public static void AnotherSpectralLibrarySearchTestDecoy()
{
var testDir = Path.Combine(TestContext.CurrentContext.TestDirectory, @"TestData\SpectralLibrarySearch");
string myFile = Path.Combine(testDir, @"slicedMouse.raw");
MyFileManager myFileManager = new MyFileManager(true);
CommonParameters commonParameters = new CommonParameters(maxThreadsToUsePerFile: 1, scoreCutoff: 1);
MsDataFile myMsDataFile = myFileManager.LoadFile(myFile, commonParameters);
var variableModifications = new List <Modification>();
var fixedModifications = new List <Modification>();
var proteinList = new List <Protein> {
new Protein("QTATIAHVTTMLGEVIGFNDHIVK", "P16858")
};
string targetSpectralLibrary = Path.Combine(testDir, @"P16858_target.msp");
string decoySpectralLibrary = Path.Combine(testDir, @"P16858_decoy.msp");
List <string> specLibs = new List <string> {
targetSpectralLibrary, decoySpectralLibrary
};
SpectralLibrary sl = new SpectralLibrary(specLibs);
var searchModes = new SinglePpmAroundZeroSearchMode(5);
var listOfSortedms2Scans = MetaMorpheusTask.GetMs2Scans(myMsDataFile, null, new CommonParameters()).OrderBy(b => b.PrecursorMass).ToArray();
PeptideSpectralMatch[] allPsmsArray = new PeptideSpectralMatch[listOfSortedms2Scans.Length];
bool writeSpectralLibrary = false;
new ClassicSearchEngine(allPsmsArray, listOfSortedms2Scans, variableModifications, fixedModifications, null, null, null,
proteinList, searchModes, commonParameters, null, sl, new List <string>(), writeSpectralLibrary).Run();
// Single search mode
Assert.AreEqual(7, allPsmsArray.Length);
Assert.IsTrue(allPsmsArray[5].Score > 38);
Assert.AreEqual("VIHDNFGIVEGLMTTVHAITATQK", allPsmsArray[5].BaseSequence);
Assert.IsTrue(allPsmsArray[5].IsDecoy);
SpectralLibrarySearchFunction.CalculateSpectralAngles(sl, allPsmsArray, listOfSortedms2Scans, commonParameters);
Assert.That(allPsmsArray[5].SpectralAngle, Is.EqualTo(0.69).Within(0.01));
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
if (SearchParameters.DoQuantification)
{
// disable quantification if a .mgf is being used
if (currentRawFileList.Any(x => Path.GetExtension(x).Equals(".mgf", StringComparison.OrdinalIgnoreCase)))
{
SearchParameters.DoQuantification = false;
}
//if we're doing SILAC, assign and add the silac labels to the residue dictionary
else if (SearchParameters.SilacLabels != null || SearchParameters.StartTurnoverLabel != null || SearchParameters.EndTurnoverLabel != null)
{
char heavyLabel = 'a'; //char to assign
//add the Turnoverlabels to the silacLabels list. They weren't there before just to prevent duplication in the tomls
if (SearchParameters.StartTurnoverLabel != null || SearchParameters.EndTurnoverLabel != null)
{
//original silacLabels object is null, so we need to initialize it
SearchParameters.SilacLabels = new List <SilacLabel>();
if (SearchParameters.StartTurnoverLabel != null)
{
var updatedLabel = SilacConversions.UpdateAminoAcidLabel(SearchParameters.StartTurnoverLabel, heavyLabel);
heavyLabel = updatedLabel.nextHeavyLabel;
SearchParameters.StartTurnoverLabel = updatedLabel.updatedLabel;
SearchParameters.SilacLabels.Add(SearchParameters.StartTurnoverLabel);
}
if (SearchParameters.EndTurnoverLabel != null)
{
var updatedLabel = SilacConversions.UpdateAminoAcidLabel(SearchParameters.EndTurnoverLabel, heavyLabel);
heavyLabel = updatedLabel.nextHeavyLabel;
SearchParameters.EndTurnoverLabel = updatedLabel.updatedLabel;
SearchParameters.SilacLabels.Add(SearchParameters.EndTurnoverLabel);
}
}
else
{
//change the silac residues to lower case amino acids (currently null)
List <SilacLabel> updatedLabels = new List <SilacLabel>();
for (int i = 0; i < SearchParameters.SilacLabels.Count; i++)
{
var updatedLabel = SilacConversions.UpdateAminoAcidLabel(SearchParameters.SilacLabels[i], heavyLabel);
heavyLabel = updatedLabel.nextHeavyLabel;
updatedLabels.Add(updatedLabel.updatedLabel);
}
SearchParameters.SilacLabels = updatedLabels;
}
}
}
//if no quant, remove any silac labels that may have been added, because they screw up downstream analysis
if (!SearchParameters.DoQuantification) //using "if" instead of "else", because DoQuantification can change if it's an mgf
{
SearchParameters.SilacLabels = null;
}
LoadModifications(taskId, out var variableModifications, out var fixedModifications, out var localizeableModificationTypes);
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, SearchParameters.SearchTarget, SearchParameters.DecoyType, localizeableModificationTypes, CommonParameters);
SanitizeProteinDatabase(proteinList, SearchParameters.TCAmbiguity);
// load spectral libraries
var spectralLibrary = LoadSpectralLibraries(taskId, dbFilenameList);
// write prose settings
ProseCreatedWhileRunning.Append("The following search 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("fixed modifications = " + string.Join(", ", fixedModifications.Select(m => m.IdWithMotif)) + "; ");
ProseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.IdWithMotif)) + "; ");
ProseCreatedWhileRunning.Append("max mods per peptide = " + CommonParameters.DigestionParams.MaxModsForPeptide + "; ");
ProseCreatedWhileRunning.Append("max modification isoforms = " + CommonParameters.DigestionParams.MaxModificationIsoforms + "; ");
ProseCreatedWhileRunning.Append("precursor mass tolerance = " + CommonParameters.PrecursorMassTolerance + "; ");
ProseCreatedWhileRunning.Append("product mass tolerance = " + CommonParameters.ProductMassTolerance + "; ");
ProseCreatedWhileRunning.Append("report PSM ambiguity = " + CommonParameters.ReportAllAmbiguity + ". ");
ProseCreatedWhileRunning.Append("The combined search database contained " + proteinList.Count(p => !p.IsDecoy)
+ " non-decoy protein entries including " + proteinList.Count(p => p.IsContaminant) + " contaminant sequences. ");
// start the search task
MyTaskResults = new MyTaskResults(this);
List <PeptideSpectralMatch> allPsms = new List <PeptideSpectralMatch>();
//generate an array to store category specific fdr values (for speedy semi/nonspecific searches)
int numFdrCategories = (int)(Enum.GetValues(typeof(FdrCategory)).Cast <FdrCategory>().Last() + 1); //+1 because it starts at zero
List <PeptideSpectralMatch>[] allCategorySpecificPsms = new List <PeptideSpectralMatch> [numFdrCategories];
for (int i = 0; i < numFdrCategories; i++)
{
allCategorySpecificPsms[i] = new List <PeptideSpectralMatch>();
}
FlashLfqResults flashLfqResults = null;
MyFileManager myFileManager = new MyFileManager(SearchParameters.DisposeOfFileWhenDone);
var fileSpecificCommonParams = fileSettingsList.Select(b => SetAllFileSpecificCommonParams(CommonParameters, b));
int completedFiles = 0;
object indexLock = new object();
object psmLock = new object();
Status("Searching files...", taskId);
Status("Searching files...", new List <string> {
taskId, "Individual Spectra Files"
});
Dictionary <string, int[]> numMs2SpectraPerFile = new Dictionary <string, int[]>();
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
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 massDiffAcceptor = GetMassDiffAcceptor(combinedParams.PrecursorMassTolerance, SearchParameters.MassDiffAcceptorType, SearchParameters.CustomMdac);
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();
numMs2SpectraPerFile.Add(Path.GetFileNameWithoutExtension(origDataFile), new int[] { myMsDataFile.GetAllScansList().Count(p => p.MsnOrder == 2), arrayOfMs2ScansSortedByMass.Length });
myFileManager.DoneWithFile(origDataFile);
PeptideSpectralMatch[] fileSpecificPsms = new PeptideSpectralMatch[arrayOfMs2ScansSortedByMass.Length];
// modern search
if (SearchParameters.SearchType == SearchType.Modern)
{
for (int currentPartition = 0; currentPartition < combinedParams.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, SearchParameters.SilacLabels,
SearchParameters.StartTurnoverLabel, SearchParameters.EndTurnoverLabel, currentPartition, SearchParameters.DecoyType, combinedParams, FileSpecificParameters,
SearchParameters.MaxFragmentSize, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), SearchParameters.TCAmbiguity, new List <string> {
taskId
});
List <int>[] fragmentIndex = null;
List <int>[] precursorIndex = null;
lock (indexLock)
{
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, ref precursorIndex, proteinList, taskId);
}
Status("Searching files...", taskId);
new ModernSearchEngine(fileSpecificPsms, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, currentPartition,
combinedParams, this.FileSpecificParameters, massDiffAcceptor, SearchParameters.MaximumMassThatFragmentIonScoreIsDoubled, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + combinedParams.TotalPartitions + "!", thisId));
if (GlobalVariables.StopLoops)
{
break;
}
}
}
// nonspecific search
else if (SearchParameters.SearchType == SearchType.NonSpecific)
{
PeptideSpectralMatch[][] fileSpecificPsmsSeparatedByFdrCategory = new PeptideSpectralMatch[numFdrCategories][]; //generate an array of all possible locals
for (int i = 0; i < numFdrCategories; i++) //only add if we're using for FDR, else ignore it as null.
{
fileSpecificPsmsSeparatedByFdrCategory[i] = new PeptideSpectralMatch[arrayOfMs2ScansSortedByMass.Length];
}
//create params for N, C, or both if semi
List <CommonParameters> paramsToUse = new List <CommonParameters> {
combinedParams
};
if (combinedParams.DigestionParams.SearchModeType == CleavageSpecificity.Semi) //if semi, we need to do both N and C to hit everything
{
paramsToUse.Clear();
List <FragmentationTerminus> terminiToUse = new List <FragmentationTerminus> {
FragmentationTerminus.N, FragmentationTerminus.C
};
foreach (FragmentationTerminus terminus in terminiToUse) //set both termini
{
paramsToUse.Add(combinedParams.CloneWithNewTerminus(terminus));
}
}
//Compress array of deconvoluted ms2 scans to avoid searching the same ms2 multiple times while still identifying coisolated peptides
List <int>[] coisolationIndex = new List <int>[] { new List <int>() };
if (arrayOfMs2ScansSortedByMass.Length != 0)
{
int maxScanNumber = arrayOfMs2ScansSortedByMass.Max(x => x.OneBasedScanNumber);
coisolationIndex = new List <int> [maxScanNumber + 1];
for (int i = 0; i < arrayOfMs2ScansSortedByMass.Length; i++)
{
int scanNumber = arrayOfMs2ScansSortedByMass[i].OneBasedScanNumber;
if (coisolationIndex[scanNumber] == null)
{
coisolationIndex[scanNumber] = new List <int> {
i
};
}
else
{
coisolationIndex[scanNumber].Add(i);
}
}
coisolationIndex = coisolationIndex.Where(x => x != null).ToArray();
}
//foreach terminus we're going to look at
foreach (CommonParameters paramToUse in paramsToUse)
{
//foreach database partition
for (int currentPartition = 0; currentPartition < paramToUse.TotalPartitions; currentPartition++)
{
List <PeptideWithSetModifications> peptideIndex = null;
List <Protein> proteinListSubset = proteinList.GetRange(currentPartition * proteinList.Count / paramToUse.TotalPartitions,
((currentPartition + 1) * proteinList.Count / paramToUse.TotalPartitions) - (currentPartition * proteinList.Count / paramToUse.TotalPartitions));
List <int>[] fragmentIndex = null;
List <int>[] precursorIndex = null;
Status("Getting fragment dictionary...", new List <string> {
taskId
});
var indexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, SearchParameters.SilacLabels,
SearchParameters.StartTurnoverLabel, SearchParameters.EndTurnoverLabel, currentPartition, SearchParameters.DecoyType, paramToUse, FileSpecificParameters,
SearchParameters.MaxFragmentSize, true, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), SearchParameters.TCAmbiguity, new List <string> {
taskId
});
lock (indexLock)
{
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, ref precursorIndex, proteinList, taskId);
}
Status("Searching files...", taskId);
new NonSpecificEnzymeSearchEngine(fileSpecificPsmsSeparatedByFdrCategory, arrayOfMs2ScansSortedByMass, coisolationIndex, peptideIndex, fragmentIndex,
precursorIndex, currentPartition, paramToUse, this.FileSpecificParameters, variableModifications, massDiffAcceptor,
SearchParameters.MaximumMassThatFragmentIonScoreIsDoubled, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + paramToUse.TotalPartitions + "!", thisId));
if (GlobalVariables.StopLoops)
{
break;
}
}
}
lock (psmLock)
{
for (int i = 0; i < allCategorySpecificPsms.Length; i++)
{
if (allCategorySpecificPsms[i] != null)
{
allCategorySpecificPsms[i].AddRange(fileSpecificPsmsSeparatedByFdrCategory[i]);
}
}
}
}
// classic search
else
{
Status("Starting search...", thisId);
var newClassicSearchEngine = new ClassicSearchEngine(fileSpecificPsms, arrayOfMs2ScansSortedByMass, variableModifications, fixedModifications, SearchParameters.SilacLabels,
SearchParameters.StartTurnoverLabel, SearchParameters.EndTurnoverLabel, proteinList, massDiffAcceptor, combinedParams, this.FileSpecificParameters, spectralLibrary, thisId, SearchParameters.WriteSpectralLibrary);
newClassicSearchEngine.Run();
ReportProgress(new ProgressEventArgs(100, "Done with search!", thisId));
}
//look for internal fragments
if (SearchParameters.MinAllowedInternalFragmentLength != 0)
{
MatchInternalFragmentIons(fileSpecificPsms, arrayOfMs2ScansSortedByMass, combinedParams, SearchParameters.MinAllowedInternalFragmentLength);
}
// calculate/set spectral angles if there is a spectral library being used
if (spectralLibrary != null)
{
Status("Calculating spectral library similarity...", thisId);
}
SpectralLibrarySearchFunction.CalculateSpectralAngles(spectralLibrary, fileSpecificPsms, arrayOfMs2ScansSortedByMass, combinedParams);
lock (psmLock)
{
allPsms.AddRange(fileSpecificPsms);
}
completedFiles++;
FinishedDataFile(origDataFile, new List <string> {
taskId, "Individual Spectra Files", origDataFile
});
ReportProgress(new ProgressEventArgs(completedFiles / currentRawFileList.Count, "Searching...", new List <string> {
taskId, "Individual Spectra Files"
}));
}
if (spectralLibrary != null)
{
spectralLibrary.CloseConnections();
}
ReportProgress(new ProgressEventArgs(100, "Done with all searches!", new List <string> {
taskId, "Individual Spectra Files"
}));
int numNotches = GetNumNotches(SearchParameters.MassDiffAcceptorType, SearchParameters.CustomMdac);
//resolve category specific fdrs (for speedy semi and nonspecific
if (SearchParameters.SearchType == SearchType.NonSpecific)
{
allPsms = NonSpecificEnzymeSearchEngine.ResolveFdrCategorySpecificPsms(allCategorySpecificPsms, numNotches, taskId, CommonParameters, FileSpecificParameters);
}
PostSearchAnalysisParameters parameters = new PostSearchAnalysisParameters
{
SearchTaskResults = MyTaskResults,
SearchTaskId = taskId,
SearchParameters = SearchParameters,
ProteinList = proteinList,
AllPsms = allPsms,
VariableModifications = variableModifications,
FixedModifications = fixedModifications,
ListOfDigestionParams = new HashSet <DigestionParams>(fileSpecificCommonParams.Select(p => p.DigestionParams)),
CurrentRawFileList = currentRawFileList,
MyFileManager = myFileManager,
NumNotches = numNotches,
OutputFolder = OutputFolder,
IndividualResultsOutputFolder = Path.Combine(OutputFolder, "Individual File Results"),
FlashLfqResults = flashLfqResults,
FileSettingsList = fileSettingsList,
NumMs2SpectraPerFile = numMs2SpectraPerFile,
DatabaseFilenameList = dbFilenameList
};
PostSearchAnalysisTask postProcessing = new PostSearchAnalysisTask
{
Parameters = parameters,
FileSpecificParameters = this.FileSpecificParameters,
CommonParameters = CommonParameters
};
return(postProcessing.Run());
}
