private static MassDiffAcceptor ParseSearchMode(string text)
{
MassDiffAcceptor massDiffAcceptor = null;
try
{
var split = text.Split(' ');
switch (split[1])
{
case "dot":
double[] massShifts = split[4].Split(',').Select(p => double.Parse(p, CultureInfo.InvariantCulture)).ToArray();
string newString = split[2].Replace("�", "");
double toleranceValue = double.Parse(newString, CultureInfo.InvariantCulture);
if (split[3].ToUpperInvariant().Equals("PPM"))
{
massDiffAcceptor = new DotMassDiffAcceptor(split[0], massShifts, new PpmTolerance(toleranceValue));
}
else if (split[3].ToUpperInvariant().Equals("DA"))
{
massDiffAcceptor = new DotMassDiffAcceptor(split[0], massShifts, new AbsoluteTolerance(toleranceValue));
}
break;
case "interval":
IEnumerable <DoubleRange> doubleRanges = Array.ConvertAll(split[2].Split(';'), b => new DoubleRange(double.Parse(b.Trim(new char[] { '[', ']' }).Split(',')[0],
CultureInfo.InvariantCulture), double.Parse(b.Trim(new char[] { '[', ']' }).Split(',')[1], CultureInfo.InvariantCulture)));
massDiffAcceptor = new IntervalMassDiffAcceptor(split[0], doubleRanges);
break;
case "OpenSearch":
massDiffAcceptor = new OpenSearchMode();
break;
case "daltonsAroundZero":
massDiffAcceptor = new SingleAbsoluteAroundZeroSearchMode(double.Parse(split[2], CultureInfo.InvariantCulture));
break;
case "ppmAroundZero":
massDiffAcceptor = new SinglePpmAroundZeroSearchMode(double.Parse(split[2], CultureInfo.InvariantCulture));
break;
default:
throw new MetaMorpheusException("Unrecognized search mode type: " + split[1]);
}
}
catch (Exception e)
{
throw new MetaMorpheusException("Could not parse search mode string: " + e.Message);
}
return(massDiffAcceptor);
}
public static void TestDotSearchMode()
{
var dsm1 = new DotMassDiffAcceptor("test1", new double[] { 0, 1 }, new AbsoluteTolerance(0.1));
Assert.IsTrue(dsm1.Accepts(1000, 1000) >= 0);
Assert.IsTrue(dsm1.Accepts(1000, 1000 + 0.1 / 2) >= 0);
Assert.IsFalse(dsm1.Accepts(1000, 1000 + 0.1 * 2) >= 0);
Assert.IsTrue(dsm1.Accepts(1000 + 0.1 / 2, 1000) >= 0);
Assert.IsFalse(dsm1.Accepts(1000 + 0.1 * 2, 1000) >= 0);
Assert.IsTrue(dsm1.Accepts(1000 + 1, 1000 + 0.1 / 2) >= 0);
Assert.IsFalse(dsm1.Accepts(1000 + 1, 1000 + 0.1 * 2) >= 0);
Assert.IsTrue(dsm1.Accepts(1000 + 1 + 0.1 / 2, 1000) >= 0);
Assert.IsFalse(dsm1.Accepts(1000 + 1 + 0.1 * 2, 1000) >= 0);
var theList = dsm1.GetAllowedPrecursorMassIntervals(100).ToList();
Assert.AreEqual(99.9, theList[0].allowedInterval.Minimum);
Assert.AreEqual(100.1, theList[0].allowedInterval.Maximum);
Assert.AreEqual(100.9, theList[1].allowedInterval.Minimum);
Assert.AreEqual(101.1, theList[1].allowedInterval.Maximum);
var dsm2 = new DotMassDiffAcceptor("test2", new double[] { 0, 1 }, new PpmTolerance(5));
Assert.IsTrue(dsm2.Accepts(1000, 1000) >= 0);
Assert.IsTrue(dsm2.Accepts(1000 * (1 + 5.0 / 1e6 / 1.0000001), 1000) >= 0); // FIRST VARIES WITHIN 5 PPM OF SECOND
Assert.IsTrue(dsm2.Accepts(1000 * (1 - 5.0 / 1e6 / 1.0000001), 1000) >= 0); // FIRST VARIES WITHIN 5 PPM OF SECOND
Assert.IsFalse(dsm2.Accepts(1000, 1000 * (1 - 5.0 / 1e6 / 1.0000001)) >= 0); // VERY CAREFUL
Assert.IsFalse(dsm2.Accepts(1000 * (1 + 5.0 / 1e6 * 1.0000001), 1000) >= 0); // FIRST VARIES WITHIN 5 PPM OF SECOND
Assert.IsFalse(dsm2.Accepts(1000 * (1 - 5.0 / 1e6 * 1.0000001), 1000) >= 0); // FIRST VARIES WITHIN 5 PPM OF SECOND
Assert.IsTrue(dsm2.Accepts(1000, 1000 * (1 + 5.0 / 1e6 * 1.0000001)) >= 0); // VERY CAREFUL
var theList2 = dsm2.GetAllowedPrecursorMassIntervals(1000).ToList();
Assert.IsTrue(theList2[0].allowedInterval.Contains(1000));
Assert.IsTrue(1000 * (1 + 5.0 / 1e6 / 1.0000001) < theList2[0].allowedInterval.Maximum);
Assert.IsTrue(1000 * (1 - 5.0 / 1e6 / 1.0000001) > theList2[0].allowedInterval.Minimum);
Assert.IsTrue(1000 * (1 + 5.0 / 1e6 * 1.0000001) > theList2[0].allowedInterval.Maximum);
Assert.IsTrue(1000 * (1 - 5.0 / 1e6 * 1.0000001) < theList2[0].allowedInterval.Minimum);
Assert.IsTrue(theList2[1].allowedInterval.Contains(1001));
}
private static MassDiffAcceptor ParseSearchMode(string text)
{
MassDiffAcceptor ye = null;
var split = text.Split(' ');
switch (split[1])
{
case "dot":
var massShifts = Array.ConvertAll(split[4].Split(','), Double.Parse);
var newString = split[2].Replace("�", "");
var toleranceValue = double.Parse(newString, CultureInfo.InvariantCulture);
if (split[3].ToUpperInvariant().Equals("PPM"))
{
ye = new DotMassDiffAcceptor(split[0], massShifts, new PpmTolerance(toleranceValue));
}
else if (split[3].ToUpperInvariant().Equals("DA"))
{
ye = new DotMassDiffAcceptor(split[0], massShifts, new AbsoluteTolerance(toleranceValue));
}
break;
case "interval":
IEnumerable <DoubleRange> doubleRanges = Array.ConvertAll(split[2].Split(','), b => new DoubleRange(double.Parse(b.Trim(new char[] { '[', ']' }).Split(';')[0], CultureInfo.InvariantCulture), double.Parse(b.Trim(new char[] { '[', ']' }).Split(';')[1], CultureInfo.InvariantCulture)));
ye = new IntervalMassDiffAcceptor(split[0], doubleRanges);
break;
case "OpenSearch":
ye = new OpenSearchMode();
break;
case "daltonsAroundZero":
ye = new SingleAbsoluteAroundZeroSearchMode(double.Parse(split[2], CultureInfo.InvariantCulture));
break;
case "ppmAroundZero":
ye = new SinglePpmAroundZeroSearchMode(double.Parse(split[2], CultureInfo.InvariantCulture));
break;
default:
throw new MetaMorpheusException("Could not parse search mode string");
}
return(ye);
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
LoadModifications(taskId, out var variableModifications, out var fixedModifications, out var localizeableModificationTypes);
// TODO: print error messages loading GPTMD mods
List <Modification> gptmdModifications = GlobalVariables.AllModsKnown.OfType <Modification>().Where(b => GptmdParameters.ListOfModsGptmd.Contains((b.ModificationType, b.IdWithMotif))).ToList();
IEnumerable <Tuple <double, double> > combos = LoadCombos(gptmdModifications).ToList();
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, DecoyType.Reverse, localizeableModificationTypes, CommonParameters);
List <PeptideSpectralMatch> allPsms = new List <PeptideSpectralMatch>();
var numRawFiles = currentRawFileList.Count;
// write prose settings
ProseCreatedWhileRunning.Append("The following G-PTM-D settings were used: "); 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("G-PTM-D modifications count = " + gptmdModifications.Count + "; ");
// temporary search type for writing prose
// the actual search type is technically file-specific but we don't allow file-specific notches, so it's safe to do this
MassDiffAcceptor tempSearchMode = new DotMassDiffAcceptor("", GetAcceptableMassShifts(fixedModifications, variableModifications, gptmdModifications, combos), CommonParameters.PrecursorMassTolerance);
ProseCreatedWhileRunning.Append("precursor mass tolerance(s) = {" + tempSearchMode.ToProseString() + "}; ");
ProseCreatedWhileRunning.Append("product mass tolerance = " + CommonParameters.ProductMassTolerance + ". ");
ProseCreatedWhileRunning.Append("The combined search database contained " + proteinList.Count(p => !p.IsDecoy) + " non-decoy protein entries including " + proteinList.Where(p => p.IsContaminant).Count() + " contaminant sequences. ");
// start the G-PTM-D task
Status("Running G-PTM-D...", new List <string> {
taskId
});
MyTaskResults = new MyTaskResults(this)
{
NewDatabases = new List <DbForTask>()
};
var fileSpecificCommonParams = fileSettingsList.Select(b => SetAllFileSpecificCommonParams(CommonParameters, b));
HashSet <DigestionParams> ListOfDigestionParams = new HashSet <DigestionParams>(fileSpecificCommonParams.Select(p => p.DigestionParams));
MyFileManager myFileManager = new MyFileManager(true);
object lock1 = new object();
object lock2 = new object();
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
// Stop if canceled
if (GlobalVariables.StopLoops)
{
break;
}
var origDataFile = currentRawFileList[spectraFileIndex];
// mark the file as in-progress
StartingDataFile(origDataFile, new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
MassDiffAcceptor searchMode = new DotMassDiffAcceptor("", GetAcceptableMassShifts(fixedModifications, variableModifications, gptmdModifications, combos), combinedParams.PrecursorMassTolerance);
NewCollection(Path.GetFileName(origDataFile), new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
Status("Loading spectra file...", new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
MsDataFile myMsDataFile = myFileManager.LoadFile(origDataFile, combinedParams);
Status("Getting ms2 scans...", new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass = GetMs2Scans(myMsDataFile, origDataFile, combinedParams).OrderBy(b => b.PrecursorMass).ToArray();
myFileManager.DoneWithFile(origDataFile);
PeptideSpectralMatch[] allPsmsArray = new PeptideSpectralMatch[arrayOfMs2ScansSortedByMass.Length];
new ClassicSearchEngine(allPsmsArray, arrayOfMs2ScansSortedByMass, variableModifications, fixedModifications, null, null, null, proteinList, searchMode, combinedParams, new List <string> {
taskId, "Individual Spectra Files", origDataFile
}).Run();
allPsms.AddRange(allPsmsArray.Where(p => p != null));
FinishedDataFile(origDataFile, new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files", origDataFile
}));
}
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files"
}));
allPsms = allPsms.OrderByDescending(b => b.Score)
.ThenBy(b => b.PeptideMonisotopicMass.HasValue ? Math.Abs(b.ScanPrecursorMass - b.PeptideMonisotopicMass.Value) : double.MaxValue)
.GroupBy(b => new Tuple <string, int, double?>(b.FullFilePath, b.ScanNumber, b.PeptideMonisotopicMass))
.Select(b => b.First()).ToList();
new FdrAnalysisEngine(allPsms, tempSearchMode.NumNotches, CommonParameters, new List <string> {
taskId
}).Run();
var writtenFile = Path.Combine(OutputFolder, "GPTMD_Candidates.psmtsv");
WritePsmsToTsv(allPsms, writtenFile, new Dictionary <string, int>());
FinishedWritingFile(writtenFile, new List <string> {
taskId
});
// get file-specific precursor mass tolerances for the GPTMD engine
var filePathToPrecursorMassTolerance = new Dictionary <string, Tolerance>();
for (int i = 0; i < currentRawFileList.Count; i++)
{
string filePath = currentRawFileList[i];
Tolerance fileTolerance = CommonParameters.PrecursorMassTolerance;
if (fileSettingsList[i] != null && fileSettingsList[i].PrecursorMassTolerance != null)
{
fileTolerance = fileSettingsList[i].PrecursorMassTolerance;
}
filePathToPrecursorMassTolerance.Add(filePath, fileTolerance);
}
// run GPTMD engine
var gptmdResults = (GptmdResults) new GptmdEngine(allPsms, gptmdModifications, combos, filePathToPrecursorMassTolerance, CommonParameters, new List <string> {
taskId
}).Run();
// Stop if canceled
if (GlobalVariables.StopLoops)
{
return(MyTaskResults);
}
// write GPTMD databases
if (dbFilenameList.Any(b => !b.IsContaminant))
{
List <string> databaseNames = new List <string>();
foreach (var nonContaminantDb in dbFilenameList.Where(p => !p.IsContaminant))
{
var dbName = Path.GetFileNameWithoutExtension(nonContaminantDb.FilePath);
var theExtension = Path.GetExtension(nonContaminantDb.FilePath).ToLowerInvariant();
bool compressed = theExtension.EndsWith("gz");
databaseNames.Add(compressed ? Path.GetFileNameWithoutExtension(dbName) : dbName);
}
string outputXMLdbFullName = Path.Combine(OutputFolder, string.Join("-", databaseNames) + "GPTMD.xml");
var newModsActuallyWritten = ProteinDbWriter.WriteXmlDatabase(gptmdResults.Mods, proteinList.Where(b => !b.IsDecoy && !b.IsContaminant).ToList(), outputXMLdbFullName);
FinishedWritingFile(outputXMLdbFullName, new List <string> {
taskId
});
MyTaskResults.NewDatabases.Add(new DbForTask(outputXMLdbFullName, false));
MyTaskResults.AddTaskSummaryText("Modifications added: " + newModsActuallyWritten.Select(b => b.Value).Sum());
MyTaskResults.AddTaskSummaryText("Mods types and counts:");
MyTaskResults.AddTaskSummaryText(string.Join(Environment.NewLine, newModsActuallyWritten.OrderByDescending(b => b.Value).Select(b => "\t" + b.Key + "\t" + b.Value)));
}
if (dbFilenameList.Any(b => b.IsContaminant))
{
// do NOT use this code (Path.GetFilenameWithoutExtension) because GPTMD on .xml.gz will result in .xml.xml file type being written
//string outputXMLdbFullNameContaminants = Path.Combine(OutputFolder, string.Join("-", dbFilenameList.Where(b => b.IsContaminant).Select(b => Path.GetFileNameWithoutExtension(b.FilePath))) + "GPTMD.xml");
List <string> databaseNames = new List <string>();
foreach (var contaminantDb in dbFilenameList.Where(p => p.IsContaminant))
{
var dbName = Path.GetFileName(contaminantDb.FilePath);
int indexOfFirstDot = dbName.IndexOf(".");
databaseNames.Add(dbName.Substring(0, indexOfFirstDot));
}
string outputXMLdbFullNameContaminants = Path.Combine(OutputFolder, string.Join("-", databaseNames) + "GPTMD.xml");
var newModsActuallyWritten = ProteinDbWriter.WriteXmlDatabase(gptmdResults.Mods, proteinList.Where(b => !b.IsDecoy && b.IsContaminant).ToList(), outputXMLdbFullNameContaminants);
FinishedWritingFile(outputXMLdbFullNameContaminants, new List <string> {
taskId
});
MyTaskResults.NewDatabases.Add(new DbForTask(outputXMLdbFullNameContaminants, true));
MyTaskResults.AddTaskSummaryText("Contaminant modifications added: " + newModsActuallyWritten.Select(b => b.Value).Sum());
MyTaskResults.AddTaskSummaryText("Mods types and counts:");
MyTaskResults.AddTaskSummaryText(string.Join(Environment.NewLine, newModsActuallyWritten.OrderByDescending(b => b.Value).Select(b => "\t" + b.Key + "\t" + b.Value)));
}
return(MyTaskResults);
}
public static void FdrTestMethod()
{
MassDiffAcceptor searchModes = new DotMassDiffAcceptor(null, new List <double> {
0, 1.0029
}, new PpmTolerance(5));
List <string> nestedIds = new List <string>();
Protein p = new Protein("MNKNNKNNNKNNNNK", null);
DigestionParams digestionParams = new DigestionParams();
var digested = p.Digest(digestionParams, new List <Modification>(), new List <Modification>()).ToList();
PeptideWithSetModifications pep1 = digested[0];
PeptideWithSetModifications pep2 = digested[1];
PeptideWithSetModifications pep3 = digested[2];
PeptideWithSetModifications pep4 = digested[3];
TestDataFile t = new TestDataFile(new List <PeptideWithSetModifications> {
pep1, pep2, pep3
});
MsDataScan mzLibScan1 = t.GetOneBasedScan(2);
Ms2ScanWithSpecificMass scan1 = new Ms2ScanWithSpecificMass(mzLibScan1, pep1.MonoisotopicMass.ToMz(1), 1, null, new CommonParameters());
PeptideSpectralMatch psm1 = new PeptideSpectralMatch(pep1, 0, 3, 0, scan1, digestionParams, new List <MatchedFragmentIon>());
MsDataScan mzLibScan2 = t.GetOneBasedScan(4);
Ms2ScanWithSpecificMass scan2 = new Ms2ScanWithSpecificMass(mzLibScan2, pep2.MonoisotopicMass.ToMz(1), 1, null, new CommonParameters());
PeptideSpectralMatch psm2 = new PeptideSpectralMatch(pep2, 1, 2, 1, scan2, digestionParams, new List <MatchedFragmentIon>());
MsDataScan mzLibScan3 = t.GetOneBasedScan(6);
Ms2ScanWithSpecificMass scan3 = new Ms2ScanWithSpecificMass(mzLibScan3, pep3.MonoisotopicMass.ToMz(1), 1, null, new CommonParameters());
PeptideSpectralMatch psm3 = new PeptideSpectralMatch(pep3, 0, 1, 2, scan3, digestionParams, new List <MatchedFragmentIon>());
psm3.AddOrReplace(pep4, 1, 1, true, new List <MatchedFragmentIon>(), 0);
var newPsms = new List <PeptideSpectralMatch> {
psm1, psm2, psm3
};
foreach (PeptideSpectralMatch psm in newPsms)
{
psm.ResolveAllAmbiguities();
}
FdrAnalysisEngine fdr = new FdrAnalysisEngine(newPsms, searchModes.NumNotches, new CommonParameters(), nestedIds);
fdr.Run();
Assert.AreEqual(2, searchModes.NumNotches);
Assert.AreEqual(0, newPsms[0].FdrInfo.CumulativeDecoyNotch);
Assert.AreEqual(1, newPsms[0].FdrInfo.CumulativeTargetNotch);
Assert.AreEqual(0, newPsms[1].FdrInfo.CumulativeDecoyNotch);
Assert.AreEqual(1, newPsms[1].FdrInfo.CumulativeTargetNotch);
Assert.AreEqual(0, newPsms[2].FdrInfo.CumulativeDecoyNotch);
Assert.AreEqual(1, newPsms[2].FdrInfo.CumulativeTargetNotch);
Assert.AreEqual(0, newPsms[0].FdrInfo.CumulativeDecoy);
Assert.AreEqual(1, newPsms[0].FdrInfo.CumulativeTarget);
Assert.AreEqual(0, newPsms[1].FdrInfo.CumulativeDecoy);
Assert.AreEqual(2, newPsms[1].FdrInfo.CumulativeTarget);
Assert.AreEqual(0, newPsms[2].FdrInfo.CumulativeDecoy);
Assert.AreEqual(3, newPsms[2].FdrInfo.CumulativeTarget);
}
public static void FdrTestMethod()
{
MassDiffAcceptor searchModes = new DotMassDiffAcceptor(null, new List <double> {
0, 1.0029
}, new PpmTolerance(5));
List <string> nestedIds = new List <string>();
Protein p = new Protein("MNKNNKNNNKNNNNK", null);
DigestionParams digestionParams = new DigestionParams();
var digested = p.Digest(digestionParams, new List <ModificationWithMass>(), new List <ModificationWithMass>()).ToList();
PeptideWithSetModifications pep1 = digested[0];
PeptideWithSetModifications pep2 = digested[1];
PeptideWithSetModifications pep3 = digested[2];
PeptideWithSetModifications pep4 = digested[3];
TestDataFile t = new TestDataFile(new List <PeptideWithSetModifications> {
pep1, pep2, pep3
});
CompactPeptide peptide1 = new CompactPeptide(pep1, TerminusType.None);
IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> > mzLibScan1 = t.GetOneBasedScan(2) as IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> >;
Ms2ScanWithSpecificMass scan1 = new Ms2ScanWithSpecificMass(mzLibScan1, peptide1.MonoisotopicMassIncludingFixedMods.ToMz(1), 1, null);
PeptideSpectralMatch psm1 = new PeptideSpectralMatch(peptide1, 0, 3, 0, scan1);
CompactPeptide peptide2 = new CompactPeptide(pep2, TerminusType.None);
IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> > mzLibScan2 = t.GetOneBasedScan(4) as IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> >;
Ms2ScanWithSpecificMass scan2 = new Ms2ScanWithSpecificMass(mzLibScan2, peptide2.MonoisotopicMassIncludingFixedMods.ToMz(1), 1, null);
PeptideSpectralMatch psm2 = new PeptideSpectralMatch(peptide2, 1, 2, 1, scan2);
CompactPeptide peptide3 = new CompactPeptide(pep3, TerminusType.None);
IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> > mzLibScan3 = t.GetOneBasedScan(6) as IMsDataScanWithPrecursor <IMzSpectrum <IMzPeak> >;
Ms2ScanWithSpecificMass scan3 = new Ms2ScanWithSpecificMass(mzLibScan3, peptide3.MonoisotopicMassIncludingFixedMods.ToMz(1), 1, null);
PeptideSpectralMatch psm3 = new PeptideSpectralMatch(peptide3, 0, 1, 2, scan3);
CompactPeptide peptide4 = new CompactPeptide(pep4, TerminusType.None);
psm3.AddOrReplace(peptide4, 1, 1, true);
Dictionary <CompactPeptideBase, HashSet <PeptideWithSetModifications> > matching = new Dictionary <CompactPeptideBase, HashSet <PeptideWithSetModifications> >
{
{
peptide1, new HashSet <PeptideWithSetModifications> {
pep1
}
},
{
peptide2, new HashSet <PeptideWithSetModifications> {
pep2
}
},
{
peptide3, new HashSet <PeptideWithSetModifications> {
pep3
}
},
{
peptide4, new HashSet <PeptideWithSetModifications> {
pep4
}
},
};
psm1.MatchToProteinLinkedPeptides(matching);
psm2.MatchToProteinLinkedPeptides(matching);
psm3.MatchToProteinLinkedPeptides(matching);
var newPsms = new List <PeptideSpectralMatch> {
psm1, psm2, psm3
};
FdrAnalysisEngine fdr = new FdrAnalysisEngine(newPsms, searchModes.NumNotches, true, nestedIds);
fdr.Run();
Assert.AreEqual(2, searchModes.NumNotches);
Assert.AreEqual(0, newPsms[0].FdrInfo.CumulativeDecoyNotch);
Assert.AreEqual(1, newPsms[0].FdrInfo.CumulativeTargetNotch);
Assert.AreEqual(0, newPsms[1].FdrInfo.CumulativeDecoyNotch);
Assert.AreEqual(1, newPsms[1].FdrInfo.CumulativeTargetNotch);
Assert.AreEqual(0, newPsms[2].FdrInfo.CumulativeDecoyNotch);
Assert.AreEqual(1, newPsms[2].FdrInfo.CumulativeTargetNotch);
Assert.AreEqual(0, newPsms[0].FdrInfo.CumulativeDecoy);
Assert.AreEqual(1, newPsms[0].FdrInfo.CumulativeTarget);
Assert.AreEqual(0, newPsms[1].FdrInfo.CumulativeDecoy);
Assert.AreEqual(2, newPsms[1].FdrInfo.CumulativeTarget);
Assert.AreEqual(0, newPsms[2].FdrInfo.CumulativeDecoy);
Assert.AreEqual(3, newPsms[2].FdrInfo.CumulativeTarget);
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificSettings[] fileSettingsList)
{
myTaskResults = new MyTaskResults(this)
{
newDatabases = new List <DbForTask>()
};
Status("Loading modifications...", new List <string> {
taskId
});
List <ModificationWithMass> variableModifications = GlobalVariables.AllModsKnown.OfType <ModificationWithMass>().Where(b => CommonParameters.ListOfModsVariable.Contains((b.modificationType, b.id))).ToList();
List <ModificationWithMass> fixedModifications = GlobalVariables.AllModsKnown.OfType <ModificationWithMass>().Where(b => CommonParameters.ListOfModsFixed.Contains((b.modificationType, b.id))).ToList();
List <string> localizeableModificationTypes = CommonParameters.LocalizeAll ? GlobalVariables.AllModTypesKnown.ToList() : CommonParameters.ListOfModTypesLocalize.ToList();
List <ModificationWithMass> gptmdModifications = GlobalVariables.AllModsKnown.OfType <ModificationWithMass>().Where(b => GptmdParameters.ListOfModsGptmd.Contains((b.modificationType, b.id))).ToList();
IEnumerable <Tuple <double, double> > combos = LoadCombos(gptmdModifications).ToList();
List <PeptideSpectralMatch> allPsms = new List <PeptideSpectralMatch>();
List <ProductType> ionTypes = new List <ProductType>();
if (CommonParameters.BIons)
{
ionTypes.Add(ProductType.B);
}
if (CommonParameters.YIons)
{
ionTypes.Add(ProductType.Y);
}
if (CommonParameters.CIons)
{
ionTypes.Add(ProductType.C);
}
if (CommonParameters.ZdotIons)
{
ionTypes.Add(ProductType.Zdot);
}
Status("Loading proteins...", new List <string> {
taskId
});
Dictionary <string, Modification> um = null;
//Decoys are currently not being searched with DecoyType.None
var proteinList = dbFilenameList.SelectMany(b => LoadProteinDb(b.FilePath, true, DecoyType.Reverse, localizeableModificationTypes, b.IsContaminant, out um)).ToList();
var numRawFiles = currentRawFileList.Count;
proseCreatedWhileRunning.Append("The following G-PTM-D settings were used: "); proseCreatedWhileRunning.Append("protease = " + CommonParameters.DigestionParams.Protease + "; ");
proseCreatedWhileRunning.Append("maximum missed cleavages = " + CommonParameters.DigestionParams.MaxMissedCleavages + "; ");
proseCreatedWhileRunning.Append("minimum peptide length = " + CommonParameters.DigestionParams.MinPeptideLength + "; ");
if (CommonParameters.DigestionParams.MaxPeptideLength == null)
{
proseCreatedWhileRunning.Append("maximum peptide length = unspecified; ");
}
else
{
proseCreatedWhileRunning.Append("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.id)) + "; ");
proseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.id)) + "; ");
proseCreatedWhileRunning.Append("G-PTM-D modifications count = " + gptmdModifications.Count + "; ");
//puppet searchmode for writing files. Actual searchmode is filespecific
MassDiffAcceptor tempSearchMode = new DotMassDiffAcceptor("", GetAcceptableMassShifts(fixedModifications, variableModifications, gptmdModifications, combos), CommonParameters.PrecursorMassTolerance);
proseCreatedWhileRunning.Append("parent mass tolerance(s) = {" + tempSearchMode.ToProseString() + "}; ");
proseCreatedWhileRunning.Append("product mass tolerance = " + CommonParameters.ProductMassTolerance + " Da. ");
proseCreatedWhileRunning.Append("The combined search database contained " + proteinList.Count + " total entries including " + proteinList.Where(p => p.IsContaminant).Count() + " contaminant sequences. ");
Status("Running G-PTM-D...", new List <string> {
taskId
});
HashSet <IDigestionParams> ListOfDigestionParams = GetListOfDistinctDigestionParams(CommonParameters, fileSettingsList.Select(b => SetAllFileSpecificCommonParams(CommonParameters, b)));
MyFileManager myFileManager = new MyFileManager(true);
object lock1 = new object();
object lock2 = new object();
ParallelOptions parallelOptions = new ParallelOptions();
if (CommonParameters.MaxParallelFilesToAnalyze.HasValue)
{
parallelOptions.MaxDegreeOfParallelism = CommonParameters.MaxParallelFilesToAnalyze.Value;
}
Parallel.For(0, currentRawFileList.Count, parallelOptions, spectraFileIndex =>
{
var origDataFile = currentRawFileList[spectraFileIndex];
ICommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
MassDiffAcceptor searchMode = new DotMassDiffAcceptor("", GetAcceptableMassShifts(fixedModifications, variableModifications, gptmdModifications, combos), combinedParams.PrecursorMassTolerance);
NewCollection(Path.GetFileName(origDataFile), new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
StartingDataFile(origDataFile, new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
Status("Loading spectra file...", new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
IMsDataFile <IMsDataScan <IMzSpectrum <IMzPeak> > > myMsDataFile = myFileManager.LoadFile(origDataFile, combinedParams.TopNpeaks, combinedParams.MinRatio, combinedParams.TrimMs1Peaks, combinedParams.TrimMsMsPeaks);
Status("Getting ms2 scans...", new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass = GetMs2Scans(myMsDataFile, origDataFile, combinedParams.DoPrecursorDeconvolution, combinedParams.UseProvidedPrecursorInfo, combinedParams.DeconvolutionIntensityRatio, combinedParams.DeconvolutionMaxAssumedChargeState, combinedParams.DeconvolutionMassTolerance).OrderBy(b => b.PrecursorMass).ToArray();
myFileManager.DoneWithFile(origDataFile);
PeptideSpectralMatch[] allPsmsArray = new PeptideSpectralMatch[arrayOfMs2ScansSortedByMass.Length];
new ClassicSearchEngine(allPsmsArray, arrayOfMs2ScansSortedByMass, variableModifications, fixedModifications, proteinList, ionTypes, searchMode, false, combinedParams, combinedParams.ProductMassTolerance, new List <string> {
taskId, "Individual Spectra Files", origDataFile
}).Run();
lock (lock2)
{
allPsms.AddRange(allPsmsArray);
}
FinishedDataFile(origDataFile, new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files", origDataFile
}));
});
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files"
}));
// Group and order psms
SequencesToActualProteinPeptidesEngine sequencesToActualProteinPeptidesEngineTest = new SequencesToActualProteinPeptidesEngine(allPsms, proteinList, fixedModifications, variableModifications, ionTypes, ListOfDigestionParams, CommonParameters.ReportAllAmbiguity, new List <string> {
taskId
});
var resTest = (SequencesToActualProteinPeptidesEngineResults)sequencesToActualProteinPeptidesEngineTest.Run();
Dictionary <CompactPeptideBase, HashSet <PeptideWithSetModifications> > compactPeptideToProteinPeptideMatchingTest = resTest.CompactPeptideToProteinPeptideMatching;
foreach (var huh in allPsms)
{
if (huh != null)
{
huh.MatchToProteinLinkedPeptides(compactPeptideToProteinPeptideMatchingTest);
}
}
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 => new Tuple <string, int, double?>(b.FullFilePath, b.ScanNumber, b.PeptideMonisotopicMass)).Select(b => b.First()).ToList();
new FdrAnalysisEngine(allPsms, tempSearchMode.NumNotches, false, new List <string> {
taskId
}).Run();
var writtenFile = Path.Combine(OutputFolder, "GPTMD_Candidates.psmtsv");
WritePsmsToTsv(allPsms, writtenFile, new Dictionary <string, int>());
SucessfullyFinishedWritingFile(writtenFile, new List <string> {
taskId
});
var gptmdResults = (GptmdResults) new GptmdEngine(allPsms, gptmdModifications, combos, CommonParameters.PrecursorMassTolerance, new List <string> {
taskId
}).Run();
if (dbFilenameList.Any(b => !b.IsContaminant))
{
// do NOT use this code (Path.GetFilenameWithoutExtension) because GPTMD on .xml.gz will result in .xml.xml file type being written
//string outputXMLdbFullName = Path.Combine(OutputFolder, string.Join("-", dbFilenameList.Where(b => !b.IsContaminant).Select(b => Path.GetFileNameWithoutExtension(b.FilePath))) + "GPTMD.xml");
List <string> databaseNames = new List <string>();
foreach (var nonContaminantDb in dbFilenameList.Where(p => !p.IsContaminant))
{
var dbName = Path.GetFileName(nonContaminantDb.FilePath);
int indexOfFirstDot = dbName.IndexOf(".");
databaseNames.Add(dbName.Substring(0, indexOfFirstDot));
}
string outputXMLdbFullName = Path.Combine(OutputFolder, string.Join("-", databaseNames) + "GPTMD.xml");
var newModsActuallyWritten = ProteinDbWriter.WriteXmlDatabase(gptmdResults.Mods, proteinList.Where(b => !b.IsDecoy && !b.IsContaminant).ToList(), outputXMLdbFullName);
SucessfullyFinishedWritingFile(outputXMLdbFullName, new List <string> {
taskId
});
myTaskResults.newDatabases.Add(new DbForTask(outputXMLdbFullName, false));
myTaskResults.AddNiceText("Modifications added: " + newModsActuallyWritten.Select(b => b.Value).Sum());
myTaskResults.AddNiceText("Mods types and counts:");
myTaskResults.AddNiceText(string.Join(Environment.NewLine, newModsActuallyWritten.OrderByDescending(b => b.Value).Select(b => "\t" + b.Key + "\t" + b.Value)));
}
if (dbFilenameList.Any(b => b.IsContaminant))
{
// do NOT use this code (Path.GetFilenameWithoutExtension) because GPTMD on .xml.gz will result in .xml.xml file type being written
//string outputXMLdbFullNameContaminants = Path.Combine(OutputFolder, string.Join("-", dbFilenameList.Where(b => b.IsContaminant).Select(b => Path.GetFileNameWithoutExtension(b.FilePath))) + "GPTMD.xml");
List <string> databaseNames = new List <string>();
foreach (var contaminantDb in dbFilenameList.Where(p => p.IsContaminant))
{
var dbName = Path.GetFileName(contaminantDb.FilePath);
int indexOfFirstDot = dbName.IndexOf(".");
databaseNames.Add(dbName.Substring(0, indexOfFirstDot));
}
string outputXMLdbFullNameContaminants = Path.Combine(OutputFolder, string.Join("-", databaseNames) + "GPTMD.xml");
var newModsActuallyWritten = ProteinDbWriter.WriteXmlDatabase(gptmdResults.Mods, proteinList.Where(b => !b.IsDecoy && b.IsContaminant).ToList(), outputXMLdbFullNameContaminants);
SucessfullyFinishedWritingFile(outputXMLdbFullNameContaminants, new List <string> {
taskId
});
myTaskResults.newDatabases.Add(new DbForTask(outputXMLdbFullNameContaminants, true));
myTaskResults.AddNiceText("Contaminant modifications added: " + newModsActuallyWritten.Select(b => b.Value).Sum());
myTaskResults.AddNiceText("Mods types and counts:");
myTaskResults.AddNiceText(string.Join(Environment.NewLine, newModsActuallyWritten.OrderByDescending(b => b.Value).Select(b => "\t" + b.Key + "\t" + b.Value)));
}
return(myTaskResults);
}
