public static List <CrosslinkSpectralMatch> SortOneListCsmsSetSecondBestScore(List <CrosslinkSpectralMatch> csmsPerScan, CommonParameters commonParameters)
{
List <double> xlTotalScores = csmsPerScan.Select(s => s.XLTotalScore).OrderByDescending(s => s).ToList();
xlTotalScores.RemoveAt(0);
//This possibly needs to be doubled for xlinks. But, since each list can be a mix of xlinks and nonxlinks we just leave as is for now.
double secondBestSore = commonParameters.ScoreCutoff;
if (xlTotalScores.Count() > 0)
{
secondBestSore = xlTotalScores[0];
}
foreach (CrosslinkSpectralMatch csm in csmsPerScan)
{
csm.SecondBestXlScore = secondBestSore;
}
csmsPerScan = csmsPerScan.OrderByDescending(c => c.XLTotalScore).ThenBy(c => c.FullSequence + (c.BetaPeptide != null ? c.BetaPeptide.FullSequence : "")).ToList();
return(csmsPerScan);
}
public static void WritePepXML_xl(List <CrosslinkSpectralMatch> items, List <Protein> proteinList, string databasePath, List <Modification> variableModifications, List <Modification> fixedModifications, List <string> localizeableModificationTypes, string outputFolder, string fileName, CommonParameters CommonParameters, XlSearchParameters XlSearchParameters)
{
if (!items.Any())
{
return;
}
XmlSerializer _indexedSerializer = new XmlSerializer(typeof(pepXML.Generated.msms_pipeline_analysis));
var _pepxml = new pepXML.Generated.msms_pipeline_analysis();
_pepxml.date = DateTime.Now;
_pepxml.summary_xml = items[0].FullFilePath + ".pep.XML";
string proteaseC = ""; string proteaseNC = "";
foreach (var x in CommonParameters.DigestionParams.Protease.DigestionMotifs.Select(m => m.InducingCleavage))
{
proteaseC += x;
}
foreach (var x in CommonParameters.DigestionParams.Protease.DigestionMotifs.Select(m => m.PreventingCleavage))
{
proteaseNC += x;
}
Crosslinker crosslinker = XlSearchParameters.Crosslinker;
string fileNameNoExtension = Path.GetFileNameWithoutExtension(items[0].FullFilePath);
string filePathNoExtension = Path.ChangeExtension(items[0].FullFilePath, null);
string modSites = crosslinker.CrosslinkerModSites.ToCharArray().Concat(crosslinker.CrosslinkerModSites2.ToCharArray()).Distinct().ToString();
var para = new List <pepXML.Generated.nameValueType>();
{
para.Add(new pepXML.Generated.nameValueType {
name = "threads", value = CommonParameters.MaxThreadsToUsePerFile.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "database", value = databasePath
});
para.Add(new pepXML.Generated.nameValueType {
name = "MS_data_file", value = items[0].FullFilePath
});
para.Add(new pepXML.Generated.nameValueType {
name = "Cross-link precursor Mass Tolerance", value = CommonParameters.PrecursorMassTolerance.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Cross-linker type", value = crosslinker.CrosslinkerName
});
para.Add(new pepXML.Generated.nameValueType {
name = "Cross-linker mass", value = crosslinker.TotalMass.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Cross-linker cleavable", value = crosslinker.Cleavable.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Cross-linker cleavable long mass", value = crosslinker.CleaveMassLong.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Cross-linker cleavable short mass", value = crosslinker.CleaveMassShort.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Cross-linker xl site", value = modSites
});
para.Add(new pepXML.Generated.nameValueType {
name = "Generate decoy proteins", value = XlSearchParameters.DecoyType.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "MaxMissed Cleavages", value = CommonParameters.DigestionParams.MaxMissedCleavages.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Protease", value = CommonParameters.DigestionParams.Protease.Name
});
para.Add(new pepXML.Generated.nameValueType {
name = "Initiator Methionine", value = CommonParameters.DigestionParams.InitiatorMethionineBehavior.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Max Modification Isoforms", value = CommonParameters.DigestionParams.MaxModificationIsoforms.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Min Peptide Len", value = CommonParameters.DigestionParams.MinPeptideLength.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Max Peptide Len", value = CommonParameters.DigestionParams.MaxPeptideLength.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Product Mass Tolerance", value = CommonParameters.ProductMassTolerance.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Ions to search", value = String.Join(", ", DissociationTypeCollection.ProductsFromDissociationType[CommonParameters.DissociationType])
});
foreach (var fixedMod in fixedModifications)
{
para.Add(new pepXML.Generated.nameValueType {
name = "Fixed Modifications: " + fixedMod.IdWithMotif, value = fixedMod.MonoisotopicMass.ToString()
});
}
foreach (var variableMod in variableModifications)
{
para.Add(new pepXML.Generated.nameValueType {
name = "Variable Modifications: " + variableMod.IdWithMotif, value = variableMod.MonoisotopicMass.ToString()
});
}
para.Add(new pepXML.Generated.nameValueType {
name = "Localize All Modifications", value = "true"
});
}
_pepxml.msms_run_summary = new pepXML.Generated.msms_pipeline_analysisMsms_run_summary[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summary
{
base_name = filePathNoExtension,
raw_data_type = "raw",
raw_data = ".mzML",
sample_enzyme = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySample_enzyme()
{
name = CommonParameters.DigestionParams.Protease.Name,
specificity = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySample_enzymeSpecificity[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySample_enzymeSpecificity
{
cut = proteaseC,
no_cut = proteaseNC,
}
}
},
search_summary = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summary[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summary
{
base_name = filePathNoExtension,
search_engine_version = GlobalVariables.MetaMorpheusVersion,
precursor_mass_type = pepXML.Generated.massType.monoisotopic,
fragment_mass_type = pepXML.Generated.massType.monoisotopic,
search_id = 1,
search_database = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summarySearch_database
{
local_path = databasePath,
type = pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summarySearch_databaseType.AA,
},
enzymatic_search_constraint = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summaryEnzymatic_search_constraint
{
enzyme = CommonParameters.DigestionParams.Protease.Name,
max_num_internal_cleavages = CommonParameters.DigestionParams.MaxMissedCleavages.ToString(),
//min_number_termini = "2"
},
parameter = para.ToArray()
}
},
}
};
_pepxml.msms_run_summary[0].spectrum_query = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_query[items.Count];
var searchHits = new List <pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit>();
for (int i = 0; i < items.Count; i++)
{
var mods = new List <pepXML.Generated.modInfoDataTypeMod_aminoacid_mass>();
var alphaPeptide = items[i].BestMatchingPeptides.First().Peptide;
foreach (var modification in alphaPeptide.AllModsOneIsNterminus)
{
var mod = new pepXML.Generated.modInfoDataTypeMod_aminoacid_mass
{
mass = modification.Value.MonoisotopicMass.Value,
// convert from one-based to zero-based (N-term is zero in the pepXML output)
position = (modification.Key - 1).ToString()
};
mods.Add(mod);
}
if (items[i].CrossType == PsmCrossType.Single)
{
var searchHit = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit
{
hit_rank = 1,
peptide = alphaPeptide.BaseSequence,
peptide_prev_aa = alphaPeptide.PreviousAminoAcid.ToString(),
peptide_next_aa = alphaPeptide.NextAminoAcid.ToString(),
protein = alphaPeptide.Protein.Accession,
num_tot_proteins = 1,
calc_neutral_pep_mass = (float)items[i].ScanPrecursorMass,
massdiff = (items[i].ScanPrecursorMass - items[i].PeptideMonisotopicMass.Value).ToString(),
xlink_typeSpecified = true,
xlink_type = pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlink_type.na,
modification_info = new pepXML.Generated.modInfoDataType {
mod_aminoacid_mass = mods.ToArray()
},
search_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "xlTotalScore", value = items[i].XLTotalScore.ToString()
},
new pepXML.Generated.nameValueType {
name = "Qvalue", value = items[i].FdrInfo.QValue.ToString()
}
},
};
searchHits.Add(searchHit);
}
else if (items[i].CrossType == PsmCrossType.DeadEnd || items[i].CrossType == PsmCrossType.DeadEndH2O || items[i].CrossType == PsmCrossType.DeadEndNH2 || items[i].CrossType == PsmCrossType.DeadEndTris)
{
double crosslinkerDeadEndMass = 0;
switch (items[i].CrossType)
{
case PsmCrossType.DeadEndNH2:
crosslinkerDeadEndMass = crosslinker.DeadendMassNH2;
break;
case PsmCrossType.DeadEndTris:
crosslinkerDeadEndMass = crosslinker.DeadendMassTris;
break;
default:
crosslinkerDeadEndMass = crosslinker.DeadendMassH2O;
break;
}
var searchHit = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit
{
hit_rank = 1,
peptide = alphaPeptide.BaseSequence,
peptide_prev_aa = alphaPeptide.PreviousAminoAcid.ToString(),
peptide_next_aa = alphaPeptide.NextAminoAcid.ToString(),
protein = alphaPeptide.Protein.Accession,
num_tot_proteins = 1,
calc_neutral_pep_mass = (float)items[i].ScanPrecursorMass,
massdiff = (items[i].ScanPrecursorMass - items[i].PeptideMonisotopicMass.Value - crosslinkerDeadEndMass).ToString(),
xlink_typeSpecified = true,
xlink_type = pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlink_type.na,
modification_info = new pepXML.Generated.modInfoDataType {
mod_aminoacid_mass = mods.ToArray()
},
search_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "xlTotalScore", value = items[i].XLTotalScore.ToString()
},
new pepXML.Generated.nameValueType {
name = "Qvalue", value = items[i].FdrInfo.QValue.ToString()
}
},
};
searchHits.Add(searchHit);
}
else if (items[i].CrossType == PsmCrossType.Inter || items[i].CrossType == PsmCrossType.Intra || items[i].CrossType == PsmCrossType.Cross)
{
var betaPeptide = items[i].BetaPeptide.BestMatchingPeptides.First().Peptide;
var modsBeta = new List <pepXML.Generated.modInfoDataTypeMod_aminoacid_mass>();
foreach (var mod in betaPeptide.AllModsOneIsNterminus)
{
var modBeta = new pepXML.Generated.modInfoDataTypeMod_aminoacid_mass
{
mass = mod.Value.MonoisotopicMass.Value,
// convert from one-based to zero-based (N-term is zero in the pepXML output)
position = (mod.Key - 1).ToString()
};
modsBeta.Add(modBeta);
}
var alpha = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlinkLinked_peptide
{
peptide = alphaPeptide.BaseSequence,
peptide_prev_aa = alphaPeptide.PreviousAminoAcid.ToString(),
peptide_next_aa = alphaPeptide.NextAminoAcid.ToString(),
protein = alphaPeptide.Protein.Accession,
num_tot_proteins = 1,
calc_neutral_pep_mass = (float)items[i].PeptideMonisotopicMass.Value,
complement_mass = (float)(items[i].ScanPrecursorMass - alphaPeptide.MonoisotopicMass),
designation = "alpha",
modification_info = new pepXML.Generated.modInfoDataType {
mod_aminoacid_mass = mods.ToArray()
},
xlink_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "xlscore", value = items[i].XLTotalScore.ToString()
},
new pepXML.Generated.nameValueType {
name = "link", value = items[i].LinkPositions.First().ToString()
},
}
};
var beta = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlinkLinked_peptide
{
peptide = betaPeptide.BaseSequence,
peptide_prev_aa = betaPeptide.PreviousAminoAcid.ToString(),
peptide_next_aa = betaPeptide.NextAminoAcid.ToString(),
protein = betaPeptide.Protein.Accession,
num_tot_proteins = 1,
calc_neutral_pep_mass = (float)betaPeptide.MonoisotopicMass,
complement_mass = (float)(items[i].ScanPrecursorMass - betaPeptide.MonoisotopicMass),
designation = "beta",
modification_info = new pepXML.Generated.modInfoDataType {
mod_aminoacid_mass = modsBeta.ToArray()
},
xlink_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "xlscore", value = items[i].BetaPeptide.Score.ToString()
},
new pepXML.Generated.nameValueType {
name = "link", value = items[i].BetaPeptide.LinkPositions.First().ToString()
},
}
};
var cross = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlinkLinked_peptide[2] {
alpha, beta
};
var searchHit = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit
{
hit_rank = 1,
peptide = "-",
peptide_prev_aa = "-",
peptide_next_aa = "-",
protein = "-",
num_tot_proteins = 1,
calc_neutral_pep_mass = (float)items[i].ScanPrecursorMass,
massdiff = (items[i].ScanPrecursorMass - betaPeptide.MonoisotopicMass - alphaPeptide.MonoisotopicMass - crosslinker.TotalMass).ToString(),
xlink_typeSpecified = true,
xlink_type = pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlink_type.xl,
xlink = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlink
{
identifier = crosslinker.CrosslinkerName,
mass = (float)crosslinker.TotalMass,
linked_peptide = cross
},
search_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "xlTotalScore", value = items[i].XLTotalScore.ToString()
},
new pepXML.Generated.nameValueType {
name = "Qvalue", value = items[i].FdrInfo.QValue.ToString()
}
}
};
searchHits.Add(searchHit);
}
else if (items[i].CrossType == PsmCrossType.Loop)
{
var thePeptide = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlinkLinked_peptide
{
xlink_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "link", value = items[i].LinkPositions.First().ToString()
},
new pepXML.Generated.nameValueType {
name = "link", value = items[i].LinkPositions[1].ToString()
}
}
};
var cross = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlinkLinked_peptide[1] {
thePeptide
};
var searchHit = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit
{
hit_rank = 1,
peptide = alphaPeptide.BaseSequence,
peptide_prev_aa = alphaPeptide.PreviousAminoAcid.ToString(),
peptide_next_aa = alphaPeptide.NextAminoAcid.ToString(),
protein = alphaPeptide.Protein.Accession,
num_tot_proteins = 1,
calc_neutral_pep_mass = (float)items[i].ScanPrecursorMass,
massdiff = (items[i].ScanPrecursorMass - alphaPeptide.MonoisotopicMass - crosslinker.LoopMass).ToString(),
xlink_typeSpecified = true,
xlink_type = pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlink_type.loop,
modification_info = new pepXML.Generated.modInfoDataType {
mod_aminoacid_mass = mods.ToArray()
},
xlink = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hitXlink
{
identifier = crosslinker.CrosslinkerName,
mass = (float)crosslinker.TotalMass,
linked_peptide = cross
},
search_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "xlTotalScore", value = items[i].XLTotalScore.ToString()
},
new pepXML.Generated.nameValueType {
name = "Qvalue", value = items[i].FdrInfo.QValue.ToString()
}
}
};
searchHits.Add(searchHit);
}
}
for (int i = 0; i < items.Count; i++)
{
_pepxml.msms_run_summary[0].spectrum_query[i] = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_query()
{
spectrum = fileNameNoExtension + "." + items[i].ScanNumber.ToString(),
start_scan = Convert.ToUInt32(items[i].ScanNumber),
end_scan = Convert.ToUInt32(items[i].ScanNumber),
precursor_neutral_mass = (float)items[i].ScanPrecursorMass,
assumed_charge = items[i].ScanPrecursorCharge.ToString(),
index = Convert.ToUInt32(i + 1),
retention_time_sec = (float)(items[i].ScanRetentionTime * 60),
search_result = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_result[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_result
{
search_hit = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit[1]
{
searchHits[i]
}
}
}
};
}
TextWriter writer = new StreamWriter(Path.Combine(outputFolder, fileName + ".pep.XML"));
_indexedSerializer.Serialize(writer, _pepxml);
writer.Close();
}
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());
}
public static List <List <CrosslinkSpectralMatch> > SortListsOfCsms(List <List <CrosslinkSpectralMatch> > ListOfCsmsPerMS2Scan, CommonParameters commonParameters)
{
List <List <CrosslinkSpectralMatch> > newLists = new List <List <CrosslinkSpectralMatch> >();
foreach (var csmsPerScan in ListOfCsmsPerMS2Scan)
{
newLists.Add(SortOneListCsmsSetSecondBestScore(csmsPerScan, commonParameters));
}
return(newLists.OrderByDescending(c => c.First().XLTotalScore).ThenByDescending(c => c.First().FullSequence + (c.First().BetaPeptide != null ? c.First().BetaPeptide.FullSequence : "")).ToList());
}
public void ComputeXlinkQandPValues(List <CrosslinkSpectralMatch> allPsms, List <CrosslinkSpectralMatch> intraCsms, List <CrosslinkSpectralMatch> interCsms, CommonParameters commonParameters, string taskId)
{
List <CrosslinkSpectralMatch> crossCsms = allPsms.Where(p => p.CrossType == PsmCrossType.Inter || p.CrossType == PsmCrossType.Intra).OrderByDescending(p => p.XLTotalScore).ToList();
new FdrAnalysisEngine(crossCsms.ToList <PeptideSpectralMatch>(), 0, commonParameters, this.FileSpecificParameters, new List <string> {
taskId
}, "crosslink").Run();
List <CrosslinkSpectralMatch> singles = allPsms.Where(p => p.CrossType != PsmCrossType.Inter).Where(p => p.CrossType != PsmCrossType.Intra).OrderByDescending(p => p.Score).ToList();
new FdrAnalysisEngine(singles.ToList <PeptideSpectralMatch>(), 0, commonParameters, this.FileSpecificParameters, new List <string> {
taskId
}, "PSM").Run();
SingleFDRAnalysis(singles, commonParameters, new List <string> {
taskId
});
// calculate FDR
DoCrosslinkFdrAnalysis(interCsms);
DoCrosslinkFdrAnalysis(intraCsms);
}
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, 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.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);
FinishedWritingFile(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);
}
private DataPointAquisitionResults GetDataAcquisitionResults(MsDataFile myMsDataFile, string currentDataFile, List <Modification> variableModifications, List <Modification> fixedModifications, List <Protein> proteinList, string taskId, CommonParameters combinedParameters, Tolerance initPrecTol, Tolerance initProdTol)
{
var fileNameWithoutExtension = Path.GetFileNameWithoutExtension(currentDataFile);
MassDiffAcceptor searchMode = initPrecTol is PpmTolerance ?
(MassDiffAcceptor) new SinglePpmAroundZeroSearchMode(initPrecTol.Value) :
new SingleAbsoluteAroundZeroSearchMode(initPrecTol.Value);
var listOfSortedms2Scans = GetMs2Scans(myMsDataFile, currentDataFile, combinedParameters).OrderBy(b => b.PrecursorMass).ToArray();
PeptideSpectralMatch[] allPsmsArray = new PeptideSpectralMatch[listOfSortedms2Scans.Length];
Log("Searching with searchMode: " + searchMode, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
});
Log("Searching with productMassTolerance: " + initProdTol, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
});
new ClassicSearchEngine(allPsmsArray, listOfSortedms2Scans, variableModifications, fixedModifications, null, null, null, proteinList, searchMode, combinedParameters, FileSpecificParameters, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
List <PeptideSpectralMatch> allPsms = allPsmsArray.Where(b => b != null).ToList();
allPsms = allPsms.OrderByDescending(b => b.Score)
.ThenBy(b => b.PeptideMonisotopicMass.HasValue ? Math.Abs(b.ScanPrecursorMass - b.PeptideMonisotopicMass.Value) : double.MaxValue)
.GroupBy(b => (b.FullFilePath, b.ScanNumber, b.PeptideMonisotopicMass)).Select(b => b.First()).ToList();
new FdrAnalysisEngine(allPsms, searchMode.NumNotches, CommonParameters, FileSpecificParameters, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
List <PeptideSpectralMatch> goodIdentifications = allPsms.Where(b => b.FdrInfo.QValueNotch < 0.001 && !b.IsDecoy && b.FullSequence != null).ToList();
if (!goodIdentifications.Any())
{
return(new DataPointAquisitionResults(null, new List <PeptideSpectralMatch>(), new List <LabeledDataPoint>(), new List <LabeledDataPoint>(), 0, 0, 0, 0));
}
//get the deconvoluted ms2scans for the good identifications
List <Ms2ScanWithSpecificMass> goodScans = new List <Ms2ScanWithSpecificMass>();
List <PeptideSpectralMatch> unfilteredPsms = allPsmsArray.ToList();
foreach (PeptideSpectralMatch psm in goodIdentifications)
{
goodScans.Add(listOfSortedms2Scans[unfilteredPsms.IndexOf(psm)]);
}
DataPointAquisitionResults currentResult = (DataPointAquisitionResults) new DataPointAcquisitionEngine(
goodIdentifications,
goodScans,
myMsDataFile,
initPrecTol,
initProdTol,
CalibrationParameters.MinMS1IsotopicPeaksNeededForConfirmedIdentification,
CommonParameters,
FileSpecificParameters,
new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
return(currentResult);
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
MyTaskResults = new MyTaskResults(this);
List <List <CrosslinkSpectralMatch> > ListOfCsmsPerMS2Scan = new List <List <CrosslinkSpectralMatch> >();
LoadModifications(taskId, out var variableModifications, out var fixedModifications, out var localizeableModificationTypes);
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, XlSearchParameters.DecoyType, localizeableModificationTypes, CommonParameters);
var crosslinker = XlSearchParameters.Crosslinker;
MyFileManager myFileManager = new MyFileManager(true);
var fileSpecificCommonParams = fileSettingsList.Select(b => SetAllFileSpecificCommonParams(CommonParameters, b));
HashSet <DigestionParams> ListOfDigestionParams = new HashSet <DigestionParams>(fileSpecificCommonParams.Select(p => p.DigestionParams));
int completedFiles = 0;
object indexLock = new object();
object psmLock = new object();
Status("Searching files...", taskId);
ProseCreatedWhileRunning.Append("The following crosslink discovery were used: ");
ProseCreatedWhileRunning.Append("crosslinker name = " + crosslinker.CrosslinkerName + "; ");
ProseCreatedWhileRunning.Append("crosslinker type = " + crosslinker.Cleavable + "; ");
ProseCreatedWhileRunning.Append("crosslinker mass = " + crosslinker.TotalMass + "; ");
ProseCreatedWhileRunning.Append("crosslinker modification site(s) = " + crosslinker.CrosslinkerModSites + "; ");
ProseCreatedWhileRunning.Append("protease = " + CommonParameters.DigestionParams.Protease + "; ");
ProseCreatedWhileRunning.Append("maximum missed cleavages = " + CommonParameters.DigestionParams.MaxMissedCleavages + "; ");
ProseCreatedWhileRunning.Append("minimum peptide length = " + CommonParameters.DigestionParams.MinPeptideLength + "; ");
ProseCreatedWhileRunning.Append(CommonParameters.DigestionParams.MaxPeptideLength == int.MaxValue ?
"maximum peptide length = unspecified; " :
"maximum peptide length = " + CommonParameters.DigestionParams.MaxPeptideLength + "; ");
ProseCreatedWhileRunning.Append("initiator methionine behavior = " + CommonParameters.DigestionParams.InitiatorMethionineBehavior + "; ");
ProseCreatedWhileRunning.Append("max modification isoforms = " + CommonParameters.DigestionParams.MaxModificationIsoforms + "; ");
ProseCreatedWhileRunning.Append("fixed modifications = " + string.Join(", ", fixedModifications.Select(m => m.IdWithMotif) + "; "));
ProseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.IdWithMotif)) + "; ");
ProseCreatedWhileRunning.Append("parent mass tolerance(s) = " + CommonParameters.PrecursorMassTolerance + "; ");
ProseCreatedWhileRunning.Append("product mass tolerance = " + CommonParameters.ProductMassTolerance + "; ");
ProseCreatedWhileRunning.Append("The combined search database contained " + proteinList.Count + " total entries including " + proteinList.Where(p => p.IsContaminant).Count() + " contaminant sequences. ");
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
var origDataFile = currentRawFileList[spectraFileIndex];
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
var thisId = new List <string> {
taskId, "Individual Spectra Files", origDataFile
};
NewCollection(Path.GetFileName(origDataFile), thisId);
Status("Loading spectra file...", thisId);
MsDataFile myMsDataFile = myFileManager.LoadFile(origDataFile, combinedParams);
Status("Getting ms2 scans...", thisId);
Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass = GetMs2Scans(myMsDataFile, origDataFile, combinedParams).OrderBy(b => b.PrecursorMass).ToArray();
List <CrosslinkSpectralMatch>[] newCsmsPerMS2ScanPerFile = new List <CrosslinkSpectralMatch> [arrayOfMs2ScansSortedByMass.Length];
for (int currentPartition = 0; currentPartition < CommonParameters.TotalPartitions; currentPartition++)
{
List <PeptideWithSetModifications> peptideIndex = null;
//When partition, the proteinList will be split for each Thread.
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
});
//Only reverse Decoy for crosslink search has been tested and are set as fixed parameter.
var indexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, null, null, null, currentPartition, UsefulProteomicsDatabases.DecoyType.Reverse, combinedParams, 30000.0, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), new List <string> {
taskId
});
List <int>[] fragmentIndex = null;
List <int>[] precursorIndex = null;
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, ref precursorIndex, proteinList, taskId);
//The second Fragment index is for 'MS1-HCD_MS1-ETD_MS2s' type of data. If LowCID is used for MS1, ion-index is not allowed to use.
List <int>[] secondFragmentIndex = null;
if (combinedParams.ChildScanDissociationType != DissociationType.LowCID &&
!CrosslinkSearchEngine.DissociationTypeGenerateSameTypeOfIons(combinedParams.DissociationType, combinedParams.ChildScanDissociationType))
{
//Becuase two different type of dissociation methods are used, the parameters are changed with different dissociation type.
var secondCombinedParams = CommonParameters.CloneWithNewDissociationType(combinedParams.ChildScanDissociationType);
var secondIndexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, null, null, null, currentPartition, UsefulProteomicsDatabases.DecoyType.Reverse, secondCombinedParams, 30000.0, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), new List <string> {
taskId
});
GenerateSecondIndexes(indexEngine, secondIndexEngine, dbFilenameList, ref secondFragmentIndex, proteinList, taskId);
}
Status("Searching files...", taskId);
new CrosslinkSearchEngine(newCsmsPerMS2ScanPerFile, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, secondFragmentIndex, currentPartition, combinedParams, crosslinker,
XlSearchParameters.RestrictToTopNHits, XlSearchParameters.CrosslinkSearchTopNum, XlSearchParameters.CrosslinkAtCleavageSite,
XlSearchParameters.XlQuench_H2O, XlSearchParameters.XlQuench_NH2, XlSearchParameters.XlQuench_Tris, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + CommonParameters.TotalPartitions + "!", thisId));
if (GlobalVariables.StopLoops)
{
break;
}
}
ListOfCsmsPerMS2Scan.AddRange(newCsmsPerMS2ScanPerFile.Where(p => p != null).ToList());
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"
}));
List <List <CrosslinkSpectralMatch> > ListOfCsmsPerMS2ScanParsimony = new List <List <CrosslinkSpectralMatch> >();
//For every Ms2Scans, each have a list of candidates psms. The allPsms from CrosslinkSearchEngine is the list (all ms2scans) of list (each ms2scan) of psm (all candidate psm).
//The allPsmsList is same as allPsms after ResolveAmbiguities.
foreach (var csmsPerScan in ListOfCsmsPerMS2Scan)
{
foreach (var csm in csmsPerScan)
{
csm.ResolveAllAmbiguities();
if (csm.BetaPeptide != null)
{
csm.BetaPeptide.ResolveAllAmbiguities();
}
}
ListOfCsmsPerMS2ScanParsimony.Add(RemoveDuplicateFromCsmsPerScan(csmsPerScan));
}
var filteredAllPsms = new List <CrosslinkSpectralMatch>();
//For each ms2scan, try to find the best candidate psm from the psms list. Add it into filteredAllPsms
//This function is for current usage, this can be replaced with PEP value.
foreach (var csmsPerScan in ListOfCsmsPerMS2ScanParsimony)
{
filteredAllPsms.Add(csmsPerScan[0]);
}
PostXLSearchAnalysisTask postXLSearchAnalysisTask = new PostXLSearchAnalysisTask();
return(postXLSearchAnalysisTask.Run(OutputFolder, dbFilenameList, currentRawFileList, taskId, fileSettingsList, filteredAllPsms.OrderByDescending(p => p.XLTotalScore).ToList(), CommonParameters, XlSearchParameters, proteinList, variableModifications, fixedModifications, localizeableModificationTypes, MyTaskResults));
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
// load modifications
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 = GlobalVariables.AllModTypesKnown.ToList();
// what types of fragment ions to search for
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);
}
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, DecoyType.Reverse, localizeableModificationTypes, CommonParameters);
// write prose settings
ProseCreatedWhileRunning.Append("The following calibration 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("The combined search database contained " + proteinList.Count(p => !p.IsDecoy) + " non-decoy protein entries including " + proteinList.Count(p => p.IsContaminant) + " contaminant sequences. ");
// start the calibration task
Status("Calibrating...", new List <string> {
taskId
});
MyTaskResults = new MyTaskResults(this)
{
NewSpectra = new List <string>(),
NewFileSpecificTomls = new List <string>()
};
object lock1 = new object();
var myFileManager = new MyFileManager(true);
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
if (GlobalVariables.StopLoops)
{
break;
}
// get filename stuff
var originalUncalibratedFilePath = currentRawFileList[spectraFileIndex];
var originalUncalibratedFilenameWithoutExtension = Path.GetFileNameWithoutExtension(originalUncalibratedFilePath);
string calibratedFilePath = Path.Combine(OutputFolder, originalUncalibratedFilenameWithoutExtension + CalibSuffix + ".mzML");
// mark the file as in-progress
StartingDataFile(originalUncalibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilePath
});
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
MsDataFile myMsDataFile;
// load the file
Status("Loading spectra file...", new List <string> {
taskId, "Individual Spectra Files"
});
lock (lock1)
{
myMsDataFile = myFileManager.LoadFile(originalUncalibratedFilePath, CommonParameters.TopNpeaks, CommonParameters.MinRatio, CommonParameters.TrimMs1Peaks, CommonParameters.TrimMsMsPeaks, CommonParameters);
}
// get datapoints to fit calibration function to
Status("Acquiring calibration data points...", new List <string> {
taskId, "Individual Spectra Files"
});
DataPointAquisitionResults acquisitionResults = null;
for (int i = 1; i <= 5; i++)
{
acquisitionResults = GetDataAcquisitionResults(myMsDataFile, originalUncalibratedFilePath, variableModifications, fixedModifications, proteinList, taskId, combinedParams, combinedParams.PrecursorMassTolerance, combinedParams.ProductMassTolerance);
// enough data points to calibrate?
if (acquisitionResults.Psms.Count >= NumRequiredPsms && acquisitionResults.Ms1List.Count > NumRequiredMs1Datapoints && acquisitionResults.Ms2List.Count > NumRequiredMs2Datapoints)
{
break;
}
if (i == 1) // failed round 1
{
CommonParameters.PrecursorMassTolerance = new PpmTolerance(20);
CommonParameters.ProductMassTolerance = new PpmTolerance(50);
}
else if (i == 2) // failed round 2
{
CommonParameters.PrecursorMassTolerance = new PpmTolerance(30);
CommonParameters.ProductMassTolerance = new PpmTolerance(100);
}
else if (i == 3) // failed round 3
{
CommonParameters.PrecursorMassTolerance = new PpmTolerance(40);
CommonParameters.ProductMassTolerance = new PpmTolerance(150);
}
else // failed round 4
{
if (acquisitionResults.Psms.Count < NumRequiredPsms)
{
Warn("Calibration failure! Could not find enough high-quality PSMs. Required " + NumRequiredPsms + ", saw " + acquisitionResults.Psms.Count);
}
if (acquisitionResults.Ms1List.Count < NumRequiredMs1Datapoints)
{
Warn("Calibration failure! Could not find enough MS1 datapoints. Required " + NumRequiredMs1Datapoints + ", saw " + acquisitionResults.Ms1List.Count);
}
if (acquisitionResults.Ms2List.Count < NumRequiredMs2Datapoints)
{
Warn("Calibration failure! Could not find enough MS2 datapoints. Required " + NumRequiredMs2Datapoints + ", saw " + acquisitionResults.Ms2List.Count);
}
FinishedDataFile(originalUncalibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilePath
});
return(MyTaskResults);
}
Warn("Could not find enough PSMs to calibrate with; opening up tolerances to " +
Math.Round(CommonParameters.PrecursorMassTolerance.Value, 2) + " ppm precursor and " +
Math.Round(CommonParameters.ProductMassTolerance.Value, 2) + " ppm product");
}
// stats before calibration
int prevPsmCount = acquisitionResults.Psms.Count;
double preCalibrationPrecursorErrorIqr = acquisitionResults.PsmPrecursorIqrPpmError;
double preCalibrationProductErrorIqr = acquisitionResults.PsmProductIqrPpmError;
// generate calibration function and shift data points
Status("Calibrating...", new List <string> {
taskId, "Individual Spectra Files"
});
new CalibrationEngine(myMsDataFile, acquisitionResults, CommonParameters, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
}).Run();
// do another search to evaluate calibration results
Status("Post-calibration search...", new List <string> {
taskId, "Individual Spectra Files"
});
acquisitionResults = GetDataAcquisitionResults(myMsDataFile, originalUncalibratedFilePath, variableModifications, fixedModifications, proteinList, taskId, combinedParams, combinedParams.PrecursorMassTolerance, combinedParams.ProductMassTolerance);
// stats after calibration
int postCalibrationPsmCount = acquisitionResults.Psms.Count;
double postCalibrationPrecursorErrorIqr = acquisitionResults.PsmPrecursorIqrPpmError;
double postCalibrationProductErrorIqr = acquisitionResults.PsmProductIqrPpmError;
// did the data improve? (not used for anything yet...)
bool improvement = ImprovGlobal(preCalibrationPrecursorErrorIqr, preCalibrationProductErrorIqr, prevPsmCount, postCalibrationPsmCount, postCalibrationPrecursorErrorIqr, postCalibrationProductErrorIqr);
// write toml settings for the calibrated file
var newTomlFileName = Path.Combine(OutputFolder, originalUncalibratedFilenameWithoutExtension + CalibSuffix + ".toml");
var fileSpecificParams = new FileSpecificParameters();
// carry over file-specific parameters from the uncalibrated file to the calibrated one
if (fileSettingsList[spectraFileIndex] != null)
{
fileSpecificParams = fileSettingsList[spectraFileIndex].Clone();
}
// don't write over ppm tolerances if they've been specified by the user already in the file-specific settings
// otherwise, suggest 4 * interquartile range as the ppm tolerance
if (fileSpecificParams.PrecursorMassTolerance == null)
{
fileSpecificParams.PrecursorMassTolerance = new PpmTolerance((4.0 * postCalibrationPrecursorErrorIqr) + Math.Abs(acquisitionResults.PsmPrecursorMedianPpmError));
}
if (fileSpecificParams.ProductMassTolerance == null)
{
fileSpecificParams.ProductMassTolerance = new PpmTolerance((4.0 * postCalibrationProductErrorIqr) + Math.Abs(acquisitionResults.PsmProductMedianPpmError));
}
Toml.WriteFile(fileSpecificParams, newTomlFileName, tomlConfig);
FinishedWritingFile(newTomlFileName, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
});
// write the calibrated mzML file
MzmlMethods.CreateAndWriteMyMzmlWithCalibratedSpectra(myMsDataFile, calibratedFilePath, false);
myFileManager.DoneWithFile(originalUncalibratedFilePath);
// finished calibrating this file
FinishedWritingFile(calibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
});
MyTaskResults.NewSpectra.Add(calibratedFilePath);
MyTaskResults.NewFileSpecificTomls.Add(newTomlFileName);
FinishedDataFile(originalUncalibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilePath
});
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
}));
}
// re-write experimental design (if it has been defined) with new calibrated file names
string assumedPathToExperDesign = Directory.GetParent(currentRawFileList.First()).FullName;
assumedPathToExperDesign = Path.Combine(assumedPathToExperDesign, GlobalVariables.ExperimentalDesignFileName);
List <string> newExperimentalDesignOutput = new List <string>();
if (File.Exists(assumedPathToExperDesign))
{
var lines = File.ReadAllLines(assumedPathToExperDesign);
for (int i = 0; i < lines.Length; i++)
{
// header of experimental design file
if (i == 0)
{
newExperimentalDesignOutput.Add(lines[i]);
}
else
{
var split = lines[i].Split('\t');
string newline = Path.GetFileNameWithoutExtension(split[0]) + CalibSuffix + "\t";
for (int j = 1; j < split.Length; j++)
{
newline += split[j] + "\t";
}
newExperimentalDesignOutput.Add(newline);
}
}
}
File.WriteAllLines(Path.Combine(OutputFolder, GlobalVariables.ExperimentalDesignFileName), newExperimentalDesignOutput);
// finished calibrating all files for the task
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files"
}));
return(MyTaskResults);
}
public static void WritePepXml(List <PeptideSpectralMatch> psms, List <DbForTask> database, List <Modification> variableModifications, List <Modification> fixedModifications, CommonParameters CommonParameters, string outputPath, double qValueFilter)
{
// TODO: needs a unit test
psms = psms.Where(p => p.FdrInfo.QValue <= qValueFilter && p.FdrInfo.QValueNotch < qValueFilter).ToList();
if (!psms.Any())
{
return;
}
XmlSerializer _indexedSerializer = new XmlSerializer(typeof(pepXML.Generated.msms_pipeline_analysis));
var _pepxml = new pepXML.Generated.msms_pipeline_analysis();
_pepxml.date = DateTime.Now;
_pepxml.summary_xml = psms[0].FullFilePath + ".pep.XML";
string proteaseNC = string.Join(string.Empty, CommonParameters.DigestionParams.Protease.DigestionMotifs.Select(m => m.InducingCleavage));
string proteaseC = string.Join(string.Empty, CommonParameters.DigestionParams.Protease.DigestionMotifs.Select(m => m.InducingCleavage));
string fileNameNoExtension = Path.GetFileNameWithoutExtension(psms[0].FullFilePath);
string filePathNoExtension = Path.ChangeExtension(psms[0].FullFilePath, null);
var para = new List <pepXML.Generated.nameValueType>();
{
para.Add(new pepXML.Generated.nameValueType {
name = "threads", value = CommonParameters.MaxThreadsToUsePerFile.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "database", value = database.First().FilePath
});
para.Add(new pepXML.Generated.nameValueType {
name = "MS_data_file", value = psms[0].FullFilePath
});
para.Add(new pepXML.Generated.nameValueType {
name = "MaxMissed Cleavages", value = CommonParameters.DigestionParams.MaxMissedCleavages.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Protease", value = CommonParameters.DigestionParams.Protease.Name
});
para.Add(new pepXML.Generated.nameValueType {
name = "Initiator Methionine", value = CommonParameters.DigestionParams.InitiatorMethionineBehavior.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Max Modification Isoforms", value = CommonParameters.DigestionParams.MaxModificationIsoforms.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Min Peptide Len", value = CommonParameters.DigestionParams.MinPeptideLength.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Max Peptide Len", value = CommonParameters.DigestionParams.MaxPeptideLength.ToString()
});
para.Add(new pepXML.Generated.nameValueType {
name = "Product Mass Tolerance", value = CommonParameters.ProductMassTolerance.ToString()
});
// TODO: check this
para.Add(new pepXML.Generated.nameValueType {
name = "Ions to search", value = string.Join(", ", DissociationTypeCollection.ProductsFromDissociationType[CommonParameters.DissociationType])
});
para.Add(new pepXML.Generated.nameValueType {
name = "Q-value Filter", value = CommonParameters.QValueOutputFilter.ToString()
});
foreach (var item in fixedModifications)
{
para.Add(new pepXML.Generated.nameValueType {
name = "Fixed Modifications: " + item.IdWithMotif, value = item.MonoisotopicMass.ToString()
});
}
foreach (var item in variableModifications)
{
para.Add(new pepXML.Generated.nameValueType {
name = "Variable Modifications: " + item.IdWithMotif, value = item.MonoisotopicMass.ToString()
});
}
para.Add(new pepXML.Generated.nameValueType {
name = "Localize All Modifications", value = "true"
});
}
_pepxml.msms_run_summary = new pepXML.Generated.msms_pipeline_analysisMsms_run_summary[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summary
{
base_name = filePathNoExtension,
raw_data_type = "raw",
raw_data = ".mzML", //TODO: use file format of spectra file used
sample_enzyme = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySample_enzyme()
{
name = CommonParameters.DigestionParams.Protease.Name,
specificity = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySample_enzymeSpecificity[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySample_enzymeSpecificity
{
cut = proteaseC,
no_cut = proteaseNC,
}
}
},
search_summary = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summary[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summary
{
base_name = filePathNoExtension,
// TODO: get MetaMorpheus recognized as a search engine type
//search_engine = pepXML.Generated.engineType.MetaMorpheus
search_engine_version = GlobalVariables.MetaMorpheusVersion,
precursor_mass_type = pepXML.Generated.massType.monoisotopic,
fragment_mass_type = pepXML.Generated.massType.monoisotopic,
search_id = 1,
//generate database information
//TODO: multiple databases
search_database = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summarySearch_database
{
local_path = database.First().FilePath,
type = pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summarySearch_databaseType.AA,
},
enzymatic_search_constraint = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySearch_summaryEnzymatic_search_constraint
{
enzyme = CommonParameters.DigestionParams.Protease.Name,
max_num_internal_cleavages = CommonParameters.DigestionParams.MaxMissedCleavages.ToString(),
},
parameter = para.ToArray()
}
},
}
};
_pepxml.msms_run_summary[0].spectrum_query = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_query[psms.Count];
var searchHits = new List <pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit>();
foreach (var psm in psms)
{
PeptideWithSetModifications peptide = psm.BestMatchingPeptides.First().Peptide;
var mods = new List <pepXML.Generated.modInfoDataTypeMod_aminoacid_mass>();
foreach (var mod in peptide.AllModsOneIsNterminus)
{
var pepXmlMod = new pepXML.Generated.modInfoDataTypeMod_aminoacid_mass
{
mass = (double)mod.Value.MonoisotopicMass,
position = (mod.Key - 1).ToString()
};
mods.Add(pepXmlMod);
}
var proteinAccessions = psm.BestMatchingPeptides.Select(p => p.Peptide.Protein.Accession).Distinct();
var searchHit = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit
{
// TODO: handle PSM ambiguity if pepXML supports it (base sequence, mod localization, protein)
// TODO: add target/decoy/contaminant designation for each PSM
// TODO: add amino acid substitution
hit_rank = 1,
peptide = ((psm.BaseSequence != null) ? psm.BaseSequence : "Ambiguous"),
peptide_prev_aa = peptide.PreviousAminoAcid.ToString(),
peptide_next_aa = peptide.NextAminoAcid.ToString(),
protein = ((peptide.Protein.Accession != null) ? peptide.Protein.Accession : string.Join("|", proteinAccessions)),
num_tot_proteins = (uint)proteinAccessions.Count(),
calc_neutral_pep_mass = (float)((psm.PeptideMonisotopicMass != null) ? psm.PeptideMonisotopicMass : float.NaN),
massdiff = ((psm.PeptideMonisotopicMass != null) ? (psm.ScanPrecursorMass - psm.PeptideMonisotopicMass.Value).ToString() : "Ambiguous"),
modification_info = (mods.Count == 0 ? new pepXML.Generated.modInfoDataType {
mod_aminoacid_mass = mods.ToArray()
} : null),
search_score = new pepXML.Generated.nameValueType[]
{
new pepXML.Generated.nameValueType {
name = "Score", value = psm.Score.ToString()
},
new pepXML.Generated.nameValueType {
name = "Qvalue", value = psm.FdrInfo.QValue.ToString()
}
},
};
searchHits.Add(searchHit);
}
for (int i = 0; i < psms.Count; i++)
{
_pepxml.msms_run_summary[0].spectrum_query[i] = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_query()
{
spectrum = fileNameNoExtension + "." + psms[i].ScanNumber.ToString(),
start_scan = Convert.ToUInt32(psms[i].ScanNumber),
end_scan = Convert.ToUInt32(psms[i].ScanNumber),
precursor_neutral_mass = (float)psms[i].ScanPrecursorMass,
assumed_charge = psms[i].ScanPrecursorCharge.ToString(),
index = Convert.ToUInt32(i + 1),
retention_time_sec = (float)(psms[i].ScanRetentionTime * 60),
search_result = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_result[1]
{
new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_result
{
search_hit = new pepXML.Generated.msms_pipeline_analysisMsms_run_summarySpectrum_querySearch_resultSearch_hit[1]
{
searchHits[i]
}
}
}
};
}
TextWriter writer = new StreamWriter(Path.Combine(outputPath));
_indexedSerializer.Serialize(writer, _pepxml);
writer.Close();
}
private DataPointAquisitionResults GetDataAcquisitionResults(MsDataFile myMsDataFile, string currentDataFile, List <ModificationWithMass> variableModifications, List <ModificationWithMass> fixedModifications, List <Protein> proteinList, string taskId, CommonParameters combinedParameters, Tolerance initPrecTol, Tolerance initProdTol)
{
var fileNameWithoutExtension = Path.GetFileNameWithoutExtension(currentDataFile);
MassDiffAcceptor searchMode = initPrecTol is PpmTolerance ?
(MassDiffAcceptor) new SinglePpmAroundZeroSearchMode(initPrecTol.Value) :
new SingleAbsoluteAroundZeroSearchMode(initPrecTol.Value);
FragmentTypes fragmentTypesForCalibration = FragmentTypes.None;
if (combinedParameters.BIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.b;
}
if (combinedParameters.YIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.y;
}
if (combinedParameters.CIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.c;
}
if (combinedParameters.ZdotIons)
{
fragmentTypesForCalibration = fragmentTypesForCalibration | FragmentTypes.zdot;
}
var listOfSortedms2Scans = GetMs2Scans(myMsDataFile, currentDataFile, combinedParameters.DoPrecursorDeconvolution, combinedParameters.UseProvidedPrecursorInfo, combinedParameters.DeconvolutionIntensityRatio, combinedParameters.DeconvolutionMaxAssumedChargeState, combinedParameters.DeconvolutionMassTolerance).OrderBy(b => b.PrecursorMass).ToArray();
PeptideSpectralMatch[] allPsmsArray = new PeptideSpectralMatch[listOfSortedms2Scans.Length];
List <ProductType> productTypes = new List <ProductType>();
if (combinedParameters.BIons)
{
productTypes.Add(ProductType.B);
}
if (combinedParameters.YIons)
{
productTypes.Add(ProductType.Y);
}
if (combinedParameters.CIons)
{
productTypes.Add(ProductType.C);
}
if (combinedParameters.ZdotIons)
{
productTypes.Add(ProductType.Zdot);
}
Log("Searching with searchMode: " + searchMode, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
});
Log("Searching with productMassTolerance: " + initProdTol, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
});
new ClassicSearchEngine(allPsmsArray, listOfSortedms2Scans, variableModifications, fixedModifications, proteinList, productTypes, searchMode, combinedParameters, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
List <PeptideSpectralMatch> allPsms = allPsmsArray.Where(p => p != null).ToList();
var compactPeptideToProteinPeptideMatching = ((SequencesToActualProteinPeptidesEngineResults) new SequencesToActualProteinPeptidesEngine
(allPsms, proteinList, fixedModifications, variableModifications, productTypes, new List <DigestionParams> {
combinedParameters.DigestionParams
},
combinedParameters.ReportAllAmbiguity, combinedParameters, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run()).CompactPeptideToProteinPeptideMatching;
foreach (var huh in allPsms)
{
huh.MatchToProteinLinkedPeptides(compactPeptideToProteinPeptideMatching);
}
allPsms = allPsms.OrderByDescending(b => b.Score)
.ThenBy(b => b.PeptideMonisotopicMass.HasValue ? Math.Abs(b.ScanPrecursorMass - b.PeptideMonisotopicMass.Value) : double.MaxValue)
.GroupBy(b => (b.FullFilePath, b.ScanNumber, b.PeptideMonisotopicMass)).Select(b => b.First()).ToList();
new FdrAnalysisEngine(allPsms, searchMode.NumNotches, CommonParameters, new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
List <PeptideSpectralMatch> goodIdentifications = allPsms.Where(b => b.FdrInfo.QValueNotch < 0.01 && !b.IsDecoy && b.FullSequence != null).ToList();
if (!goodIdentifications.Any())
{
return(new DataPointAquisitionResults(null, new List <PeptideSpectralMatch>(), new List <LabeledDataPoint>(), new List <LabeledDataPoint>(), 0, 0, 0, 0));
}
DataPointAquisitionResults currentResult = (DataPointAquisitionResults) new DataPointAcquisitionEngine(
goodIdentifications,
myMsDataFile,
initPrecTol,
initProdTol,
CalibrationParameters.NumFragmentsNeededForEveryIdentification,
CalibrationParameters.MinMS1IsotopicPeaksNeededForConfirmedIdentification,
CalibrationParameters.MinMS2IsotopicPeaksNeededForConfirmedIdentification,
fragmentTypesForCalibration,
CommonParameters,
new List <string> {
taskId, "Individual Spectra Files", fileNameWithoutExtension
}).Run();
return(currentResult);
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
MyTaskResults = new MyTaskResults(this);
if (NeoType.Equals(NeoTaskType.AggregateTargetDecoyFiles))
{
//getfolders
if (NeoParameters.DecoyFilePath == null)
{
NeoParameters.DecoyFilePath = new DirectoryInfo(OutputFolder).Name;
string taskString = NeoParameters.DecoyFilePath.Split('-')[0];
int taskNum = Convert.ToInt32(taskString.Substring(4, taskString.Length - 4));
taskNum--;
NeoParameters.DecoyFilePath = OutputFolder.Substring(0, OutputFolder.Length - NeoParameters.DecoyFilePath.Length) + "Task" + taskNum + "-SearchTask\\" + Path.GetFileNameWithoutExtension(currentRawFileList[0]) + "_PSMs.psmtsv";
if (NeoParameters.TargetFilePath == null)
{
NeoParameters.TargetFilePath = new DirectoryInfo(OutputFolder).Name;
taskNum--;
NeoParameters.TargetFilePath = OutputFolder.Substring(0, OutputFolder.Length - NeoParameters.TargetFilePath.Length) + "Task" + taskNum + "-SearchTask\\" + Path.GetFileNameWithoutExtension(currentRawFileList[0]) + "_PSMs.psmtsv";
}
}
if (NeoParameters.TargetFilePath == null)
{
NeoParameters.TargetFilePath = new DirectoryInfo(OutputFolder).Name;
string taskString = NeoParameters.TargetFilePath.Split('-')[0];
int taskNum = Convert.ToInt32(taskString.Substring(4, taskString.Length - 4));
taskNum--;
NeoParameters.TargetFilePath = OutputFolder.Substring(0, OutputFolder.Length - NeoParameters.TargetFilePath.Length) + "Task" + taskNum + "-SearchTask\\" + Path.GetFileNameWithoutExtension(currentRawFileList[0]) + "_PSMs.psmtsv";
}
AggregateSearchFiles.Combine(NeoParameters.TargetFilePath, NeoParameters.DecoyFilePath, OutputFolder + "\\" + Path.GetFileNameWithoutExtension(currentRawFileList[0]));
}
else if (NeoType.Equals(NeoTaskType.AggregateNormalSplicedFiles))
{
//reset database
dbFilenameList = StoredDatabases;
string normalPath = "";
string cisPath = new DirectoryInfo(OutputFolder).Name;
string taskString = cisPath.Split('-')[0];
int taskNum = Convert.ToInt32(taskString.Substring(4, taskString.Length - 4));
taskNum -= 2;
string transPath = OutputFolder.Substring(0, OutputFolder.Length - cisPath.Length) + "Task" + (taskNum + 1) + "-SearchTask\\" + Path.GetFileNameWithoutExtension(currentRawFileList[0]) + "_PSMs.psmtsv";
cisPath = OutputFolder.Substring(0, OutputFolder.Length - cisPath.Length) + "Task" + taskNum + "-SearchTask\\" + Path.GetFileNameWithoutExtension(currentRawFileList[0]) + "_PSMs.psmtsv";
AggregateSearchFiles.RecursiveNeoAggregation(normalPath, cisPath, OutputFolder, "CisResults.psmtsv");
AggregateSearchFiles.RecursiveNeoAggregation(normalPath, transPath, OutputFolder, "TransResults.psmtsv");
}
else if (NeoType.Equals(NeoTaskType.GenerateSplicedPeptides))
{
NeoMassCalculator.ImportMasses();
MyFileManager myFileManager = new MyFileManager(true);
//Import Spectra
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
var origDataFile = currentRawFileList[spectraFileIndex];
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
var thisId = new List <string> {
taskId, "Individual Spectra Files", origDataFile
};
NewCollection(Path.GetFileName(origDataFile), thisId);
Status("Loading spectra file...", thisId);
MsDataFile myMsDataFile = myFileManager.LoadFile(origDataFile, combinedParams.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();
//Import Database
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, DecoyType.None, localizeableModificationTypes, combinedParams);
//Read N and C files
string nPath = NeoParameters.NFilePath;
string cPath = NeoParameters.CFilePath;
//if termini input
if (nPath == null || cPath == null)
{
//if no termini input
string taskHeader = "Task";
string[] pathArray = OutputFolder.Split('\\');
string basePath = "";
for (int i = 0; i < pathArray.Length - 1; i++)
{
basePath += pathArray[i] + '\\';
}
string currentTaskNumber = pathArray[pathArray.Length - 1].Split('-')[0];
currentTaskNumber = currentTaskNumber.Substring(taskHeader.Length, currentTaskNumber.Length - taskHeader.Length);
string NHeader = "";
string CHeader = "";
if (cPath == null)
{
CHeader = taskHeader + (Convert.ToInt16(currentTaskNumber) - 1);
if (nPath == null)
{
NHeader = taskHeader + (Convert.ToInt16(currentTaskNumber) - 2);
}
}
else
{
NHeader = taskHeader + (Convert.ToInt16(currentTaskNumber) - 1);
}
foreach (string s in Directory.GetDirectories(basePath))
{
if (s.Contains(NHeader))
{
nPath = s;
}
else if (s.Contains(CHeader))
{
cPath = s;
}
}
string fileName = Path.GetFileNameWithoutExtension(currentRawFileList[0]) + "_PSMs.psmtsv";
nPath += "\\" + fileName;
cPath += "\\" + fileName;
}
Status("Importing Search Results...", taskId);
List <NeoPsm> psms = ImportPsmtsv.ImportNeoPsms(nPath, cPath);
//Splice
Status("Splicing Fragments...", taskId);
List <NeoPsm> candidates = NeoSplicePeptides.SplicePeptides(psms);
//Find Ambiguity
Status("Identifying Ambiguity...", taskId);
NeoFindAmbiguity.FindAmbiguity(candidates, proteinList, arrayOfMs2ScansSortedByMass, dbFilenameList[0].FilePath);
//Export Results
Status("Exporting Results...", taskId);
NeoExport.ExportAll(candidates, arrayOfMs2ScansSortedByMass, OutputFolder);
//Switch databases
string outputFolder = NeoExport.path + NeoExport.folder + @"\" + NeoExport.folder + "FusionDatabaseAppendixNC.fasta";
dbFilenameList = new List <DbForTask>()
{
new DbForTask(outputFolder, false)
};
}
}
else //if SearchTransDb
{
string outputFolder = NeoExport.path + NeoExport.folder + @"\" + NeoExport.folder + "FusionDatabaseAppendixTS.fasta";
dbFilenameList = new List <DbForTask>()
{
new DbForTask(outputFolder, false)
};
}
return(MyTaskResults);
}
public MyTaskResults Run(string outputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, List <CrosslinkSpectralMatch> allPsms, CommonParameters commonParameters, XlSearchParameters xlSearchParameters, List <Protein> proteinList, List <Modification> variableModifications, List <Modification> fixedModifications, List <string> localizeableModificationTypes, MyTaskResults MyTaskResults)
{
// inter-crosslinks; different proteins are linked
List <CrosslinkSpectralMatch> interCsms = allPsms.Where(p => p.CrossType == PsmCrossType.Inter).OrderByDescending(p => p.XLTotalScore).ToList();
// intra-crosslinks; crosslinks within a protein
List <CrosslinkSpectralMatch> intraCsms = allPsms.Where(p => p.CrossType == PsmCrossType.Intra).OrderByDescending(p => p.XLTotalScore).ToList();
var singlePsms = allPsms.Where(p => p.CrossType == PsmCrossType.Single).OrderByDescending(p => p.Score).ToList();
var loopPsms = allPsms.Where(p => p.CrossType == PsmCrossType.Loop).OrderByDescending(p => p.Score).ToList();
var deadendPsms = allPsms.Where(p => p.CrossType == PsmCrossType.DeadEnd ||
p.CrossType == PsmCrossType.DeadEndH2O ||
p.CrossType == PsmCrossType.DeadEndNH2 ||
p.CrossType == PsmCrossType.DeadEndTris).OrderByDescending(p => p.Score).ToList();
ComputeXlinkQandPValues(allPsms, intraCsms, interCsms, commonParameters, taskId);
WriteCsvFiles(outputFolder, interCsms, intraCsms, singlePsms, loopPsms, deadendPsms, taskId, xlSearchParameters);
MyTaskResults.AddTaskSummaryText("Target inter-crosslinks within 1% FDR: " + interCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
MyTaskResults.AddTaskSummaryText("Target intra-crosslinks within 1% FDR: " + intraCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
MyTaskResults.AddTaskSummaryText("Target single peptides within 1% FDR: " + singlePsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
MyTaskResults.AddTaskSummaryText("Target loop-linked peptides within 1% FDR: " + loopPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
MyTaskResults.AddTaskSummaryText("Target deadend peptides within 1% FDR: " + deadendPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write pepXML
if (xlSearchParameters.WritePepXml)
{
List <CrosslinkSpectralMatch> writeToXml = new List <CrosslinkSpectralMatch>();
writeToXml.AddRange(intraCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(interCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(singlePsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(loopPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(deadendPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml = writeToXml.OrderBy(p => p.ScanNumber).ToList();
foreach (var fullFilePath in currentRawFileList)
{
string fileNameNoExtension = Path.GetFileNameWithoutExtension(fullFilePath);
WriteFile.WritePepXML_xl(writeToXml.Where(p => p.FullFilePath == fullFilePath).ToList(), proteinList, dbFilenameList[0].FilePath, variableModifications, fixedModifications, localizeableModificationTypes, outputFolder, fileNameNoExtension, commonParameters, xlSearchParameters);
FinishedWritingFile(Path.Combine(outputFolder, fileNameNoExtension + ".pep.XML"), new List <string> {
taskId
});
}
}
return(MyTaskResults);
}
public SearchTask() : base(MyTask.Search)
{
CommonParameters = new CommonParameters();
SearchParameters = new SearchParameters();
}
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);
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, DecoyType.Reverse, localizeableModificationTypes, CommonParameters);
// write prose settings
ProseCreatedWhileRunning.Append("The following calibration 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("The combined search database contained " + proteinList.Count(p => !p.IsDecoy) + " non-decoy protein entries including " + proteinList.Count(p => p.IsContaminant) + " contaminant sequences. ");
// start the calibration task
Status("Calibrating...", new List <string> {
taskId
});
MyTaskResults = new MyTaskResults(this)
{
NewSpectra = new List <string>(),
NewFileSpecificTomls = new List <string>()
};
var myFileManager = new MyFileManager(true);
List <string> spectraFilesAfterCalibration = new List <string>();
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
if (GlobalVariables.StopLoops)
{
break;
}
bool couldNotFindEnoughDatapoints = false;
// get filename stuff
var originalUncalibratedFilePath = currentRawFileList[spectraFileIndex];
var originalUncalibratedFilenameWithoutExtension = Path.GetFileNameWithoutExtension(originalUncalibratedFilePath);
string calibratedFilePath = Path.Combine(OutputFolder, originalUncalibratedFilenameWithoutExtension + CalibSuffix + ".mzML");
// mark the file as in-progress
StartingDataFile(originalUncalibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilePath
});
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
// load the file
Status("Loading spectra file...", new List <string> {
taskId, "Individual Spectra Files"
});
var myMsDataFile = myFileManager.LoadFile(originalUncalibratedFilePath, CommonParameters);
// get datapoints to fit calibration function to
Status("Acquiring calibration data points...", new List <string> {
taskId, "Individual Spectra Files"
});
DataPointAquisitionResults acquisitionResults = null;
for (int i = 1; i <= 5; i++)
{
acquisitionResults = GetDataAcquisitionResults(myMsDataFile, originalUncalibratedFilePath, variableModifications, fixedModifications, proteinList, taskId, combinedParams, combinedParams.PrecursorMassTolerance, combinedParams.ProductMassTolerance);
// enough data points to calibrate?
if (acquisitionResults.Psms.Count >= NumRequiredPsms && acquisitionResults.Ms1List.Count > NumRequiredMs1Datapoints && acquisitionResults.Ms2List.Count > NumRequiredMs2Datapoints)
{
break;
}
if (i == 1) // failed round 1
{
CommonParameters.PrecursorMassTolerance = new PpmTolerance(20);
CommonParameters.ProductMassTolerance = new PpmTolerance(50);
}
else if (i == 2) // failed round 2
{
CommonParameters.PrecursorMassTolerance = new PpmTolerance(30);
CommonParameters.ProductMassTolerance = new PpmTolerance(100);
}
else if (i == 3) // failed round 3
{
CommonParameters.PrecursorMassTolerance = new PpmTolerance(40);
CommonParameters.ProductMassTolerance = new PpmTolerance(150);
}
else // failed round 4
{
if (acquisitionResults.Psms.Count < NumRequiredPsms)
{
Warn("Calibration failure! Could not find enough high-quality PSMs. Required " + NumRequiredPsms + ", saw " + acquisitionResults.Psms.Count);
}
if (acquisitionResults.Ms1List.Count < NumRequiredMs1Datapoints)
{
Warn("Calibration failure! Could not find enough MS1 datapoints. Required " + NumRequiredMs1Datapoints + ", saw " + acquisitionResults.Ms1List.Count);
}
if (acquisitionResults.Ms2List.Count < NumRequiredMs2Datapoints)
{
Warn("Calibration failure! Could not find enough MS2 datapoints. Required " + NumRequiredMs2Datapoints + ", saw " + acquisitionResults.Ms2List.Count);
}
couldNotFindEnoughDatapoints = true;
FinishedDataFile(originalUncalibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilePath
});
break;
}
Warn("Could not find enough PSMs to calibrate with; opening up tolerances to " +
Math.Round(CommonParameters.PrecursorMassTolerance.Value, 2) + " ppm precursor and " +
Math.Round(CommonParameters.ProductMassTolerance.Value, 2) + " ppm product");
}
if (couldNotFindEnoughDatapoints)
{
spectraFilesAfterCalibration.Add(Path.GetFileNameWithoutExtension(currentRawFileList[spectraFileIndex]));
ReportProgress(new ProgressEventArgs(100, "Failed to calibrate!", new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
}));
continue;
}
// stats before calibration
int prevPsmCount = acquisitionResults.Psms.Count;
double preCalibrationPrecursorErrorIqr = acquisitionResults.PsmPrecursorIqrPpmError;
double preCalibrationProductErrorIqr = acquisitionResults.PsmProductIqrPpmError;
// generate calibration function and shift data points
Status("Calibrating...", new List <string> {
taskId, "Individual Spectra Files"
});
CalibrationEngine engine = new CalibrationEngine(myMsDataFile, acquisitionResults, CommonParameters, FileSpecificParameters, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
});
engine.Run();
//update file
myMsDataFile = engine.CalibratedDataFile;
// do another search to evaluate calibration results
Status("Post-calibration search...", new List <string> {
taskId, "Individual Spectra Files"
});
acquisitionResults = GetDataAcquisitionResults(myMsDataFile, originalUncalibratedFilePath, variableModifications, fixedModifications, proteinList, taskId, combinedParams, combinedParams.PrecursorMassTolerance, combinedParams.ProductMassTolerance);
//generate calibration function and shift data points AGAIN because it's fast and contributes new data
Status("Calibrating...", new List <string> {
taskId, "Individual Spectra Files"
});
engine = new CalibrationEngine(myMsDataFile, acquisitionResults, CommonParameters, FileSpecificParameters, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
});
engine.Run();
//update file
myMsDataFile = engine.CalibratedDataFile;
// write the calibrated mzML file
MzmlMethods.CreateAndWriteMyMzmlWithCalibratedSpectra(myMsDataFile, calibratedFilePath, false);
myFileManager.DoneWithFile(originalUncalibratedFilePath);
// stats after calibration
int postCalibrationPsmCount = acquisitionResults.Psms.Count;
double postCalibrationPrecursorErrorIqr = acquisitionResults.PsmPrecursorIqrPpmError;
double postCalibrationProductErrorIqr = acquisitionResults.PsmProductIqrPpmError;
// did the data improve? (not used for anything yet...)
bool improvement = ImprovGlobal(preCalibrationPrecursorErrorIqr, preCalibrationProductErrorIqr, prevPsmCount, postCalibrationPsmCount, postCalibrationPrecursorErrorIqr, postCalibrationProductErrorIqr);
// write toml settings for the calibrated file
var newTomlFileName = Path.Combine(OutputFolder, originalUncalibratedFilenameWithoutExtension + CalibSuffix + ".toml");
var fileSpecificParams = new FileSpecificParameters();
// carry over file-specific parameters from the uncalibrated file to the calibrated one
if (fileSettingsList[spectraFileIndex] != null)
{
fileSpecificParams = fileSettingsList[spectraFileIndex].Clone();
}
//suggest 4 * interquartile range as the ppm tolerance
fileSpecificParams.PrecursorMassTolerance = new PpmTolerance((4.0 * postCalibrationPrecursorErrorIqr) + Math.Abs(acquisitionResults.PsmPrecursorMedianPpmError));
fileSpecificParams.ProductMassTolerance = new PpmTolerance((4.0 * postCalibrationProductErrorIqr) + Math.Abs(acquisitionResults.PsmProductMedianPpmError));
Toml.WriteFile(fileSpecificParams, newTomlFileName, tomlConfig);
FinishedWritingFile(newTomlFileName, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
});
// finished calibrating this file
spectraFilesAfterCalibration.Add(Path.GetFileNameWithoutExtension(calibratedFilePath));
FinishedWritingFile(calibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
});
MyTaskResults.NewSpectra.Add(calibratedFilePath);
MyTaskResults.NewFileSpecificTomls.Add(newTomlFileName);
FinishedDataFile(originalUncalibratedFilePath, new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilePath
});
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files", originalUncalibratedFilenameWithoutExtension
}));
}
// re-write experimental design (if it has been defined) with new calibrated file names
string assumedPathToExperDesign = Directory.GetParent(currentRawFileList.First()).FullName;
assumedPathToExperDesign = Path.Combine(assumedPathToExperDesign, GlobalVariables.ExperimentalDesignFileName);
if (File.Exists(assumedPathToExperDesign))
{
WriteNewExperimentalDesignFile(assumedPathToExperDesign, OutputFolder, spectraFilesAfterCalibration);
}
// finished calibrating all files for the task
ReportProgress(new ProgressEventArgs(100, "Done!", new List <string> {
taskId, "Individual Spectra Files"
}));
return(MyTaskResults);
}
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);
// 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, this.FileSpecificParameters,
SearchParameters.MaxFragmentSize, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), 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, this.FileSpecificParameters,
SearchParameters.MaxFragmentSize, true, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), 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);
new ClassicSearchEngine(fileSpecificPsms, arrayOfMs2ScansSortedByMass, variableModifications, fixedModifications, SearchParameters.SilacLabels,
SearchParameters.StartTurnoverLabel, SearchParameters.EndTurnoverLabel, proteinList, massDiffAcceptor, combinedParams, this.FileSpecificParameters, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search!", thisId));
}
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"
}));
}
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());
}
private static void UpdateTomls(string tomlFileName, string fileName, CommonParameters ye5, TerminusType terminusType, bool spliceSearch)
{
string[] oldTomlLines = File.ReadAllLines(@fileName);
List <string> newTomlLines = new List <string>();
foreach (string line in oldTomlLines)
{
if (line.Contains("LocalizeAll") && terminusType.Equals(TerminusType.None))
{
newTomlLines.Add(GetCorrectValue("LocalizeAll", tomlFileName, line));
}
else if (line.Contains("ListOfModsFixed"))
{
newTomlLines.Add(GetCorrectValue("ListOfModsFixed", tomlFileName, line));
}
else if (line.Contains("ListOfModsVariable") && terminusType.Equals(TerminusType.None) && !spliceSearch)
{
newTomlLines.Add(GetCorrectValue("ListOfModsVariable", tomlFileName, line));
}
else if (line.Contains("BIons"))
{
if (terminusType.Equals(TerminusType.N) || terminusType.Equals(TerminusType.None))
{
newTomlLines.Add(GetCorrectValue("BIons", tomlFileName, line));
}
else
{
newTomlLines.Add("BIons = false");
}
}
else if (line.Contains("YIons"))
{
if (terminusType.Equals(TerminusType.C) || terminusType.Equals(TerminusType.None))
{
newTomlLines.Add(GetCorrectValue("YIons", tomlFileName, line));
}
else
{
newTomlLines.Add("YIons = false");
}
}
else if (line.Contains("ZdotIons"))
{
if (terminusType.Equals(TerminusType.C) || terminusType.Equals(TerminusType.None))
{
newTomlLines.Add(GetCorrectValue("ZdotIons", tomlFileName, line));
}
else
{
newTomlLines.Add("ZdotIons = false");
}
}
else if (line.Contains("CIons"))
{
if (terminusType.Equals(TerminusType.N) || terminusType.Equals(TerminusType.None))
{
newTomlLines.Add(GetCorrectValue("CIons", tomlFileName, line));
}
else
{
newTomlLines.Add("CIons = false");
}
}
else if (line.Contains("ProductMassTolerance"))
{
newTomlLines.Add(GetCorrectValue("ProductMassTolerance", tomlFileName, line));
}
else if (line.Contains("PrecursorMassTolerance"))
{
newTomlLines.Add(GetCorrectValue("PrecursorMassTolerance", tomlFileName, line));
}
else if (line.Contains("MaxMissedCleavages"))
{
newTomlLines.Add(GetCorrectValue("MaxMissedCleavages", tomlFileName, line));
}
else if (line.Contains("InitiatorMethionineBehavior"))
{
newTomlLines.Add(GetCorrectValue("InitiatorMethionineBehavior", tomlFileName, line));
}
else if (line.Contains("MinPeptideLength") && !!terminusType.Equals(TerminusType.None))
{
newTomlLines.Add(GetCorrectValue("MinPeptideLength", tomlFileName, line));
}
else if (line.Contains("MaxPeptideLength"))
{
newTomlLines.Add(GetCorrectValue("MaxPeptideLength", tomlFileName, line));
}
else if (line.Contains("MaxModificationIsoforms"))
{
newTomlLines.Add(GetCorrectValue("MaxModificationIsoforms", tomlFileName, line));
}
else if (line.Contains("MaxModsForPeptide"))
{
newTomlLines.Add(GetCorrectValue("MaxModsForPeptide", tomlFileName, line));
}
else if (line.Contains("SemiProteaseDigestion"))
{
newTomlLines.Add(GetCorrectValue("SemiProteaseDigestion", tomlFileName, line));
}
else if (line.Contains("TerminusTypeSemiProtease"))
{
newTomlLines.Add(GetCorrectValue("TerminusTypeSemiProtease", tomlFileName, line));
}
else if (line.Contains("Protease") && terminusType.Equals(TerminusType.None) && !spliceSearch) //this must be last, else other names including protease will be overwritten and crash.
{
newTomlLines.Add(GetCorrectValue("Protease", tomlFileName, line));
}
else
{
newTomlLines.Add(line);
}
}
using (StreamWriter file = new StreamWriter(fileName))
foreach (string line in newTomlLines)
{
file.WriteLine(line);
}
}
public XLSearchTask() : base(MyTask.XLSearch)
{
CommonParameters = new CommonParameters();
XlSearchParameters = new XlSearchParameters();
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
MyTaskResults = new MyTaskResults(this);
List <CrosslinkSpectralMatch> allPsms = new List <CrosslinkSpectralMatch>();
LoadModifications(taskId, out var variableModifications, out var fixedModifications, out var localizeableModificationTypes);
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, XlSearchParameters.DecoyType, localizeableModificationTypes, CommonParameters);
var crosslinker = XlSearchParameters.Crosslinker;
MyFileManager myFileManager = new MyFileManager(true);
var fileSpecificCommonParams = fileSettingsList.Select(b => SetAllFileSpecificCommonParams(CommonParameters, b));
HashSet <DigestionParams> ListOfDigestionParams = new HashSet <DigestionParams>(fileSpecificCommonParams.Select(p => p.DigestionParams));
int completedFiles = 0;
object indexLock = new object();
object psmLock = new object();
Status("Searching files...", taskId);
ProseCreatedWhileRunning.Append("The following crosslink discovery were used: ");
ProseCreatedWhileRunning.Append("crosslinker name = " + crosslinker.CrosslinkerName + "; ");
ProseCreatedWhileRunning.Append("crosslinker type = " + crosslinker.Cleavable + "; ");
ProseCreatedWhileRunning.Append("crosslinker mass = " + crosslinker.TotalMass + "; ");
ProseCreatedWhileRunning.Append("crosslinker modification site(s) = " + crosslinker.CrosslinkerModSites + "; ");
ProseCreatedWhileRunning.Append("protease = " + CommonParameters.DigestionParams.Protease + "; ");
ProseCreatedWhileRunning.Append("maximum missed cleavages = " + CommonParameters.DigestionParams.MaxMissedCleavages + "; ");
ProseCreatedWhileRunning.Append("minimum peptide length = " + CommonParameters.DigestionParams.MinPeptideLength + "; ");
ProseCreatedWhileRunning.Append(CommonParameters.DigestionParams.MaxPeptideLength == int.MaxValue ?
"maximum peptide length = unspecified; " :
"maximum peptide length = " + CommonParameters.DigestionParams.MaxPeptideLength + "; ");
ProseCreatedWhileRunning.Append("initiator methionine behavior = " + CommonParameters.DigestionParams.InitiatorMethionineBehavior + "; ");
ProseCreatedWhileRunning.Append("max modification isoforms = " + CommonParameters.DigestionParams.MaxModificationIsoforms + "; ");
ProseCreatedWhileRunning.Append("fixed modifications = " + string.Join(", ", fixedModifications.Select(m => m.IdWithMotif) + "; "));
ProseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.IdWithMotif)) + "; ");
ProseCreatedWhileRunning.Append("parent mass tolerance(s) = " + CommonParameters.PrecursorMassTolerance + "; ");
ProseCreatedWhileRunning.Append("product mass tolerance = " + CommonParameters.ProductMassTolerance + "; ");
ProseCreatedWhileRunning.Append("The combined search database contained " + proteinList.Count + " total entries including " + proteinList.Where(p => p.IsContaminant).Count() + " contaminant sequences. ");
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
var origDataFile = currentRawFileList[spectraFileIndex];
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
var thisId = new List <string> {
taskId, "Individual Spectra Files", origDataFile
};
NewCollection(Path.GetFileName(origDataFile), thisId);
Status("Loading spectra file...", thisId);
MsDataFile myMsDataFile = myFileManager.LoadFile(origDataFile, combinedParams);
Status("Getting ms2 scans...", thisId);
Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass = GetMs2Scans(myMsDataFile, origDataFile, combinedParams).OrderBy(b => b.PrecursorMass).ToArray();
CrosslinkSpectralMatch[] newPsms = new CrosslinkSpectralMatch[arrayOfMs2ScansSortedByMass.Length];
for (int currentPartition = 0; currentPartition < CommonParameters.TotalPartitions; currentPartition++)
{
List <PeptideWithSetModifications> peptideIndex = null;
List <Protein> proteinListSubset = proteinList.GetRange(currentPartition * proteinList.Count() / combinedParams.TotalPartitions, ((currentPartition + 1) * proteinList.Count() / combinedParams.TotalPartitions) - (currentPartition * proteinList.Count() / combinedParams.TotalPartitions));
Status("Getting fragment dictionary...", new List <string> {
taskId
});
var indexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, null, currentPartition, UsefulProteomicsDatabases.DecoyType.Reverse, combinedParams, 30000.0, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), new List <string> {
taskId
});
List <int>[] fragmentIndex = null;
List <int>[] precursorIndex = null;
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, ref precursorIndex, proteinList, taskId);
Status("Searching files...", taskId);
new CrosslinkSearchEngine(newPsms, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, currentPartition, combinedParams, crosslinker,
XlSearchParameters.RestrictToTopNHits, XlSearchParameters.CrosslinkSearchTopNum, XlSearchParameters.XlQuench_H2O,
XlSearchParameters.XlQuench_NH2, XlSearchParameters.XlQuench_Tris, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + CommonParameters.TotalPartitions + "!", thisId));
if (GlobalVariables.StopLoops)
{
break;
}
}
allPsms.AddRange(newPsms.Where(p => p != null));
completedFiles++;
ReportProgress(new ProgressEventArgs(completedFiles / currentRawFileList.Count, "Searching...", new List <string> {
taskId, "Individual Spectra Files"
}));
}
ReportProgress(new ProgressEventArgs(100, "Done with all searches!", new List <string> {
taskId, "Individual Spectra Files"
}));
allPsms = allPsms.OrderByDescending(p => p.XLTotalScore).ToList();
var allPsmsXL = allPsms.Where(p => p.CrossType == PsmCrossType.Cross).ToList();
// inter-crosslinks; different proteins are linked
var interCsms = allPsmsXL.Where(p => !p.ProteinAccession.Equals(p.BetaPeptide.ProteinAccession)).ToList();
foreach (var item in interCsms)
{
item.CrossType = PsmCrossType.Inter;
}
// intra-crosslinks; crosslinks within a protein
var intraCsms = allPsmsXL.Where(p => p.ProteinAccession.Equals(p.BetaPeptide.ProteinAccession)).ToList();
foreach (var item in intraCsms)
{
item.CrossType = PsmCrossType.Intra;
}
// calculate FDR
DoCrosslinkFdrAnalysis(interCsms);
DoCrosslinkFdrAnalysis(intraCsms);
SingleFDRAnalysis(allPsms, new List <string> {
taskId
});
// calculate protein crosslink residue numbers
foreach (var csm in allPsmsXL)
{
// alpha peptide crosslink residue in the protein
csm.XlProteinPos = csm.OneBasedStartResidueInProtein.Value + csm.LinkPositions[0] - 1;
// beta crosslink residue in protein
csm.BetaPeptide.XlProteinPos = csm.BetaPeptide.OneBasedStartResidueInProtein.Value + csm.BetaPeptide.LinkPositions[0] - 1;
}
// write interlink CSMs
if (interCsms.Any())
{
string file = Path.Combine(OutputFolder, "XL_Interlinks.tsv");
WritePsmCrossToTsv(interCsms, file, 2);
FinishedWritingFile(file, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target inter-crosslinks within 1% FDR: " + interCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
if (XlSearchParameters.WriteOutputForPercolator)
{
var interPsmsXLPercolator = interCsms.Where(p => p.Score >= 2 && p.BetaPeptide.Score >= 2).OrderBy(p => p.ScanNumber).ToList();
WriteCrosslinkToTxtForPercolator(interPsmsXLPercolator, OutputFolder, "XL_Interlinks_Percolator", crosslinker, new List <string> {
taskId
});
}
// write intralink CSMs
if (intraCsms.Any())
{
string file = Path.Combine(OutputFolder, "XL_Intralinks.tsv");
WritePsmCrossToTsv(intraCsms, file, 2);
FinishedWritingFile(file, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target intra-crosslinks within 1% FDR: " + intraCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
if (XlSearchParameters.WriteOutputForPercolator)
{
var intraPsmsXLPercolator = intraCsms.Where(p => p.Score >= 2 && p.BetaPeptide.Score >= 2).OrderBy(p => p.ScanNumber).ToList();
WriteCrosslinkToTxtForPercolator(intraPsmsXLPercolator, OutputFolder, "XL_Intralinks_Percolator", crosslinker, new List <string> {
taskId
});
}
// write single peptides
var singlePsms = allPsms.Where(p => p.CrossType == PsmCrossType.Single).ToList();
if (singlePsms.Any())
{
string writtenFileSingle = Path.Combine(OutputFolder, "SinglePeptides" + ".tsv");
WritePsmCrossToTsv(singlePsms, writtenFileSingle, 1);
FinishedWritingFile(writtenFileSingle, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target single peptides within 1% FDR: " + singlePsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write loops
var loopPsms = allPsms.Where(p => p.CrossType == PsmCrossType.Loop).ToList();
if (loopPsms.Any())
{
string writtenFileLoop = Path.Combine(OutputFolder, "Looplinks" + ".tsv");
WritePsmCrossToTsv(loopPsms, writtenFileLoop, 1);
FinishedWritingFile(writtenFileLoop, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target loop-linked peptides within 1% FDR: " + loopPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write deadends
var deadendPsms = allPsms.Where(p => p.CrossType == PsmCrossType.DeadEnd ||
p.CrossType == PsmCrossType.DeadEndH2O ||
p.CrossType == PsmCrossType.DeadEndNH2 ||
p.CrossType == PsmCrossType.DeadEndTris).ToList();
if (deadendPsms.Any())
{
string writtenFileDeadend = Path.Combine(OutputFolder, "Deadends" + ".tsv");
WritePsmCrossToTsv(deadendPsms, writtenFileDeadend, 1);
FinishedWritingFile(writtenFileDeadend, new List <string> {
taskId
});
}
MyTaskResults.AddNiceText("Target deadend peptides within 1% FDR: " + deadendPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write pepXML
if (XlSearchParameters.WritePepXml)
{
List <CrosslinkSpectralMatch> writeToXml = new List <CrosslinkSpectralMatch>();
writeToXml.AddRange(intraCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(interCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(singlePsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(loopPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(deadendPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml = writeToXml.OrderBy(p => p.ScanNumber).ToList();
foreach (var fullFilePath in currentRawFileList)
{
string fileNameNoExtension = Path.GetFileNameWithoutExtension(fullFilePath);
WritePepXML_xl(writeToXml.Where(p => p.FullFilePath == fullFilePath).ToList(), proteinList, dbFilenameList[0].FilePath, variableModifications, fixedModifications, localizeableModificationTypes, OutputFolder, fileNameNoExtension, new List <string> {
taskId
});
}
}
return(MyTaskResults);
}
public GptmdTask() : base(MyTask.Gptmd)
{
CommonParameters = new CommonParameters();
GptmdParameters = new GptmdParameters();
}
public MsDataFile LoadFile(string origDataFile, int?topNpeaks, double?minRatio, bool trimMs1Peaks, bool trimMsMsPeaks, CommonParameters commonParameters)
{
FilteringParams filter = new FilteringParams(topNpeaks, minRatio, 1, trimMs1Peaks, trimMsMsPeaks);
if (MyMsDataFiles.TryGetValue(origDataFile, out MsDataFile value) && value != null)
{
return(value);
}
// By now know that need to load this file!!!
lock (FileLoadingLock) // Lock because reading is sequential
{
if (Path.GetExtension(origDataFile).Equals(".mzML", StringComparison.OrdinalIgnoreCase))
{
MyMsDataFiles[origDataFile] = Mzml.LoadAllStaticData(origDataFile, filter, commonParameters.MaxThreadsToUsePerFile);
}
else if (Path.GetExtension(origDataFile).Equals(".mgf", StringComparison.OrdinalIgnoreCase))
{
MyMsDataFiles[origDataFile] = Mgf.LoadAllStaticData(origDataFile, filter);
}
else
{
#if NETFRAMEWORK
MyMsDataFiles[origDataFile] = ThermoStaticData.LoadAllStaticData(origDataFile, filter);
#else
Warn("No capability for reading " + origDataFile);
#endif
}
return(MyMsDataFiles[origDataFile]);
}
}
public MyTaskResults Run(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList, List <CrosslinkSpectralMatch> allPsms, CommonParameters commonParameters, XlSearchParameters xlSearchParameters, List <Protein> proteinList, List <Modification> variableModifications, List <Modification> fixedModifications, List <string> localizeableModificationTypes, MyTaskResults MyTaskResults)
{
foreach (var csm in allPsms)
{
csm.ResolveProteinPosAmbiguitiesForXl();
}
var allPsmsXL = allPsms.Where(p => p.CrossType == PsmCrossType.Cross).ToList();
// inter-crosslinks; different proteins are linked
var interCsms = allPsmsXL.Where(p => !p.IsIntraCsm()).ToList();
foreach (var item in interCsms)
{
item.CrossType = PsmCrossType.Inter;
}
// intra-crosslinks; crosslinks within a protein
var intraCsms = allPsmsXL.Where(p => p.IsIntraCsm()).ToList();
foreach (var item in intraCsms)
{
item.CrossType = PsmCrossType.Intra;
}
// calculate FDR
DoCrosslinkFdrAnalysis(interCsms);
DoCrosslinkFdrAnalysis(intraCsms);
SingleFDRAnalysis(allPsms, commonParameters, new List <string> {
taskId
});
// write interlink CSMs
if (interCsms.Any())
{
string file = Path.Combine(OutputFolder, "XL_Interlinks.tsv");
WriteFile.WritePsmCrossToTsv(interCsms, file, 2);
FinishedWritingFile(file, new List <string> {
taskId
});
}
MyTaskResults.AddTaskSummaryText("Target inter-crosslinks within 1% FDR: " + interCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
if (xlSearchParameters.WriteOutputForPercolator)
{
var interPsmsXLPercolator = interCsms.Where(p => p.Score >= 2 && p.BetaPeptide.Score >= 2).OrderBy(p => p.ScanNumber).ToList();
WriteFile.WriteCrosslinkToTxtForPercolator(interPsmsXLPercolator, OutputFolder, "XL_Interlinks_Percolator", xlSearchParameters.Crosslinker);
FinishedWritingFile(Path.Combine(OutputFolder, "XL_Interlinks_Percolator.txt"), new List <string> {
taskId
});
}
// write intralink CSMs
if (intraCsms.Any())
{
string file = Path.Combine(OutputFolder, "XL_Intralinks.tsv");
WriteFile.WritePsmCrossToTsv(intraCsms, file, 2);
FinishedWritingFile(file, new List <string> {
taskId
});
}
MyTaskResults.AddTaskSummaryText("Target intra-crosslinks within 1% FDR: " + intraCsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy && !p.BetaPeptide.IsDecoy));
if (xlSearchParameters.WriteOutputForPercolator)
{
var intraPsmsXLPercolator = intraCsms.Where(p => p.Score >= 2 && p.BetaPeptide.Score >= 2).OrderBy(p => p.ScanNumber).ToList();
WriteFile.WriteCrosslinkToTxtForPercolator(intraPsmsXLPercolator, OutputFolder, "XL_Intralinks_Percolator", xlSearchParameters.Crosslinker);
FinishedWritingFile(Path.Combine(OutputFolder, "XL_Intralinks_Percolator.txt"), new List <string> {
taskId
});
}
// write single peptides
var singlePsms = allPsms.Where(p => p.CrossType == PsmCrossType.Single).ToList();
if (singlePsms.Any())
{
string writtenFileSingle = Path.Combine(OutputFolder, "SinglePeptides" + ".tsv");
WriteFile.WritePsmCrossToTsv(singlePsms, writtenFileSingle, 1);
FinishedWritingFile(writtenFileSingle, new List <string> {
taskId
});
}
MyTaskResults.AddTaskSummaryText("Target single peptides within 1% FDR: " + singlePsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write loops
var loopPsms = allPsms.Where(p => p.CrossType == PsmCrossType.Loop).ToList();
if (loopPsms.Any())
{
string writtenFileLoop = Path.Combine(OutputFolder, "Looplinks" + ".tsv");
WriteFile.WritePsmCrossToTsv(loopPsms, writtenFileLoop, 1);
FinishedWritingFile(writtenFileLoop, new List <string> {
taskId
});
}
MyTaskResults.AddTaskSummaryText("Target loop-linked peptides within 1% FDR: " + loopPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write deadends
var deadendPsms = allPsms.Where(p => p.CrossType == PsmCrossType.DeadEnd ||
p.CrossType == PsmCrossType.DeadEndH2O ||
p.CrossType == PsmCrossType.DeadEndNH2 ||
p.CrossType == PsmCrossType.DeadEndTris).ToList();
if (deadendPsms.Any())
{
string writtenFileDeadend = Path.Combine(OutputFolder, "Deadends" + ".tsv");
WriteFile.WritePsmCrossToTsv(deadendPsms, writtenFileDeadend, 1);
FinishedWritingFile(writtenFileDeadend, new List <string> {
taskId
});
}
MyTaskResults.AddTaskSummaryText("Target deadend peptides within 1% FDR: " + deadendPsms.Count(p => p.FdrInfo.QValue <= 0.01 && !p.IsDecoy));
// write pepXML
if (xlSearchParameters.WritePepXml)
{
List <CrosslinkSpectralMatch> writeToXml = new List <CrosslinkSpectralMatch>();
writeToXml.AddRange(intraCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(interCsms.Where(p => !p.IsDecoy && !p.BetaPeptide.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(singlePsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(loopPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml.AddRange(deadendPsms.Where(p => !p.IsDecoy && p.FdrInfo.QValue <= 0.05));
writeToXml = writeToXml.OrderBy(p => p.ScanNumber).ToList();
foreach (var fullFilePath in currentRawFileList)
{
string fileNameNoExtension = Path.GetFileNameWithoutExtension(fullFilePath);
WriteFile.WritePepXML_xl(writeToXml.Where(p => p.FullFilePath == fullFilePath).ToList(), proteinList, dbFilenameList[0].FilePath, variableModifications, fixedModifications, localizeableModificationTypes, OutputFolder, fileNameNoExtension, commonParameters, xlSearchParameters);
FinishedWritingFile(Path.Combine(OutputFolder, fileNameNoExtension + ".pep.XML"), new List <string> {
taskId
});
}
}
return(MyTaskResults);
}
protected override MyTaskResults RunSpecific(string OutputFolder, List <DbForTask> dbFilenameList, List <string> currentRawFileList, string taskId, FileSpecificParameters[] fileSettingsList)
{
MyTaskResults = new MyTaskResults(this);
List <List <GlycoSpectralMatch> > ListOfGsmsPerMS2Scan = new List <List <GlycoSpectralMatch> >();
LoadModifications(taskId, out var variableModifications, out var fixedModifications, out var localizeableModificationTypes);
// load proteins
List <Protein> proteinList = LoadProteins(taskId, dbFilenameList, true, _glycoSearchParameters.DecoyType, localizeableModificationTypes, CommonParameters);
MyFileManager myFileManager = new MyFileManager(true);
int completedFiles = 0;
Status("Searching files...", taskId);
ProseCreatedWhileRunning.Append("\n");
ProseCreatedWhileRunning.Append("protease = " + CommonParameters.DigestionParams.Protease + "; \n");
ProseCreatedWhileRunning.Append("maximum missed cleavages = " + CommonParameters.DigestionParams.MaxMissedCleavages + "; \n");
ProseCreatedWhileRunning.Append("minimum peptide length = " + CommonParameters.DigestionParams.MinPeptideLength + "; \n");
ProseCreatedWhileRunning.Append(CommonParameters.DigestionParams.MaxPeptideLength == int.MaxValue ?
"maximum peptide length = unspecified; " :
"maximum peptide length = " + CommonParameters.DigestionParams.MaxPeptideLength + "; \n");
ProseCreatedWhileRunning.Append("initiator methionine behavior = " + CommonParameters.DigestionParams.InitiatorMethionineBehavior + "; \n");
ProseCreatedWhileRunning.Append("max modification isoforms = " + CommonParameters.DigestionParams.MaxModificationIsoforms + "; \n");
ProseCreatedWhileRunning.Append("fixed modifications = " + string.Join(", ", fixedModifications.Select(m => m.IdWithMotif)) + "; \n");
ProseCreatedWhileRunning.Append("variable modifications = " + string.Join(", ", variableModifications.Select(m => m.IdWithMotif)) + "; \n");
ProseCreatedWhileRunning.Append("parent mass tolerance(s) = " + CommonParameters.PrecursorMassTolerance + "; \n");
ProseCreatedWhileRunning.Append("product mass tolerance = " + CommonParameters.ProductMassTolerance + "; \n");
ProseCreatedWhileRunning.Append("The combined search database contained " + proteinList.Count + " total entries including " + proteinList.Where(p => p.IsContaminant).Count() + " contaminant sequences. \n");
if (_glycoSearchParameters.GlycoSearchType == GlycoSearchType.OGlycanSearch)
{
ProseCreatedWhileRunning.Append("The O-glycan database: " + _glycoSearchParameters.OGlycanDatabasefile + "\n");
}
else if (_glycoSearchParameters.GlycoSearchType == GlycoSearchType.NGlycanSearch)
{
ProseCreatedWhileRunning.Append("The N-glycan database: " + _glycoSearchParameters.OGlycanDatabasefile + "\n");
}
else
{
ProseCreatedWhileRunning.Append("The O-glycan database: " + _glycoSearchParameters.OGlycanDatabasefile + "\n");
ProseCreatedWhileRunning.Append("The N-glycan database: " + _glycoSearchParameters.NGlycanDatabasefile + "\n");
}
ProseCreatedWhileRunning.Append("\n");
for (int spectraFileIndex = 0; spectraFileIndex < currentRawFileList.Count; spectraFileIndex++)
{
var origDataFile = currentRawFileList[spectraFileIndex];
CommonParameters combinedParams = SetAllFileSpecificCommonParams(CommonParameters, fileSettingsList[spectraFileIndex]);
var thisId = new List <string> {
taskId, "Individual Spectra Files", origDataFile
};
NewCollection(Path.GetFileName(origDataFile), thisId);
Status("Loading spectra file...", thisId);
MsDataFile myMsDataFile = myFileManager.LoadFile(origDataFile, combinedParams);
Status("Getting ms2 scans...", thisId);
Ms2ScanWithSpecificMass[] arrayOfMs2ScansSortedByMass = GetMs2Scans(myMsDataFile, origDataFile, combinedParams).OrderBy(b => b.PrecursorMass).ToArray();
List <GlycoSpectralMatch>[] newCsmsPerMS2ScanPerFile = new List <GlycoSpectralMatch> [arrayOfMs2ScansSortedByMass.Length];
for (int currentPartition = 0; currentPartition < CommonParameters.TotalPartitions; currentPartition++)
{
List <PeptideWithSetModifications> peptideIndex = null;
//When partition, the proteinList will be split for each Thread.
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
});
//Only reverse Decoy for glyco search has been tested and are set as fixed parameter.
var indexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, null, null, null, currentPartition, _glycoSearchParameters.DecoyType, combinedParams, this.FileSpecificParameters, 30000.0, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), TargetContaminantAmbiguity.RemoveContaminant, new List <string> {
taskId
});
List <int>[] fragmentIndex = null;
List <int>[] precursorIndex = null;
GenerateIndexes(indexEngine, dbFilenameList, ref peptideIndex, ref fragmentIndex, ref precursorIndex, proteinList, taskId);
//The second Fragment index is for 'MS1-HCD_MS1-ETD_MS2s' type of data. If LowCID is used for MS1, ion-index is not allowed to use.
List <int>[] secondFragmentIndex = null;
//if (combinedParams.MS2ChildScanDissociationType != DissociationType.LowCID
//&& !CrosslinkSearchEngine.DissociationTypeGenerateSameTypeOfIons(combinedParams.DissociationType, combinedParams.MS2ChildScanDissociationType))
//{
// //Becuase two different type of dissociation methods are used, the parameters are changed with different dissociation type.
// var secondCombinedParams = CommonParameters.CloneWithNewDissociationType(combinedParams.MS2ChildScanDissociationType);
// var secondIndexEngine = new IndexingEngine(proteinListSubset, variableModifications, fixedModifications, null, null, null, currentPartition, _glycoSearchParameters.DecoyType, secondCombinedParams, this.FileSpecificParameters, 30000.0, false, dbFilenameList.Select(p => new FileInfo(p.FilePath)).ToList(), new List<string> { taskId });
// GenerateSecondIndexes(indexEngine, secondIndexEngine, dbFilenameList, ref secondFragmentIndex, proteinList, taskId);
//}
Status("Searching files...", taskId);
new GlycoSearchEngine(newCsmsPerMS2ScanPerFile, arrayOfMs2ScansSortedByMass, peptideIndex, fragmentIndex, secondFragmentIndex, currentPartition, combinedParams, this.FileSpecificParameters,
_glycoSearchParameters.OGlycanDatabasefile, _glycoSearchParameters.NGlycanDatabasefile, _glycoSearchParameters.GlycoSearchType, _glycoSearchParameters.GlycoSearchTopNum, _glycoSearchParameters.MaximumOGlycanAllowed, _glycoSearchParameters.OxoniumIonFilt, thisId).Run();
ReportProgress(new ProgressEventArgs(100, "Done with search " + (currentPartition + 1) + "/" + CommonParameters.TotalPartitions + "!", thisId));
if (GlobalVariables.StopLoops)
{
break;
}
}
ListOfGsmsPerMS2Scan.AddRange(newCsmsPerMS2ScanPerFile.Where(p => p != null).ToList());
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"
}));
//For every Ms2Scans, each have a list of candidates psms. The allPsms from GlycoSearchEngine is the list (all ms2scans) of list (each ms2scan) of psm (all candidate psm).
//Currently, only keep the first scan for consideration.
List <GlycoSpectralMatch> GsmPerScans = ListOfGsmsPerMS2Scan.Select(p => p.First()).ToList();
var filteredAllPsms = new List <GlycoSpectralMatch>();
//For each ms2scan, try to find the best candidate psm from the psms list. Do the localizaiton analysis. Add it into filteredAllPsms.
foreach (var gsmsPerScan in GsmPerScans.GroupBy(p => p.ScanNumber))
{
var glycos = RemoveSimilarSequenceDuplicates(gsmsPerScan.OrderByDescending(p => p.Score).ToList());
foreach (var glycoSpectralMatch in glycos)
{
if (glycoSpectralMatch.LocalizationGraphs != null)
{
List <Route> localizationCandidates = new List <Route>();
for (int i = 0; i < glycoSpectralMatch.LocalizationGraphs.Count; i++)
{
var allPathWithMaxScore = LocalizationGraph.GetAllHighestScorePaths(glycoSpectralMatch.LocalizationGraphs[i].array, glycoSpectralMatch.LocalizationGraphs[i].ChildModBoxes);
foreach (var path in allPathWithMaxScore)
{
var local = LocalizationGraph.GetLocalizedPath(glycoSpectralMatch.LocalizationGraphs[i], path);
local.ModBoxId = glycoSpectralMatch.LocalizationGraphs[i].ModBoxId;
localizationCandidates.Add(local);
}
}
glycoSpectralMatch.Routes = localizationCandidates;
}
if (glycoSpectralMatch.Routes != null)
{
LocalizationLevel localLevel;
glycoSpectralMatch.LocalizedGlycan = GlycoSpectralMatch.GetLocalizedGlycan(glycoSpectralMatch.Routes, out localLevel);
glycoSpectralMatch.LocalizationLevel = localLevel;
//Localization PValue.
if (localLevel == LocalizationLevel.Level1 || localLevel == LocalizationLevel.Level2)
{
List <Route> allRoutes = new List <Route>();
foreach (var graph in glycoSpectralMatch.LocalizationGraphs)
{
allRoutes.AddRange(LocalizationGraph.GetAllPaths_CalP(graph, glycoSpectralMatch.ScanInfo_p, glycoSpectralMatch.Thero_n));
}
glycoSpectralMatch.SiteSpeciLocalProb = LocalizationGraph.CalSiteSpecificLocalizationProbability(allRoutes, glycoSpectralMatch.LocalizationGraphs.First().ModPos);
}
}
filteredAllPsms.Add(glycoSpectralMatch);
}
}
PostGlycoSearchAnalysisTask postGlycoSearchAnalysisTask = new PostGlycoSearchAnalysisTask();
postGlycoSearchAnalysisTask.FileSpecificParameters = this.FileSpecificParameters;
return(postGlycoSearchAnalysisTask.Run(OutputFolder, dbFilenameList, currentRawFileList, taskId, fileSettingsList, filteredAllPsms.OrderByDescending(p => p.Score).ToList(), CommonParameters, _glycoSearchParameters, proteinList, variableModifications, fixedModifications, localizeableModificationTypes, MyTaskResults));
}