public static void TestRunSpecificPostSearchAnalysis()
{
//code coverage unit test for an unused abstract method in post search analysis
var task = new PostSearchAnalysisTask();
task.RunTask(TestContext.CurrentContext.TestDirectory, new List <DbForTask>(), new List <string>(), "");
}
public static void TestProteinPrunedWithModSelectionAndVariants()
{
var modToWrite = GlobalVariables.AllModsKnown.Where(p => p.ModificationType == "UniProt" && p.Target.ToString() == "T").First();
var modToNotWrite = GlobalVariables.AllModsKnown.Where(p => p.ModificationType == "Common Artifact" && p.Target.ToString() == "X").First();
Dictionary <int, List <Modification> > variantMods = new Dictionary <int, List <Modification> >();
variantMods.Add(1, new List <Modification>()
{
modToNotWrite
});
List <SequenceVariation> variants = new List <SequenceVariation> {
new SequenceVariation(4, 4, "V", "T", @"20\t41168825\t.\tT\tC\t14290.77\t.\tANN=C|missense_variant|MODERATE|PLCG1|ENSG00000124181|transcript|ENST00000244007.7|protein_coding|22/33|c.2438T>C|p.Ile813Thr|2635/5285|2438/3876|813/1291||\tGT:AD:DP:GQ:PL\t1/1:1,392:393:99:14319,1142,0", variantMods)
};
var protein1 = new Protein("PEPVIDEKPEPT", "1", oneBasedModifications: new Dictionary <int, List <Modification> > {
{ 1, new List <Modification> {
modToNotWrite
} }, { 12, new List <Modification> {
modToWrite
} }
}, sequenceVariations: variants);
var protein2 = new Protein("PEPIDPEPT", "2", oneBasedModifications: new Dictionary <int, List <Modification> > {
{ 1, new List <Modification> {
modToNotWrite
} }, { 9, new List <Modification> {
modToWrite
} }
});
var protein1Variants = protein1.GetVariantProteins(1, 0);
string path = @"temp";
var proteinList = new List <Protein> {
protein1, protein2
};
proteinList.AddRange(protein1Variants);
ProteinDbWriter.WriteXmlDatabase(new Dictionary <string, HashSet <Tuple <int, Modification> > >(), proteinList, path);
Directory.CreateDirectory(Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTestVariant"));
Dictionary <string, HashSet <Tuple <int, Modification> > > modList = new Dictionary <string, HashSet <Tuple <int, Modification> > >();
var Hash = new HashSet <Tuple <int, Modification> >
{
new Tuple <int, Modification>(1, modToWrite),
new Tuple <int, Modification>(2, modToNotWrite),
};
var db = ProteinDbWriter.WriteXmlDatabase(modList, proteinList, Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTestVariant/fakeDb.xml"));
var peptideObserved = protein1Variants.First().Digest(new DigestionParams(minPeptideLength: 1), new List <Modification>(), new List <Modification>())
.Where(p => p.BaseSequence == "PEPT").First();
PostSearchAnalysisParameters testPostTaskParameters = new PostSearchAnalysisParameters();
CommonParameters commonParam = new CommonParameters(useDeltaScore: false);
double[,] noiseData = new double[10000, 10000];
noiseData[0, 0] = 1.0;
List <Proteomics.Fragmentation.MatchedFragmentIon> matchedFragmentIons = new List <Proteomics.Fragmentation.MatchedFragmentIon>()
{
};
MzSpectrum spectrum = new MzSpectrum(noiseData);
MsDataScan scan = new MsDataScan(spectrum, 1, 1, true, Polarity.Unknown, 2, new MzLibUtil.MzRange(10, 1000), "", MZAnalyzerType.Orbitrap, 10000, null, noiseData, "");
testPostTaskParameters.ProteinList = proteinList;
testPostTaskParameters.AllPsms = new List <PeptideSpectralMatch> {
new PeptideSpectralMatch(peptideObserved, 0, 20, 1, new Ms2ScanWithSpecificMass(scan, 100, 1, @"", commonParam), commonParam, matchedFragmentIons)
};
testPostTaskParameters.SearchParameters = new SearchParameters();
testPostTaskParameters.SearchParameters.WritePrunedDatabase = true;
testPostTaskParameters.SearchParameters.DoQuantification = false;
testPostTaskParameters.SearchParameters.WriteMzId = false;
testPostTaskParameters.DatabaseFilenameList = new List <DbForTask>()
{
new DbForTask(Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest/fakeDb.xml"), false)
};
testPostTaskParameters.OutputFolder = Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest");
Directory.CreateDirectory(Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest/individual"));
testPostTaskParameters.IndividualResultsOutputFolder = Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest/individual");
int[] stuffForSpectraFile = new int[2];
stuffForSpectraFile[0] = 10;
stuffForSpectraFile[1] = 10;
Dictionary <string, int[]> numSpectraPerFile = new Dictionary <string, int[]>();
numSpectraPerFile.Add("", stuffForSpectraFile);
testPostTaskParameters.NumMs2SpectraPerFile = numSpectraPerFile;
MsDataFile myMsDataFile = new TestDataFile(new List <PeptideWithSetModifications>
{
peptideObserved
});
string mzmlName = @"newMzml.mzML";
IO.MzML.MzmlMethods.CreateAndWriteMyMzmlWithCalibratedSpectra(myMsDataFile, mzmlName, false);
modList.Add("test", Hash);
testPostTaskParameters.CurrentRawFileList = new List <string>()
{
mzmlName
};
SearchTask task5 = new SearchTask
{
SearchParameters = new SearchParameters
{
WritePrunedDatabase = true,
SearchTarget = true,
MassDiffAcceptorType = MassDiffAcceptorType.Exact,
},
CommonParameters = new CommonParameters()
};
var test = task5.RunTask(Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest"), new List <DbForTask>()
{
new DbForTask(Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest/fakeDb.xml"), false)
}, new List <string>()
{
mzmlName
}, "name");
testPostTaskParameters.SearchTaskResults = test;
PostSearchAnalysisTask testPostTask = new PostSearchAnalysisTask();
testPostTask.Parameters = testPostTaskParameters;
testPostTask.CommonParameters = commonParam;
testPostTask.FileSpecificParameters = new List <(string FileName, CommonParameters Parameters)> {
("newMzMl.mzml", commonParam)
};
testPostTask.Run();
var proteinsLoaded = ProteinDbLoader.LoadProteinXML(path, true, DecoyType.None, GlobalVariables.AllModsKnown, false, new List <string>(), out var unknownMods);
// assert that mods on proteins are the same before/after task is run
Assert.AreEqual(protein1Variants.First().Accession, proteinsLoaded.First().Accession);
Assert.AreEqual(protein1Variants.First().OneBasedPossibleLocalizedModifications.Count(), proteinsLoaded.First().OneBasedPossibleLocalizedModifications.Count());
Assert.AreEqual(protein2.OneBasedPossibleLocalizedModifications.Count(), proteinsLoaded.ElementAt(1).OneBasedPossibleLocalizedModifications.Count());
// assert that protein pruned DB has correct proteins mods
var proteinPruned = ProteinDbLoader.LoadProteinXML(Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest/fakeDbproteinPruned.xml"), true, DecoyType.None, GlobalVariables.AllModsKnown, false, new List <string>(), out var unknownMods1);
Assert.That(proteinPruned.Count().Equals(1));
Assert.That(proteinPruned.FirstOrDefault().OneBasedPossibleLocalizedModifications.Count().Equals(1));
// assert that mod-pruned DB has correct proteins and mods
var modPruned = ProteinDbLoader.LoadProteinXML(Path.Combine(TestContext.CurrentContext.TestDirectory, @"PrunedDbTest/fakeDbpruned.xml"), true, DecoyType.None, GlobalVariables.AllModsKnown, false, new List <string>(), out var unknownMods2);
Assert.That(modPruned.Count().Equals(2));
Assert.That(modPruned.ElementAt(0).OneBasedPossibleLocalizedModifications.Count().Equals(1));
Assert.That(modPruned.ElementAt(1).OneBasedPossibleLocalizedModifications.Count().Equals(1));
}
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());
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
// disable quantification if a .mgf is being used
if (SearchParameters.DoQuantification && currentRawFileList.Any(x => Path.GetExtension(x).Equals(".mgf", StringComparison.OrdinalIgnoreCase)))
{
SearchParameters.DoQuantification = false;
}
// load modifications
Status("Loading modifications...", 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 = GlobalVariables.AllModTypesKnown.ToList();
// what types of fragment ions to search for
List <ProductType> ionTypes = new List <ProductType>();
if (CommonParameters.BIons && CommonParameters.AddCompIons)
{
ionTypes.Add(ProductType.B);
}
else if (CommonParameters.BIons)
{
ionTypes.Add(ProductType.BnoB1ions);
}
if (CommonParameters.YIons)
{
ionTypes.Add(ProductType.Y);
}
if (CommonParameters.ZdotIons)
{
ionTypes.Add(ProductType.Zdot);
}
if (CommonParameters.CIons)
{
ionTypes.Add(ProductType.C);
}
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, SearchParameters.SearchTarget, SearchParameters.DecoyType, localizeableModificationTypes, CommonParameters);
// 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.id)) + "; ");
ProseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.id)) + "; ");
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>();
FlashLFQResults flashLfqResults = null;
MyFileManager myFileManager = new MyFileManager(SearchParameters.DisposeOfFileWhenDone);
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);
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.TopNpeaks, combinedParams.MinRatio, combinedParams.TrimMs1Peaks, combinedParams.TrimMsMsPeaks, combinedParams);
Status("Getting ms2 scans...", thisId);
Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass = GetMs2Scans(myMsDataFile, origDataFile, combinedParams.DoPrecursorDeconvolution, combinedParams.UseProvidedPrecursorInfo, combinedParams.DeconvolutionIntensityRatio, combinedParams.DeconvolutionMaxAssumedChargeState, combinedParams.DeconvolutionMassTolerance).OrderBy(b => b.PrecursorMass).ToArray();
numMs2SpectraPerFile.Add(Path.GetFileNameWithoutExtension(origDataFile), new int[] { myMsDataFile.GetAllScansList().Count(p => p.MsnOrder == 2), arrayOfMs2ScansSortedByMass.Length });
myFileManager.DoneWithFile(origDataFile);
var fileSpecificPsms = new PeptideSpectralMatch[arrayOfMs2ScansSortedByMass.Length];
// modern search
if (SearchParameters.SearchType == SearchType.Modern)
{
for (int currentPartition = 0; currentPartition < combinedParams.TotalPartitions; currentPartition++)
{
List <CompactPeptide> 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, ionTypes, currentPartition, SearchParameters.DecoyType, ListOfDigestionParams, combinedParams, SearchParameters.MaxFragmentSize, new List <string> {
taskId
});
List <int>[] fragmentIndex = null;
lock (indexLock)
{
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, taskId);
}
Status("Searching files...", taskId);
new ModernSearchEngine(fileSpecificPsms, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, ionTypes, currentPartition, combinedParams, massDiffAcceptor, SearchParameters.MaximumMassThatFragmentIonScoreIsDoubled, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + combinedParams.TotalPartitions + "!", thisId));
}
}
// nonspecific search
else if (SearchParameters.SearchType == SearchType.NonSpecific)
{
List <List <ProductType> > terminusSeparatedIons = ProductTypeMethods.SeparateIonsByTerminus(ionTypes);
foreach (List <ProductType> terminusSpecificIons in terminusSeparatedIons)
{
for (int currentPartition = 0; currentPartition < combinedParams.TotalPartitions; currentPartition++)
{
List <CompactPeptide> peptideIndex = null;
List <Protein> proteinListSubset = proteinList.GetRange(currentPartition * proteinList.Count() / combinedParams.TotalPartitions, ((currentPartition + 1) * proteinList.Count() / combinedParams.TotalPartitions) - (currentPartition * proteinList.Count() / combinedParams.TotalPartitions));
List <int>[] fragmentIndex = new List <int> [1];
Status("Getting fragment dictionary...", new List <string> {
taskId
});
var indexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, terminusSpecificIons, currentPartition, SearchParameters.DecoyType, ListOfDigestionParams, combinedParams, SearchParameters.MaxFragmentSize, new List <string> {
taskId
});
lock (indexLock)
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, taskId);
Status("Getting precursor dictionary...", new List <string> {
taskId
});
List <CompactPeptide> peptideIndexPrecursor = null;
List <Protein> proteinListSubsetPrecursor = proteinList.GetRange(currentPartition * proteinList.Count() / combinedParams.TotalPartitions, ((currentPartition + 1) * proteinList.Count() / combinedParams.TotalPartitions) - (currentPartition * proteinList.Count() / combinedParams.TotalPartitions));
List <int>[] fragmentIndexPrecursor = new List <int> [1];
var indexEnginePrecursor = new PrecursorIndexingEngine(proteinListSubsetPrecursor, variableModifications, fixedModifications, terminusSpecificIons, currentPartition, SearchParameters.DecoyType, ListOfDigestionParams, combinedParams, 0, new List <string> {
taskId
});
lock (indexLock)
GenerateIndexes(indexEnginePrecursor, dbFilenameList, ref peptideIndexPrecursor, ref fragmentIndexPrecursor, taskId);
if (peptideIndex.Count != peptideIndexPrecursor.Count)
{
throw new MetaMorpheusException("peptideIndex not identical between indexing engines");
}
Status("Searching files...", taskId);
new NonSpecificEnzymeSearchEngine(fileSpecificPsms, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, fragmentIndexPrecursor, terminusSpecificIons, currentPartition, combinedParams, massDiffAcceptor, SearchParameters.MaximumMassThatFragmentIonScoreIsDoubled, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + combinedParams.TotalPartitions + "!", thisId));
}
}
}
// classic search
else
{
Status("Starting search...", thisId);
new ClassicSearchEngine(fileSpecificPsms, arrayOfMs2ScansSortedByMass, variableModifications, fixedModifications, proteinList, ionTypes, massDiffAcceptor, combinedParams, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search!", thisId));
}
lock (psmLock)
{
allPsms.AddRange(fileSpecificPsms.Where(p => p != null));
}
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"
}));
}
ReportProgress(new ProgressEventArgs(100, "Done with all searches!", new List <string> {
taskId, "Individual Spectra Files"
}));
PostSearchAnalysisParameters parameters = new PostSearchAnalysisParameters();
parameters.SearchTaskResults = MyTaskResults;
parameters.SearchTaskId = taskId;
parameters.SearchParameters = SearchParameters;
parameters.ProteinList = proteinList;
parameters.IonTypes = ionTypes;
parameters.AllPsms = allPsms;
parameters.FixedModifications = fixedModifications;
parameters.VariableModifications = variableModifications;
parameters.ListOfDigestionParams = ListOfDigestionParams;
parameters.CurrentRawFileList = currentRawFileList;
parameters.MyFileManager = myFileManager;
parameters.NumNotches = GetNumNotches(SearchParameters.MassDiffAcceptorType, SearchParameters.CustomMdac);
parameters.OutputFolder = OutputFolder;
parameters.IndividualResultsOutputFolder = Path.Combine(OutputFolder, "Individual File Results");
parameters.FlashLfqResults = flashLfqResults;
parameters.FileSettingsList = fileSettingsList;
parameters.NumMs2SpectraPerFile = numMs2SpectraPerFile;
parameters.DatabaseFilenameList = dbFilenameList;
PostSearchAnalysisTask postProcessing = new PostSearchAnalysisTask();
postProcessing.Parameters = parameters;
postProcessing.CommonParameters = CommonParameters;
return(postProcessing.Run());
}
