);
var productErrors = psms.Where(p => p.MatchedFragmentIons != null).SelectMany(p => p.MatchedFragmentIons).Select(p => (p.Mz.ToMass(p.Charge) - p.NeutralTheoreticalProduct.NeutralMass) / p.NeutralTheoreticalProduct.NeutralMass * 1e6).ToList();
PsmProductIqrPpmError = Statistics.InterquartileRange(productErrors);
PsmProductMedianPpmError = Statistics.Median(productErrors);
}
public DataPointAquisitionResults(
MetaMorpheusEngine dataPointAcquisitionEngine,
List <LabeledMs1DataPoint> ms1List,
List <LabeledMs2DataPoint> ms2List,
int numMs1MassChargeCombinationsConsidered,
int numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks,
int numMs2MassChargeCombinationsConsidered,
int numMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks)
: base(dataPointAcquisitionEngine)
{
Ms1List = ms1List;
Ms2List = ms2List;
Ms1InfoTh = Ms1List.Select(b => b.LabelTh).MeanStandardDeviation();
Ms2InfoTh = Ms2List.Select(b => b.LabelTh).MeanStandardDeviation();
Ms1InfoPpm = Ms1List.Select(b => b.LabelPpm).MeanStandardDeviation();
Ms2InfoPpm = Ms2List.Select(b => b.LabelPpm).MeanStandardDeviation();
NumMs1MassChargeCombinationsConsidered = numMs1MassChargeCombinationsConsidered;
NumMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks = numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks;
NumMs2MassChargeCombinationsConsidered = numMs2MassChargeCombinationsConsidered;
NumMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks = numMs2MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks;
}
public XLSearchResults(List <Tuple <PsmCross, PsmCross> > xlpsms, MetaMorpheusEngine searchParams) : base(searchParams)
{
XLPsms = xlpsms;
}
public LocalizationEngineResults(MetaMorpheusEngine s) : base(s)
{
}
public GptmdResults(MetaMorpheusEngine s, Dictionary <string, HashSet <Tuple <int, Modification> > > mods, int modsAdded) : base(s)
{
Mods = mods;
ModsAdded = modsAdded;
}
public static void OGlycoTest_Localization2()
{
//There may have a bug that MM cannot identify Peptide modified with (HexNAc), This is to test and find the bug.
//Get glycanBox
var glycanBox = OGlycanBoxes[0];
//Get unmodified peptide, products, allPossible modPos and all boxes.
Protein protein = new Protein("AATVGSLAGQPLQER", "P16150");
var peptide = protein.Digest(new DigestionParams(), new List <Modification>(), new List <Modification>()).First();
List <Product> products = new List <Product>();
peptide.Fragment(DissociationType.ETD, FragmentationTerminus.Both, products);
int[] modPos = GlycoSpectralMatch.GetPossibleModSites(peptide, new string[] { "S", "T" }).OrderBy(p => p).ToArray();
var boxes = GlycanBox.BuildChildOGlycanBoxes(glycanBox.NumberOfMods, glycanBox.ModIds).ToArray();
//Load scan.
CommonParameters commonParameters = new CommonParameters(dissociationType: DissociationType.ETD, trimMsMsPeaks: false);
string spectraFile = Path.Combine(TestContext.CurrentContext.TestDirectory, @"GlycoTestData\181217_Fusion_(LC2)_NewObj_Serum_deSA_Jacalin_HRM_4h_ETD_HCD_DDA_mz(400_1200)_21707.mgf");
var file = new MyFileManager(true).LoadFile(spectraFile, commonParameters);
var scans = MetaMorpheusTask.GetMs2Scans(file, spectraFile, commonParameters).ToArray();
//Known peptideWithMod match.
var peptideWithMod = GlycoPeptides.OGlyGetTheoreticalPeptide(new int[1] {
4
}, peptide, glycanBox);
Assert.That(peptideWithMod.FullSequence == "AAT[O-Glycosylation:N1 on X]VGSLAGQPLQER");
//List<Product> knownProducts = peptideWithMod.Fragment(DissociationType.EThcD, FragmentationTerminus.Both).ToList();
List <Product> knownProducts = GlycoPeptides.OGlyGetTheoreticalFragments(DissociationType.ETD, peptide, peptideWithMod);
var matchedKnownFragmentIons = MetaMorpheusEngine.MatchFragmentIons(scans.First(), knownProducts, commonParameters);
//Get hashset int
int obsPreviousFragmentCeilingMz = 0;
List <int> binsToSearch = new List <int>();
foreach (var envelope in scans.First().ExperimentalFragments)
{
// assume charge state 1 to calculate mass tolerance
double experimentalFragmentMass = envelope.MonoisotopicMass;
// get theoretical fragment bins within mass tolerance
int obsFragmentFloorMass = (int)Math.Floor((commonParameters.ProductMassTolerance.GetMinimumValue(experimentalFragmentMass)) * 1000);
int obsFragmentCeilingMass = (int)Math.Ceiling((commonParameters.ProductMassTolerance.GetMaximumValue(experimentalFragmentMass)) * 1000);
// prevents double-counting peaks close in m/z and lower-bound out of range exceptions
if (obsFragmentFloorMass < obsPreviousFragmentCeilingMz)
{
obsFragmentFloorMass = obsPreviousFragmentCeilingMz;
}
obsPreviousFragmentCeilingMz = obsFragmentCeilingMass + 1;
// search mass bins within a tolerance
for (int fragmentBin = obsFragmentFloorMass; fragmentBin <= obsFragmentCeilingMass; fragmentBin++)
{
binsToSearch.Add(fragmentBin);
}
}
HashSet <int> allPeaks = new HashSet <int>(binsToSearch);
//Graph Localization
LocalizationGraph localizationGraph = new LocalizationGraph(modPos, glycanBox, boxes, -1);
LocalizationGraph.LocalizeOGlycan(localizationGraph, scans.First(), commonParameters.ProductMassTolerance, products);
var allPaths = LocalizationGraph.GetAllHighestScorePaths(localizationGraph.array, localizationGraph.ChildModBoxes);
var knowPath = new int[2] {
1, 1
};
Assert.That(Enumerable.SequenceEqual(knowPath, allPaths[0]));
var local = LocalizationGraph.GetLocalizedPath(localizationGraph, allPaths.First());
Assert.That(Enumerable.SequenceEqual(local.Mods.Select(p => p.Item1), new List <int> {
4
}));
Assert.That(Enumerable.SequenceEqual(local.Mods.Select(p => p.Item2), new List <int> {
0
}));
}
public static void OGlycoTest_Localization()
{
//Get glycanBox
var glycanBox = OGlycanBoxes[19];
//Get unmodified peptide, products, allPossible modPos and all boxes.
Protein protein = new Protein("TTGSLEPSSGASGPQVSSVK", "P16150");
var peptide = protein.Digest(new DigestionParams(), new List <Modification>(), new List <Modification>()).First();
List <Product> products = new List <Product>();
peptide.Fragment(DissociationType.ETD, FragmentationTerminus.Both, products);
int[] modPos = GlycoSpectralMatch.GetPossibleModSites(peptide, new string[] { "S", "T" }).OrderBy(v => v).ToArray();
var boxes = GlycanBox.BuildChildOGlycanBoxes(3, glycanBox.ModIds).ToArray();
Assert.That(boxes.Count() == 6);
//Get Unlocal Fragment
var unlocalCost = GlycoPeptides.GetUnlocalFragment(products, modPos, glycanBox);
Assert.That(unlocalCost.Count == 4); //Basicly, the unlocal are c/z ions that don't localize glycosylation.
//Get scan
CommonParameters commonParameters = new CommonParameters(dissociationType: DissociationType.EThcD, trimMsMsPeaks: false);
string spectraFile = Path.Combine(TestContext.CurrentContext.TestDirectory, @"GlycoTestData\2019_09_16_StcEmix_35trig_EThcD25_rep1_4565.mgf");
var file = new MyFileManager(true).LoadFile(spectraFile, commonParameters);
var scans = MetaMorpheusTask.GetMs2Scans(file, spectraFile, commonParameters).ToArray();
//Known peptideWithMod match.
var peptideWithMod = GlycoPeptides.OGlyGetTheoreticalPeptide(new int[3] {
10, 2, 3
}, peptide, glycanBox);
Assert.That(peptideWithMod.FullSequence == "T[O-Glycosylation:H1N1 on X]T[O-Glycosylation:H1N1 on X]GSLEPSS[O-Glycosylation:N1 on X]GASGPQVSSVK");
List <Product> knownProducts = GlycoPeptides.OGlyGetTheoreticalFragments(DissociationType.EThcD, peptide, peptideWithMod);
var matchedKnownFragmentIons = MetaMorpheusEngine.MatchFragmentIons(scans.First(), knownProducts, commonParameters);
//Graph Localization
LocalizationGraph localizationGraph = new LocalizationGraph(modPos, glycanBox, boxes, -1);
LocalizationGraph.LocalizeOGlycan(localizationGraph, scans.First(), commonParameters.ProductMassTolerance, products);
var allPaths = LocalizationGraph.GetAllHighestScorePaths(localizationGraph.array, localizationGraph.ChildModBoxes);
var knowPath = new int[8] {
2, 4, 4, 4, 5, 5, 5, 5
};
Assert.That(Enumerable.SequenceEqual(knowPath, allPaths[0]));
//Get localized Route
var local = LocalizationGraph.GetLocalizedPath(localizationGraph, allPaths.First());
Assert.That(Enumerable.SequenceEqual(local.Mods.Select(v => v.Item1), new List <int> {
2, 3, 10
}));
Assert.That(Enumerable.SequenceEqual(local.Mods.Select(v => v.Item2), new List <int> {
1, 1, 0
}));
//Get all paths, calculate PScore and calculate position probability.
var p = scans.First().TheScan.MassSpectrum.Size *commonParameters.ProductMassTolerance.GetRange(1000).Width / scans.First().TheScan.MassSpectrum.Range.Width;
var n = knownProducts.Where(v => v.ProductType == ProductType.c || v.ProductType == ProductType.zDot).Count();
var allPathWithWeights = LocalizationGraph.GetAllPaths_CalP(localizationGraph, p, n);
Assert.That(allPathWithWeights.Count == 168);
//Calculate Site Specific Localization Probability
var y = LocalizationGraph.CalSiteSpecificLocalizationProbability(allPathWithWeights, localizationGraph.ModPos);
Assert.That(y.Count == 8);
Assert.That(y.First().Value[1].Item2 > 0.99);
}
public static void MatchInternalFragmentIons(PeptideSpectralMatch[] fileSpecificPsms, Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass, CommonParameters combinedParams, int minInternalFragmentLength)
{
//for each PSM with an ID
for (int index = 0; index < fileSpecificPsms.Length; index++)
{
PeptideSpectralMatch psm = fileSpecificPsms[index];
if (psm != null && psm.BestMatchingPeptides.Count() > 0)
{
//Get the scan
Ms2ScanWithSpecificMass scanForThisPsm = arrayOfMs2ScansSortedByMass[index];
DissociationType dissociationType = combinedParams.DissociationType == DissociationType.Autodetect ?
scanForThisPsm.TheScan.DissociationType.Value : combinedParams.DissociationType;
//Get the theoretical peptides
List <PeptideWithSetModifications> ambiguousPeptides = new List <PeptideWithSetModifications>();
List <int> notches = new List <int>();
foreach (var(Notch, Peptide) in psm.BestMatchingPeptides)
{
ambiguousPeptides.Add(Peptide);
notches.Add(Notch);
}
//get matched ions for each peptide
List <List <MatchedFragmentIon> > matchedIonsForAllAmbiguousPeptides = new List <List <MatchedFragmentIon> >();
List <Product> internalFragments = new List <Product>();
foreach (PeptideWithSetModifications peptide in ambiguousPeptides)
{
internalFragments.Clear();
peptide.FragmentInternally(combinedParams.DissociationType, minInternalFragmentLength, internalFragments);
//TODO: currently, internal and terminal ions can match to the same observed peaks (much like how b- and y-ions can match to the same peaks). Investigate if we should change that...
matchedIonsForAllAmbiguousPeptides.Add(MetaMorpheusEngine.MatchFragmentIons(scanForThisPsm, internalFragments, combinedParams));
}
//Find the max number of matched ions
int maxNumMatchedIons = matchedIonsForAllAmbiguousPeptides.Max(x => x.Count);
//remove peptides if they have fewer than max-1 matched ions, thus requiring at least two internal ions to disambiguate an ID
//if not removed, then add the matched internal ions
HashSet <PeptideWithSetModifications> PeptidesToMatchingInternalFragments = new HashSet <PeptideWithSetModifications>();
for (int peptideIndex = 0; peptideIndex < ambiguousPeptides.Count; peptideIndex++)
{
//if we should remove the theoretical, remove it
if (matchedIonsForAllAmbiguousPeptides[peptideIndex].Count + 1 < maxNumMatchedIons)
{
psm.RemoveThisAmbiguousPeptide(notches[peptideIndex], ambiguousPeptides[peptideIndex]);
}
// otherwise add the matched internal ions to the total ions
else
{
PeptideWithSetModifications currentPwsm = ambiguousPeptides[peptideIndex];
//check that we haven't already added the matched ions for this peptide
if (!PeptidesToMatchingInternalFragments.Contains(currentPwsm))
{
PeptidesToMatchingInternalFragments.Add(currentPwsm); //record that we've seen this peptide
psm.PeptidesToMatchingFragments[currentPwsm].AddRange(matchedIonsForAllAmbiguousPeptides[peptideIndex]); //add the matched ions
}
}
}
}
}
}
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);
}
public TestResults(MetaMorpheusEngine e) : base(e)
{
}