protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList) { MyTaskResults = new MyTaskResults(this); List <PsmCross> allPsms = new List <PsmCross>(); var compactPeptideToProteinPeptideMatch = new Dictionary <CompactPeptideBase, HashSet <PeptideWithSetModifications> >(); 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(); // load proteins List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, XlSearchParameters.DecoyType, localizeableModificationTypes, CommonParameters); List <ProductType> ionTypes = new List <ProductType>(); 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); } TerminusType terminusType = ProductTypeMethods.IdentifyTerminusType(ionTypes); var crosslinker = new CrosslinkerTypeClass(); crosslinker.SelectCrosslinker(XlSearchParameters.CrosslinkerType); if (XlSearchParameters.CrosslinkerType == CrosslinkerType.UserDefined) { crosslinker = GenerateUserDefinedCrosslinker(XlSearchParameters); } MyFileManager myFileManager = new MyFileManager(XlSearchParameters.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); 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.id)) + "; "); ProseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.id)) + "; "); ProseCreatedWhileRunning.Append("parent mass tolerance(s) = " + XlSearchParameters.XlPrecusorMsTl + "; "); 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++) { if (GlobalVariables.StopLoops) { break; } var origDataFile = currentRawFileList[spectraFileIndex]; CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]); List <PsmCross> newPsms = new List <PsmCross>(); 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(); for (int currentPartition = 0; currentPartition < CommonParameters.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, UsefulProteomicsDatabases.DecoyType.Reverse, ListOfDigestionParams, combinedParams, 30000.0, new List <string> { taskId }); List <int>[] fragmentIndex = null; lock (indexLock) GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, taskId); Status("Searching files...", taskId); new TwoPassCrosslinkSearchEngine(newPsms, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, ionTypes, currentPartition, combinedParams, false, XlSearchParameters.XlPrecusorMsTl, crosslinker, XlSearchParameters.CrosslinkSearchTop, XlSearchParameters.CrosslinkSearchTopNum, XlSearchParameters.XlQuench_H2O, XlSearchParameters.XlQuench_NH2, XlSearchParameters.XlQuench_Tris, XlSearchParameters.XlCharge_2_3, XlSearchParameters.XlCharge_2_3_PrimeFragment, thisId).Run(); ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + CommonParameters.TotalPartitions + "!", thisId)); } lock (psmLock) { 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" })); Status("Crosslink analysis engine", taskId); MetaMorpheusEngineResults allcrosslinkanalysisResults; allcrosslinkanalysisResults = new CrosslinkAnalysisEngine(allPsms, compactPeptideToProteinPeptideMatch, proteinList, variableModifications, fixedModifications, ionTypes, OutputFolder, crosslinker, terminusType, CommonParameters, new List <string> { taskId }).Run(); allPsms = allPsms.ToList(); if (XlSearchParameters.XlOutAll) { try { WriteAllToTsv(allPsms, OutputFolder, "allPsms", new List <string> { taskId }); } catch (Exception) { throw; } } var allPsmsXL = allPsms.Where(p => p.CrossType == PsmCrossType.Cross).Where(p => p.XLBestScore >= CommonParameters.ScoreCutoff && p.BetaPsmCross.XLBestScore >= CommonParameters.ScoreCutoff).ToList(); foreach (var item in allPsmsXL) { if (item.OneBasedStartResidueInProtein.HasValue) { item.XlProteinPos = item.OneBasedStartResidueInProtein.Value + item.XlPos - 1; } if (item.BetaPsmCross.OneBasedStartResidueInProtein.HasValue) { item.BetaPsmCross.XlProteinPos = item.BetaPsmCross.OneBasedStartResidueInProtein.Value + item.BetaPsmCross.XlPos - 1; } } //Write Inter Psms FDR var interPsmsXL = allPsmsXL.Where(p => !p.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First().Contains(p.BetaPsmCross.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First()) && !p.BetaPsmCross.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First().Contains(p.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First())).OrderByDescending(p => p.XLQvalueTotalScore).ToList(); foreach (var item in interPsmsXL) { item.CrossType = PsmCrossType.Inter; } var interPsmsXLFDR = CrosslinkDoFalseDiscoveryRateAnalysis(interPsmsXL).ToList(); //var interPsmsXLFDR = CrosslinkFDRAnalysis(interPsmsXL).ToList(); if (XlSearchParameters.XlOutCrosslink) { WriteCrosslinkToTsv(interPsmsXLFDR, OutputFolder, "xl_inter_fdr", new List <string> { taskId }); } if (XlSearchParameters.XlOutPercolator) { try { var interPsmsXLPercolator = interPsmsXL.Where(p => p.XLBestScore >= 2 && p.BetaPsmCross.XLBestScore >= 2).OrderBy(p => p.ScanNumber).ToList(); WriteCrosslinkToTxtForPercolator(interPsmsXLPercolator, OutputFolder, "xl_inter_perc", crosslinker, new List <string> { taskId }); } catch (Exception) { throw; } } //Write Intra Psms FDR var intraPsmsXL = allPsmsXL.Where(p => p.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First() == p.BetaPsmCross.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First() || p.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First().Contains(p.BetaPsmCross.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First()) || p.BetaPsmCross.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First().Contains(p.CompactPeptides.First().Value.Item2.Select(b => b.Protein.Accession).First())).OrderByDescending(p => p.XLQvalueTotalScore).ToList(); foreach (var item in intraPsmsXL) { item.CrossType = PsmCrossType.Intra; } var intraPsmsXLFDR = CrosslinkDoFalseDiscoveryRateAnalysis(intraPsmsXL).ToList(); //var intraPsmsXLFDR = CrosslinkFDRAnalysis(intraPsmsXL).ToList(); if (XlSearchParameters.XlOutCrosslink) { WriteCrosslinkToTsv(intraPsmsXLFDR, OutputFolder, "xl_intra_fdr", new List <string> { taskId }); } if (XlSearchParameters.XlOutPercolator) { try { var intraPsmsXLPercolator = intraPsmsXL.Where(p => p.XLBestScore >= 2 && p.BetaPsmCross.XLBestScore >= 2).OrderBy(p => p.ScanNumber).ToList(); WriteCrosslinkToTxtForPercolator(intraPsmsXLPercolator, OutputFolder, "xl_intra_perc", crosslinker, new List <string> { taskId }); } catch (Exception) { throw; } } var singlePsms = allPsms.Where(p => p.CrossType == PsmCrossType.Singe && p.FullSequence != null && !p.FullSequence.Contains("Crosslink")).OrderByDescending(p => p.Score).ToList(); var singlePsmsFDR = SingleFDRAnalysis(singlePsms).ToList(); if (XlSearchParameters.XlOutAll) { WriteSingleToTsv(singlePsmsFDR, OutputFolder, "single_fdr", new List <string> { taskId }); } var loopPsms = allPsms.Where(p => p.CrossType == PsmCrossType.Loop).OrderByDescending(p => p.XLTotalScore).ToList(); var loopPsmsFDR = SingleFDRAnalysis(loopPsms).ToList(); if (XlSearchParameters.XlOutAll) { WriteSingleToTsv(loopPsmsFDR, OutputFolder, "loop_fdr", new List <string> { taskId }); } var deadendPsms = allPsms.Where(p => p.CrossType == PsmCrossType.DeadEnd || p.CrossType == PsmCrossType.DeadEndH2O || p.CrossType == PsmCrossType.DeadEndNH2 || p.CrossType == PsmCrossType.DeadEndTris).OrderByDescending(p => p.XLTotalScore).ToList(); deadendPsms.AddRange(allPsms.Where(p => p.CrossType == PsmCrossType.Singe && p.FullSequence != null && p.FullSequence.Contains("Crosslink")).ToList()); var deadendPsmsFDR = SingleFDRAnalysis(deadendPsms).ToList(); if (XlSearchParameters.XlOutAll) { WriteSingleToTsv(deadendPsmsFDR, OutputFolder, "deadend_fdr", new List <string> { taskId }); } if (XlSearchParameters.XlOutPepXML) { List <PsmCross> allPsmsFDR = new List <PsmCross>(); allPsmsFDR.AddRange(intraPsmsXLFDR.Where(p => p.IsDecoy != true && p.BetaPsmCross.IsDecoy != true && p.FdrInfo.QValue <= 0.05).ToList()); allPsmsFDR.AddRange(interPsmsXLFDR.Where(p => p.IsDecoy != true && p.BetaPsmCross.IsDecoy != true && p.FdrInfo.QValue <= 0.05).ToList()); allPsmsFDR.AddRange(singlePsmsFDR.Where(p => p.IsDecoy != true && p.FdrInfo.QValue <= 0.05).ToList()); allPsmsFDR.AddRange(loopPsmsFDR.Where(p => p.IsDecoy != true && p.FdrInfo.QValue <= 0.05).ToList()); allPsmsFDR.AddRange(deadendPsmsFDR.Where(p => p.IsDecoy != true && p.FdrInfo.QValue <= 0.05).ToList()); allPsmsFDR = allPsmsFDR.OrderBy(p => p.ScanNumber).ToList(); foreach (var fullFilePath in currentRawFileList) { string fileNameNoExtension = Path.GetFileNameWithoutExtension(fullFilePath); WritePepXML_xl(allPsmsFDR.Where(p => p.FullFilePath == fullFilePath).ToList(), proteinList, dbFilenameList[0].FilePath, variableModifications, fixedModifications, localizeableModificationTypes, OutputFolder, fileNameNoExtension, new List <string> { taskId }); } } if (XlSearchParameters.XlOutAll) { List <PsmCross> allPsmsXLFDR = new List <PsmCross>(); allPsmsXLFDR.AddRange(intraPsmsXLFDR.Where(p => p.IsDecoy != true && p.BetaPsmCross.IsDecoy != true && p.FdrInfo.QValue <= 0.05).ToList()); allPsmsXLFDR.AddRange(interPsmsXLFDR.Where(p => p.IsDecoy != true && p.BetaPsmCross.IsDecoy != true && p.FdrInfo.QValue <= 0.05).ToList()); try { allPsmsXLFDR = allPsmsXLFDR.OrderByDescending(p => p.XLQvalueTotalScore).ToList(); var allPsmsXLFDRGroup = FindCrosslinks(allPsmsXLFDR); WriteCrosslinkToTsv(allPsmsXLFDRGroup, OutputFolder, "allPsmsXLFDRGroup", new List <string> { taskId }); } catch (Exception) { throw; } } return(MyTaskResults); }
public CrosslinkAnalysisResults(CrosslinkAnalysisEngine s) : base(s) { }