public static void WritePsmCrossToTsv(List <CrosslinkSpectralMatch> items, string filePath, int writeType)
{
if (items.Count == 0)
{
return;
}
using (StreamWriter output = new StreamWriter(filePath))
{
string header = "";
switch (writeType)
{
case 1:
header = CrosslinkSpectralMatch.GetTabSepHeaderSingle();
break;
case 2:
header = CrosslinkSpectralMatch.GetTabSepHeaderCross();
break;
default:
break;
}
output.WriteLine(header);
foreach (var heh in items)
{
output.WriteLine(heh.ToString());
}
}
}
public static void XlTestGenerateIntensityRanks()
{
double[] intensity = new double[] { 1.1, 1.1, 0.5, 3.2, 0.5, 6.0 };
int[] rank = CrosslinkSpectralMatch.GenerateIntensityRanks(intensity);
int[] Rank = new int[] { 4, 3, 6, 2, 5, 1 };
Assert.AreEqual(rank, Rank);
}
public static void XlTestXlPosCal()
{
var prot = new Protein("MNNNKQQQQ", null);
List <DigestionMotif> motifs = new List <DigestionMotif> {
new DigestionMotif("K", null, 1, null)
};
Protease protease = new Protease("New Custom Protease", CleavageSpecificity.Full, null, null, motifs);
ProteaseDictionary.Dictionary.Add(protease.Name, protease);
DigestionParams digestionParams = new DigestionParams(protease: protease.Name, minPeptideLength: 1, initiatorMethionineBehavior: InitiatorMethionineBehavior.Retain);
List <Modification> variableModifications = new List <Modification>();
var ye = prot.Digest(digestionParams, new List <Modification>(), variableModifications).ToList();
var pep = ye[0];
Assert.AreEqual(pep.BaseSequence, "MNNNK");
Crosslinker crosslinker = GlobalVariables.Crosslinkers.Where(p => p.CrosslinkerName == "DSS").First();
Assert.AreEqual(crosslinker.CrosslinkerModSites, "K");
Assert.AreEqual(Residue.GetResidue(crosslinker.CrosslinkerModSites).MonoisotopicMass, 128.09496301518999, 1e-9);
var n = pep.Fragment(DissociationType.HCD, FragmentationTerminus.N);
var c = pep.Fragment(DissociationType.HCD, FragmentationTerminus.C);
Assert.AreEqual(n.Count(), 4);
Assert.AreEqual(c.Count(), 4);
Assert.AreEqual(c.First().NeutralMass, 146.10552769899999, 1e-6);
var x = CrosslinkSpectralMatch.GetPossibleCrosslinkerModSites(crosslinker.CrosslinkerModSites.ToCharArray(), pep).ToArray();
Assert.AreEqual(x[0], 5);
var pep2 = ye[2];
Assert.AreEqual("MNNNKQQQQ", pep2.BaseSequence);
var n2 = pep2.Fragment(DissociationType.HCD, FragmentationTerminus.N);
var c2 = pep2.Fragment(DissociationType.HCD, FragmentationTerminus.C);
Assert.AreEqual(n2.Count(), 8);
Assert.AreEqual(c2.Count(), 8);
var x2 = CrosslinkSpectralMatch.GetPossibleCrosslinkerModSites(crosslinker.CrosslinkerModSites.ToCharArray(), pep2).ToArray();
Assert.AreEqual(x2[0], 5);
//Test crosslinker with multiple types of mod
var protSTC = new Protein("GASTACK", null);
var peps = protSTC.Digest(digestionParams, new List <Modification>(), variableModifications).ToList();
var pepSTC = peps[0];
Assert.AreEqual(pepSTC.BaseSequence, "GASTACK");
Crosslinker crosslinker2 = new Crosslinker("ST", "C", "crosslinkerSTC", false, -18.01056, 0, 0, 0, 0, 0, 0);
string crosslinkerModSitesAll = new string((crosslinker2.CrosslinkerModSites + crosslinker2.CrosslinkerModSites2).ToCharArray().Distinct().ToArray());
Assert.AreEqual(crosslinkerModSitesAll, "STC");
}
//Calculate the FDR of crosslinked peptide FP/TP
private void DoCrosslinkFdrAnalysis(List <CrosslinkSpectralMatch> csms)
{
int cumulativeTarget = 0;
int cumulativeDecoy = 0;
for (int i = 0; i < csms.Count; i++)
{
var csm = csms[i];
if (csm.IsDecoy || csm.BetaPeptide.IsDecoy)
{
cumulativeDecoy++;
}
else
{
cumulativeTarget++;
}
double qValue = Math.Min(1, (double)cumulativeDecoy / cumulativeTarget);
csm.SetFdrValues(cumulativeTarget, cumulativeDecoy, qValue, 0, 0, 0, 0, 0);
}
double qValueThreshold = 1.0;
for (int i = csms.Count - 1; i >= 0; i--)
{
CrosslinkSpectralMatch csm = csms[i];
// threshold q-values
if (csm.FdrInfo.QValue > qValueThreshold)
{
csm.FdrInfo.QValue = qValueThreshold;
}
else if (csm.FdrInfo.QValue < qValueThreshold)
{
qValueThreshold = csm.FdrInfo.QValue;
}
}
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
MyTaskResults = new MyTaskResults(this);
List <CrosslinkSpectralMatch> allPsms = new List <CrosslinkSpectralMatch>();
LoadModifications(taskId, out var variableModifications, out var fixedModifications, out var localizeableModificationTypes);
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, XlSearchParameters.DecoyType, localizeableModificationTypes, CommonParameters);
var crosslinker = XlSearchParameters.Crosslinker;
MyFileManager myFileManager = new MyFileManager(true);
var fileSpecificCommonParams = fileSettingsList.Select(b => SetAllFileSpecificCommonParams(CommonParameters, b));
HashSet <DigestionParams> ListOfDigestionParams = new HashSet <DigestionParams>(fileSpecificCommonParams.Select(p => p.DigestionParams));
int completedFiles = 0;
object indexLock = new object();
object psmLock = new object();
Status("Searching files...", taskId);
ProseCreatedWhileRunning.Append("The following crosslink discovery were used: ");
ProseCreatedWhileRunning.Append("crosslinker name = " + crosslinker.CrosslinkerName + "; ");
ProseCreatedWhileRunning.Append("crosslinker type = " + crosslinker.Cleavable + "; ");
ProseCreatedWhileRunning.Append("crosslinker mass = " + crosslinker.TotalMass + "; ");
ProseCreatedWhileRunning.Append("crosslinker modification site(s) = " + crosslinker.CrosslinkerModSites + "; ");
ProseCreatedWhileRunning.Append("protease = " + CommonParameters.DigestionParams.Protease + "; ");
ProseCreatedWhileRunning.Append("maximum missed cleavages = " + CommonParameters.DigestionParams.MaxMissedCleavages + "; ");
ProseCreatedWhileRunning.Append("minimum peptide length = " + CommonParameters.DigestionParams.MinPeptideLength + "; ");
ProseCreatedWhileRunning.Append(CommonParameters.DigestionParams.MaxPeptideLength == int.MaxValue ?
"maximum peptide length = unspecified; " :
"maximum peptide length = " + CommonParameters.DigestionParams.MaxPeptideLength + "; ");
ProseCreatedWhileRunning.Append("initiator methionine behavior = " + CommonParameters.DigestionParams.InitiatorMethionineBehavior + "; ");
ProseCreatedWhileRunning.Append("max modification isoforms = " + CommonParameters.DigestionParams.MaxModificationIsoforms + "; ");
ProseCreatedWhileRunning.Append("fixed modifications = " + string.Join(", ", fixedModifications.Select(m => m.IdWithMotif) + "; "));
ProseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.IdWithMotif)) + "; ");
ProseCreatedWhileRunning.Append("parent mass tolerance(s) = " + CommonParameters.PrecursorMassTolerance + "; ");
ProseCreatedWhileRunning.Append("product mass tolerance = " + CommonParameters.ProductMassTolerance + "; ");
ProseCreatedWhileRunning.Append("The combined search database contained " + proteinList.Count + " total entries including " + proteinList.Where(p => p.IsContaminant).Count() + " contaminant sequences. ");
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
var origDataFile = currentRawFileList[spectraFileIndex];
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
var thisId = new List <string> {
taskId, "Individual Spectra Files", origDataFile
};
NewCollection(Path.GetFileName(origDataFile), thisId);
Status("Loading spectra file...", thisId);
MsDataFile myMsDataFile = myFileManager.LoadFile(origDataFile, combinedParams);
Status("Getting ms2 scans...", thisId);
Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass = GetMs2Scans(myMsDataFile, origDataFile, combinedParams).OrderBy(b => b.PrecursorMass).ToArray();
CrosslinkSpectralMatch[] newPsms = new CrosslinkSpectralMatch[arrayOfMs2ScansSortedByMass.Length];
for (int currentPartition = 0; currentPartition < CommonParameters.TotalPartitions; currentPartition++)
{
List <PeptideWithSetModifications> peptideIndex = null;
List <Protein> proteinListSubset = proteinList.GetRange(currentPartition * proteinList.Count() / combinedParams.TotalPartitions, ((currentPartition + 1) * proteinList.Count() / combinedParams.TotalPartitions) - (currentPartition * proteinList.Count() / combinedParams.TotalPartitions));
Status("Getting fragment dictionary...", new List <string> {
taskId
});
var indexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, null, currentPartition, UsefulProteomicsDatabases.DecoyType.Reverse, combinedParams, 30000.0, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), new List <string> {
taskId
});
List <int>[] fragmentIndex = null;
List <int>[] precursorIndex = null;
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, ref precursorIndex, proteinList, taskId);
Status("Searching files...", taskId);
new CrosslinkSearchEngine(newPsms, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, currentPartition, combinedParams, crosslinker,
XlSearchParameters.RestrictToTopNHits, XlSearchParameters.CrosslinkSearchTopNum, XlSearchParameters.XlQuench_H2O,
XlSearchParameters.XlQuench_NH2, XlSearchParameters.XlQuench_Tris, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + CommonParameters.TotalPartitions + "!", thisId));
if (GlobalVariables.StopLoops)
{
break;
}
}
allPsms.AddRange(newPsms.Where(p => p != null));
completedFiles++;
ReportProgress(new ProgressEventArgs(completedFiles / currentRawFileList.Count, "Searching...", new List <string> {
taskId, "Individual Spectra Files"
}));
}
ReportProgress(new ProgressEventArgs(100, "Done with all searches!", new List <string> {
taskId, "Individual Spectra Files"
}));
allPsms = allPsms.OrderByDescending(p => p.XLTotalScore).ToList();
var allPsmsXL = allPsms.Where(p => p.CrossType == PsmCrossType.Cross).ToList();
// inter-crosslinks; different proteins are linked
var interCsms = allPsmsXL.Where(p => !p.ProteinAccession.Equals(p.BetaPeptide.ProteinAccession)).ToList();
foreach (var item in interCsms)
{
item.CrossType = PsmCrossType.Inter;
}
// intra-crosslinks; crosslinks within a protein
var intraCsms = allPsmsXL.Where(p => p.ProteinAccession.Equals(p.BetaPeptide.ProteinAccession)).ToList();
foreach (var item in intraCsms)
{
item.CrossType = PsmCrossType.Intra;
}
// calculate FDR
DoCrosslinkFdrAnalysis(interCsms);
DoCrosslinkFdrAnalysis(intraCsms);
SingleFDRAnalysis(allPsms, new List <string> {
taskId
});
// calculate protein crosslink residue numbers
foreach (var csm in allPsmsXL)
{
// alpha peptide crosslink residue in the protein
csm.XlProteinPos = csm.OneBasedStartResidueInProtein.Value + csm.LinkPositions[0] - 1;
// beta crosslink residue in protein
csm.BetaPeptide.XlProteinPos = csm.BetaPeptide.OneBasedStartResidueInProtein.Value + csm.BetaPeptide.LinkPositions[0] - 1;
}
// write interlink CSMs
if (interCsms.Any())
{
string file = Path.Combine(OutputFolder, "XL_Interlinks.tsv");
WritePsmCrossToTsv(interCsms, file, 2);
FinishedWritingFile(file, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target inter-crosslinks within 1% FDR: " + interCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
if (XlSearchParameters.WriteOutputForPercolator)
{
var interPsmsXLPercolator = interCsms.Where(p => p.Score >= 2 && p.BetaPeptide.Score >= 2).OrderBy(p => p.ScanNumber).ToList();
WriteCrosslinkToTxtForPercolator(interPsmsXLPercolator, OutputFolder, "XL_Interlinks_Percolator", crosslinker, new List <string> {
taskId
});
}
// write intralink CSMs
if (intraCsms.Any())
{
string file = Path.Combine(OutputFolder, "XL_Intralinks.tsv");
WritePsmCrossToTsv(intraCsms, file, 2);
FinishedWritingFile(file, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target intra-crosslinks within 1% FDR: " + intraCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
if (XlSearchParameters.WriteOutputForPercolator)
{
var intraPsmsXLPercolator = intraCsms.Where(p => p.Score >= 2 && p.BetaPeptide.Score >= 2).OrderBy(p => p.ScanNumber).ToList();
WriteCrosslinkToTxtForPercolator(intraPsmsXLPercolator, OutputFolder, "XL_Intralinks_Percolator", crosslinker, new List <string> {
taskId
});
}
// write single peptides
var singlePsms = allPsms.Where(p => p.CrossType == PsmCrossType.Single).ToList();
if (singlePsms.Any())
{
string writtenFileSingle = Path.Combine(OutputFolder, "SinglePeptides" + ".tsv");
WritePsmCrossToTsv(singlePsms, writtenFileSingle, 1);
FinishedWritingFile(writtenFileSingle, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target single peptides within 1% FDR: " + singlePsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write loops
var loopPsms = allPsms.Where(p => p.CrossType == PsmCrossType.Loop).ToList();
if (loopPsms.Any())
{
string writtenFileLoop = Path.Combine(OutputFolder, "Looplinks" + ".tsv");
WritePsmCrossToTsv(loopPsms, writtenFileLoop, 1);
FinishedWritingFile(writtenFileLoop, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target loop-linked peptides within 1% FDR: " + loopPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write deadends
var deadendPsms = allPsms.Where(p => p.CrossType == PsmCrossType.DeadEnd ||
p.CrossType == PsmCrossType.DeadEndH2O ||
p.CrossType == PsmCrossType.DeadEndNH2 ||
p.CrossType == PsmCrossType.DeadEndTris).ToList();
if (deadendPsms.Any())
{
string writtenFileDeadend = Path.Combine(OutputFolder, "Deadends" + ".tsv");
WritePsmCrossToTsv(deadendPsms, writtenFileDeadend, 1);
FinishedWritingFile(writtenFileDeadend, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target deadend peptides within 1% FDR: " + deadendPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write pepXML
if (XlSearchParameters.WritePepXml)
{
List <CrosslinkSpectralMatch> writeToXml = new List <CrosslinkSpectralMatch>();
writeToXml.AddRange(intraCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(interCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(singlePsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(loopPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(deadendPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml = writeToXml.OrderBy(p => p.ScanNumber).ToList();
foreach (var fullFilePath in currentRawFileList)
{
string fileNameNoExtension = Path.GetFileNameWithoutExtension(fullFilePath);
WritePepXML_xl(writeToXml.Where(p => p.FullFilePath == fullFilePath).ToList(), proteinList, dbFilenameList[0].FilePath, variableModifications, fixedModifications, localizeableModificationTypes, OutputFolder, fileNameNoExtension, new List <string> {
taskId
});
}
}
return(MyTaskResults);
}
public static void XlTest_BSA_DSSO()
{
//Generate parameters
var commonParameters = new CommonParameters(doPrecursorDeconvolution: false, dissociationType: DissociationType.EThcD,
scoreCutoff: 1, digestionParams: new DigestionParams(minPeptideLength: 5), precursorMassTolerance: new PpmTolerance(10));
var xlSearchParameters = new XlSearchParameters();
//Create databases contain two protein.
var proteinList = new List <Protein> {
new Protein("EKVLTSSAR", "Fake01"), new Protein("LSQKFPK", "Fake02")
};
ModificationMotif.TryGetMotif("M", out ModificationMotif motif1);
Modification mod1 = new Modification(_originalId: "Oxidation of M", _modificationType: "Common Variable", _target: motif1, _locationRestriction: "Anywhere.", _monoisotopicMass: 15.99491461957);
ModificationMotif.TryGetMotif("C", out ModificationMotif motif2);
Modification mod2 = new Modification(_originalId: "Carbamidomethyl of C", _modificationType: "Common Fixed", _target: motif2, _locationRestriction: "Anywhere.", _monoisotopicMass: 57.02146372068994);
var variableModifications = new List <Modification>()
{
mod1
};
var fixedModifications = new List <Modification>()
{
mod2
};
var localizeableModifications = new List <Modification>();
//Run index engine
var indexEngine = new IndexingEngine(proteinList, variableModifications, fixedModifications, null, 1, DecoyType.Reverse, commonParameters, 30000, false, new List <FileInfo>(), new List <string>());
var indexResults = (IndexingResults)indexEngine.Run();
var indexedFragments = indexResults.FragmentIndex.Where(p => p != null).SelectMany(v => v).ToList();
Assert.AreEqual(82, indexedFragments.Count);
Assert.AreEqual(3, indexResults.PeptideIndex.Count);
//Get MS2 scans.
var myMsDataFile = new XLTestDataFile();
var listOfSortedms2Scans = MetaMorpheusTask.GetMs2Scans(myMsDataFile, null, new CommonParameters()).OrderBy(b => b.PrecursorMass).ToArray();
//Generate crosslinker, which is DSSO here.
Crosslinker crosslinker = GlobalVariables.Crosslinkers.Where(p => p.CrosslinkerName == "DSSO").First();
CrosslinkSpectralMatch[] possiblePsms = new CrosslinkSpectralMatch[listOfSortedms2Scans.Length];
new CrosslinkSearchEngine(possiblePsms, listOfSortedms2Scans, indexResults.PeptideIndex, indexResults.FragmentIndex, 0, commonParameters, crosslinker, xlSearchParameters.RestrictToTopNHits, xlSearchParameters.CrosslinkSearchTopNum, xlSearchParameters.XlQuench_H2O, xlSearchParameters.XlQuench_NH2, xlSearchParameters.XlQuench_Tris, new List <string> {
}).Run();
var newPsms = possiblePsms.Where(p => p != null).ToList();
foreach (var item in newPsms)
{
item.SetFdrValues(0, 0, 0, 0, 0, 0, 0, 0, 0, false);
}
//Test newPsms
Assert.AreEqual(3, newPsms.Count);
//Test Output
var task = new XLSearchTask();
task.WritePepXML_xl(newPsms, proteinList, null, variableModifications, fixedModifications, null, TestContext.CurrentContext.TestDirectory, "pep.XML", new List <string> {
});
//Test PsmCross.XlCalculateTotalProductMasses
//var psmCrossAlpha = new CrosslinkSpectralMatch(digestedList[1], 0, 0, 0, listOfSortedms2Scans[0], commonParameters.DigestionParams, new List<MatchedFragmentIon>());
//var psmCrossBeta = new CrosslinkSpectralMatch(digestedList[2], 0, 0, 0, listOfSortedms2Scans[0], commonParameters.DigestionParams, new List<MatchedFragmentIon>());
//var linkPos = CrosslinkSpectralMatch.GetPossibleCrosslinkerModSites(crosslinker.CrosslinkerModSites.ToCharArray(), digestedList[1]);
//var productMassesAlphaList = CrosslinkedPeptide.XlGetTheoreticalFragments(DissociationType.EThcD, false, crosslinker, linkPos, digestedList[2].MonoisotopicMass, digestedList[1]);
//Assert.AreEqual(productMassesAlphaList.First().Value.Count, 50); //TO DO: The number here should be manually verified.
File.Delete(@"singlePsms.tsv");
File.Delete(@"pep.XML.pep.xml");
File.Delete(@"allPsms.tsv");
}
