public ProteinScoringGraphFactory(IMassBinning comparer, AminoAcidSet aminoAcidSet)
{
_comparer = comparer;
_adjList = new LinkedList<ScoringGraphEdge>[_comparer.NumberOfBins];
for (var i = 0; i < _comparer.NumberOfBins; i++) _adjList[i] = new LinkedList<ScoringGraphEdge>();
var terminalModifications = FilteredProteinMassBinning.GetTerminalModifications(aminoAcidSet);
var aminoAcidArray = FilteredProteinMassBinning.GetExtendedAminoAcidArray(aminoAcidSet);
for (var i = 0; i < _comparer.NumberOfBins; i++)
{
var mi = _comparer.GetMass(i);
var fineNodeMass = mi;
for (var a = 0; a < aminoAcidArray.Length; a++)
{
var aa = aminoAcidArray[a];
var j = _comparer.GetBinNumber(fineNodeMass + aa.Mass);
if (j < 0 || j >= _comparer.NumberOfBins) continue;
_adjList[j].AddLast(new ScoringGraphEdge(i));
if (i == 0 && !(aa is ModifiedAminoAcid))
{
foreach (var terminalMod in terminalModifications)
{
var modifiedAa = new ModifiedAminoAcid(aa, terminalMod);
j = _comparer.GetBinNumber(fineNodeMass + modifiedAa.Mass);
if (j < 0 || j >= _comparer.NumberOfBins) continue;
_adjList[j].AddLast(new ScoringGraphEdge(i));
}
}
}
}
}
/// <summary>
/// Initializes a new instance of the <see cref="FlipScoringGraph" />.
/// </summary>
/// <param name="nTerminalNodes">Node scores for nTerminal ions.</param>
/// <param name="cTerminalNodes">Node scores for cTerminal ions.</param>
/// <param name="massBins">The mass bins for constructing the nodes of the graph.</param>
/// <param name="edges">Mapping between sink node -> graph edge.</param>
public FlipScoringGraph(
IMassBinning massBins,
double?[] nTerminalNodes,
double?[] cTerminalNodes,
Dictionary <int, List <FlipScoringGraphEdge> > edges)
{
this.massBins = massBins;
this.nTerminalNodes = nTerminalNodes;
this.cTerminalNodes = cTerminalNodes;
this.edges = edges;
}
public CompositeScorerFactory(
ILcMsRun run,
IMassBinning comparer,
AminoAcidSet aaSet,
int minProductCharge = 1, int maxProductCharge = 20,
double productTolerancePpm = 10,
int isotopeOffsetTolerance = 2,
double filteringWindowSize = 1.1
)
: this(run, comparer, aaSet, minProductCharge, maxProductCharge, new Tolerance(productTolerancePpm), isotopeOffsetTolerance, filteringWindowSize)
{
}
public CompositeScorerFactory(
ILcMsRun run,
IMassBinning comparer,
AminoAcidSet aaSet,
int minProductCharge = 1, int maxProductCharge = 20,
double productTolerancePpm = 10,
int isotopeOffsetTolerance = 2,
double filteringWindowSize = 1.1
)
: this(run, comparer, aaSet, minProductCharge, maxProductCharge, new Tolerance(productTolerancePpm), isotopeOffsetTolerance, filteringWindowSize)
{
}
public ProteinScoringGraphFactory(IMassBinning comparer, AminoAcidSet aminoAcidSet)
{
_comparer = comparer;
_adjList = new LinkedList <ScoringGraphEdge> [_comparer.NumberOfBins];
for (var i = 0; i < _comparer.NumberOfBins; i++)
{
_adjList[i] = new LinkedList <ScoringGraphEdge>();
}
var terminalModifications = FilteredProteinMassBinning.GetTerminalModifications(aminoAcidSet);
var aminoAcidArray = FilteredProteinMassBinning.GetExtendedAminoAcidArray(aminoAcidSet);
for (var i = 0; i < _comparer.NumberOfBins; i++)
{
var mi = _comparer.GetMass(i);
var fineNodeMass = mi;
for (var a = 0; a < aminoAcidArray.Length; a++)
{
var aa = aminoAcidArray[a];
var j = _comparer.GetBinNumber(fineNodeMass + aa.Mass);
if (j < 0 || j >= _comparer.NumberOfBins)
{
continue;
}
_adjList[j].AddLast(new ScoringGraphEdge(i));
if (i == 0 && !(aa is ModifiedAminoAcid))
{
foreach (var terminalMod in terminalModifications)
{
var modifiedAa = new ModifiedAminoAcid(aa, terminalMod);
j = _comparer.GetBinNumber(fineNodeMass + modifiedAa.Mass);
if (j < 0 || j >= _comparer.NumberOfBins)
{
continue;
}
_adjList[j].AddLast(new ScoringGraphEdge(i));
}
}
}
}
}
public CompositeScorerBasedOnDeconvolutedSpectrum(DeconvolutedSpectrum deconvolutedSpectrum, ProductSpectrum spec, Tolerance productTolerance, IMassBinning comparer)
: base(deconvolutedSpectrum, productTolerance)
{
ReferencePeakIntensity = GetRefIntensity(spec.Peaks);
_comparer = comparer;
_massBinToPeakMap = new Dictionary<int, DeconvolutedPeak>();
foreach (var p in deconvolutedSpectrum.Peaks)
{
var mass = p.Mz;
var deltaMass = productTolerance.GetToleranceAsDa(mass, 1);
var minMass = mass - deltaMass;
var maxMass = mass + deltaMass;
var binNum = comparer.GetBinNumber(mass);
if (binNum < 0)
{
binNum = comparer.GetBinNumber(minMass);
if (binNum < 0) binNum = comparer.GetBinNumber(maxMass);
}
// filter out
if (binNum < 0) continue;
UpdateDeconvPeak(binNum, p as DeconvolutedPeak);
// going up
for (var nextBinNum = binNum + 1; nextBinNum < comparer.NumberOfBins; nextBinNum++)
{
var nextBinMass = comparer.GetMassStart(nextBinNum);
if (minMass < nextBinMass && nextBinMass < maxMass) UpdateDeconvPeak(nextBinNum, p as DeconvolutedPeak); //_ionMassChkBins[nextBinNum] = true;
else break;
}
// going down
for (var prevBinNum = binNum - 1; prevBinNum < comparer.NumberOfBins; prevBinNum--)
{
var prevBinMass = comparer.GetMassEnd(prevBinNum);
if (minMass < prevBinMass && prevBinMass < maxMass) UpdateDeconvPeak(prevBinNum, p as DeconvolutedPeak); //_ionMassChkBins[prevBinNum] = true;
else break;
}
}
}
public CompositeScorerFactory(
ILcMsRun run,
IMassBinning comparer,
AminoAcidSet aaSet,
int minProductCharge, int maxProductCharge,
Tolerance productTolerance,
int isotopeOffsetTolerance = 2,
double filteringWindowSize = 1.1)
{
_run = run;
_minProductCharge = minProductCharge;
_maxProductCharge = maxProductCharge;
_productTolerance = productTolerance;
FilteringWindowSize = filteringWindowSize;
IsotopeOffsetTolerance = isotopeOffsetTolerance;
_ms2Scorer = new Dictionary<int, IScorer>();
_comparer = comparer;
_scoringGraphFactory = new ProteinScoringGraphFactory(comparer, aaSet);
}
public CompositeScorerFactory(
ILcMsRun run,
IMassBinning comparer,
AminoAcidSet aaSet,
int minProductCharge, int maxProductCharge,
Tolerance productTolerance,
int isotopeOffsetTolerance = 2,
double filteringWindowSize = 1.1)
{
_run = run;
_minProductCharge = minProductCharge;
_maxProductCharge = maxProductCharge;
_productTolerance = productTolerance;
FilteringWindowSize = filteringWindowSize;
IsotopeOffsetTolerance = isotopeOffsetTolerance;
_ms2Scorer = new Dictionary <int, IScorer>();
_comparer = comparer;
_scoringGraphFactory = new ProteinScoringGraphFactory(comparer, aaSet);
}
/// <summary>
/// Initializes a new instance of the <see cref="ScoringGraphFactory" /> class.
/// </summary>
/// <param name="massBins">The mass bins for constructing the nodes of the graph.</param>
/// <param name="aminoAcidSet">Amino acid set to build the graph edges from.</param>
/// <param name="aminoAcidProbabilities">The amino acid probabilities.</param>
public ScoringGraphFactory(IMassBinning massBins, AminoAcidSet aminoAcidSet, Dictionary <char, double> aminoAcidProbabilities = null)
{
this.massBins = massBins;
this.aminoAcidSet = aminoAcidSet;
this.aminoAcidProbabilities = aminoAcidProbabilities;
}
public CompositeScorerBasedOnDeconvolutedSpectrum(DeconvolutedSpectrum deconvolutedSpectrum, ProductSpectrum spec, Tolerance productTolerance, IMassBinning comparer)
: base(deconvolutedSpectrum, productTolerance)
{
ReferencePeakIntensity = GetRefIntensity(spec.Peaks);
_comparer = comparer;
_massBinToPeakMap = new Dictionary <int, DeconvolutedPeak>();
foreach (var p in deconvolutedSpectrum.Peaks)
{
var mass = p.Mz;
var deltaMass = productTolerance.GetToleranceAsDa(mass, 1);
var minMass = mass - deltaMass;
var maxMass = mass + deltaMass;
var binNum = comparer.GetBinNumber(mass);
if (binNum < 0)
{
binNum = comparer.GetBinNumber(minMass);
if (binNum < 0)
{
binNum = comparer.GetBinNumber(maxMass);
}
}
// filter out
if (binNum < 0)
{
continue;
}
UpdateDeconvPeak(binNum, p as DeconvolutedPeak);
// going up
for (var nextBinNum = binNum + 1; nextBinNum < comparer.NumberOfBins; nextBinNum++)
{
var nextBinMass = comparer.GetMassStart(nextBinNum);
if (minMass < nextBinMass && nextBinMass < maxMass)
{
UpdateDeconvPeak(nextBinNum, p as DeconvolutedPeak); //_ionMassChkBins[nextBinNum] = true;
}
else
{
break;
}
}
// going down
for (var prevBinNum = binNum - 1; prevBinNum < comparer.NumberOfBins; prevBinNum--)
{
var prevBinMass = comparer.GetMassEnd(prevBinNum);
if (minMass < prevBinMass && prevBinMass < maxMass)
{
UpdateDeconvPeak(prevBinNum, p as DeconvolutedPeak); //_ionMassChkBins[prevBinNum] = true;
}
else
{
break;
}
}
}
}
public bool RunSearch(double corrThreshold = 0.7, CancellationToken? cancellationToken = null, IProgress<ProgressData> progress = null)
{
// Get the Normalized spec file/folder path
SpecFilePath = MassSpecDataReaderFactory.NormalizeDatasetPath(SpecFilePath);
var prog = new Progress<ProgressData>();
var progData = new ProgressData(progress);
if (progress != null)
{
prog = new Progress<ProgressData>(p =>
{
progData.Status = p.Status;
progData.StatusInternal = p.StatusInternal;
progData.Report(p.Percent);
});
}
var sw = new Stopwatch();
var swAll = new Stopwatch();
swAll.Start();
ErrorMessage = string.Empty;
Console.Write(@"Reading raw file...");
progData.Status = "Reading spectra file";
progData.StepRange(10.0);
sw.Start();
_run = PbfLcMsRun.GetLcMsRun(SpecFilePath, 0, 0, prog);
_ms2ScanNums = _run.GetScanNumbers(2).ToArray();
_isolationWindowTargetMz = new double[_run.MaxLcScan + 1];
foreach (var ms2Scan in _ms2ScanNums)
{
var ms2Spec = _run.GetSpectrum(ms2Scan) as ProductSpectrum;
if (ms2Spec == null) continue;
_isolationWindowTargetMz[ms2Scan] = ms2Spec.IsolationWindow.IsolationWindowTargetMz;
}
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
progData.StepRange(20.0);
ISequenceFilter ms1Filter;
if (this.ScanNumbers != null && this.ScanNumbers.Any())
{
ms1Filter = new SelectedMsMsFilter(this.ScanNumbers);
}
else if (string.IsNullOrWhiteSpace(FeatureFilePath))
{
// Checks whether SpecFileName.ms1ft exists
var ms1FtFilePath = MassSpecDataReaderFactory.ChangeExtension(SpecFilePath, LcMsFeatureFinderLauncher.FileExtension);
if (!File.Exists(ms1FtFilePath))
{
Console.WriteLine(@"Running ProMex...");
sw.Start();
var param = new LcMsFeatureFinderInputParameter
{
InputPath = SpecFilePath,
MinSearchMass = MinSequenceMass,
MaxSearchMass = MaxSequenceMass,
MinSearchCharge = MinPrecursorIonCharge,
MaxSearchCharge = MaxPrecursorIonCharge,
CsvOutput = false,
ScoreReport = false,
LikelihoodScoreThreshold = -10
};
var featureFinder = new LcMsFeatureFinderLauncher(param);
featureFinder.Run();
}
sw.Reset();
sw.Start();
Console.Write(@"Reading ProMex results...");
ms1Filter = new Ms1FtFilter(_run, PrecursorIonTolerance, ms1FtFilePath, -10);
}
else
{
sw.Reset();
sw.Start();
var extension = Path.GetExtension(FeatureFilePath);
if (extension.ToLower().Equals(".csv"))
{
Console.Write(@"Reading ICR2LS/Decon2LS results...");
ms1Filter = new IsosFilter(_run, PrecursorIonTolerance, FeatureFilePath);
}
else if (extension.ToLower().Equals(".ms1ft"))
{
Console.Write(@"Reading ProMex results...");
ms1Filter = new Ms1FtFilter(_run, PrecursorIonTolerance, FeatureFilePath, -10);
}
else if (extension.ToLower().Equals(".msalign"))
{
Console.Write(@"Reading MS-Align+ results...");
ms1Filter = new MsDeconvFilter(_run, PrecursorIonTolerance, FeatureFilePath);
}
else ms1Filter = null; //new Ms1FeatureMatrix(_run);
}
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
// pre-generate deconvoluted spectra for scoring
_massBinComparer = new FilteredProteinMassBinning(AminoAcidSet, MaxSequenceMass+1000);
_ms2ScorerFactory2 = new CompositeScorerFactory(_run, _massBinComparer, AminoAcidSet,
MinProductIonCharge, MaxProductIonCharge, ProductIonTolerance);
sw.Reset();
Console.WriteLine(@"Generating deconvoluted spectra for MS/MS spectra...");
sw.Start();
var pfeOptions = new ParallelOptions
{
MaxDegreeOfParallelism = MaxNumThreads,
CancellationToken = cancellationToken ?? CancellationToken.None
};
Parallel.ForEach(_ms2ScanNums, pfeOptions, ms2ScanNum =>
{
_ms2ScorerFactory2.DeconvonluteProductSpectrum(ms2ScanNum);
});
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
progData.StepRange(10.0);
progData.Status = "Reading Fasta File";
// Target database
var targetDb = new FastaDatabase(DatabaseFilePath);
targetDb.Read();
// Generate sequence tags for all MS/MS spectra
if (TagBasedSearch)
{
progData.StepRange(25.0);
progData.Status = "Generating Sequence Tags";
sw.Reset();
Console.WriteLine(@"Generating sequence tags for MS/MS spectra...");
sw.Start();
var seqTagGen = GetSequenceTagGenerator();
_tagMs2ScanNum = seqTagGen.GetMs2ScanNumsContainingTags().ToArray();
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
_tagSearchEngine = new ScanBasedTagSearchEngine(_run, seqTagGen, new LcMsPeakMatrix(_run, ms1Filter), targetDb, ProductIonTolerance, AminoAcidSet,
_ms2ScorerFactory2,
ScanBasedTagSearchEngine.DefaultMinMatchedTagLength,
MaxSequenceMass, MinProductIonCharge, MaxProductIonCharge);
}
var specFileName = MassSpecDataReaderFactory.RemoveExtension(Path.GetFileName(SpecFilePath));
var targetOutputFilePath = Path.Combine(OutputDir, specFileName + TargetFileNameEnding);
var decoyOutputFilePath = Path.Combine(OutputDir, specFileName + DecoyFileNameEnding);
var tdaOutputFilePath = Path.Combine(OutputDir, specFileName + TdaFileNameEnding);
progData.StepRange(60.0);
progData.Status = "Running Target search";
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Target))
{
sw.Reset();
Console.Write(@"Reading the target database...");
sw.Start();
targetDb.Read();
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
var targetMatches = new SortedSet<DatabaseSequenceSpectrumMatch>[_run.MaxLcScan + 1];
progData.MaxPercentage = 42.5;
if (TagBasedSearch)
{
sw.Reset();
Console.WriteLine(@"Tag-based searching the target database");
sw.Start();
RunTagBasedSearch(targetMatches, targetDb, null, prog);
Console.WriteLine(@"Target database tag-based search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.MaxPercentage = 60.0;
sw.Reset();
Console.WriteLine(@"Searching the target database");
sw.Start();
RunSearch(targetMatches, targetDb, ms1Filter, null, prog);
Console.WriteLine(@"Target database search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
// calculate spectral e-value usign generating function
sw.Reset();
Console.WriteLine(@"Calculating spectral E-values for target-spectrum matches");
sw.Start();
var bestTargetMatches = RunGeneratingFunction(targetMatches);
WriteResultsToFile(bestTargetMatches, targetOutputFilePath, targetDb);
sw.Stop();
Console.WriteLine(@"Target-spectrum match E-value calculation elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.StepRange(95.0); // total to 95%
progData.Status = "Running Decoy search";
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Decoy))
{
// Decoy database
sw.Reset();
sw.Start();
var decoyDb = targetDb.Decoy(null, true);
Console.Write(@"Reading the decoy database...");
decoyDb.Read();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
progData.MaxPercentage = 77.5;
var decoyMatches = new SortedSet<DatabaseSequenceSpectrumMatch>[_run.MaxLcScan + 1];
if (TagBasedSearch)
{
sw.Reset();
Console.WriteLine(@"Tag-based searching the decoy database");
sw.Start();
RunTagBasedSearch(decoyMatches, decoyDb, null, prog);
Console.WriteLine(@"Decoy database tag-based search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.MaxPercentage = 95.0;
sw.Reset();
Console.WriteLine(@"Searching the decoy database");
sw.Start();
RunSearch(decoyMatches, decoyDb, ms1Filter, null, prog);
Console.WriteLine(@"Decoy database search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
// calculate spectral e-value usign generating function
sw.Reset();
Console.WriteLine(@"Calculating spectral E-values for decoy-spectrum matches");
sw.Start();
var bestDecoyMatches = RunGeneratingFunction(decoyMatches);
WriteResultsToFile(bestDecoyMatches, decoyOutputFilePath, decoyDb);
sw.Stop();
Console.WriteLine(@"Decoy-spectrum match E-value calculation elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.StepRange(100.0);
progData.Status = "Writing combined results file";
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Both))
{
// Add "Qvalue" and "PepQValue"
var fdrCalculator = new FdrCalculator(targetOutputFilePath, decoyOutputFilePath);
if (fdrCalculator.HasError())
{
ErrorMessage = fdrCalculator.ErrorMessage;
Console.WriteLine(@"Error computing FDR: " + fdrCalculator.ErrorMessage);
return false;
}
fdrCalculator.WriteTo(tdaOutputFilePath);
}
progData.Report(100.0);
Console.WriteLine(@"Done.");
swAll.Stop();
Console.WriteLine(@"Total elapsed time for search: {0:f1} sec ({1:f2} min)", swAll.Elapsed.TotalSeconds, swAll.Elapsed.TotalMinutes);
return true;
}
internal ProteinScoringGraph(double?[] nodeScoresByPrefixIon, double?[] nodeScoresBySuffixIon, LinkedList <ScoringGraphEdge>[] adjList, IMassBinning comparer)
{
_nodeScoresByPrefixIon = nodeScoresByPrefixIon;
_nodeScoresBySuffixIon = nodeScoresBySuffixIon;
_adjList = adjList;
this._comparer = comparer;
}
/// <summary>
/// Initializes a new instance of the <see cref="FlipScoringGraphFactory" />.
/// </summary>
/// <param name="massBins">The mass bins for constructing the nodes of the graph.</param>
/// <param name="aminoAcidSet">Amino acid set to build the graph edges from.</param>
/// <param name="aminoAcidProbabilities">The amino acid probabilities.</param>
public FlipScoringGraphFactory(IMassBinning massBins, AminoAcidSet aminoAcidSet, Dictionary <char, double> aminoAcidProbabilities)
{
this.massBins = massBins;
this.edges = this.InitEdges(aminoAcidSet, aminoAcidProbabilities);
}
public bool RunSearch(double corrThreshold = 0.7, CancellationToken?cancellationToken = null, IProgress <ProgressData> progress = null)
{
// Get the Normalized spec file/folder path
SpecFilePath = MassSpecDataReaderFactory.NormalizeDatasetPath(SpecFilePath);
var prog = new Progress <ProgressData>();
var progData = new ProgressData(progress);
if (progress != null)
{
prog = new Progress <ProgressData>(p =>
{
progData.Status = p.Status;
progData.StatusInternal = p.StatusInternal;
progData.Report(p.Percent);
});
}
var sw = new Stopwatch();
var swAll = new Stopwatch();
swAll.Start();
ErrorMessage = string.Empty;
Console.Write(@"Reading raw file...");
progData.Status = "Reading spectra file";
progData.StepRange(10.0);
sw.Start();
_run = PbfLcMsRun.GetLcMsRun(SpecFilePath, 0, 0, prog);
_ms2ScanNums = _run.GetScanNumbers(2).ToArray();
_isolationWindowTargetMz = new double[_run.MaxLcScan + 1];
foreach (var ms2Scan in _ms2ScanNums)
{
var ms2Spec = _run.GetSpectrum(ms2Scan) as ProductSpectrum;
if (ms2Spec == null)
{
continue;
}
_isolationWindowTargetMz[ms2Scan] = ms2Spec.IsolationWindow.IsolationWindowTargetMz;
}
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
progData.StepRange(20.0);
ISequenceFilter ms1Filter;
if (this.ScanNumbers != null && this.ScanNumbers.Any())
{
ms1Filter = new SelectedMsMsFilter(this.ScanNumbers);
}
else if (string.IsNullOrWhiteSpace(FeatureFilePath))
{
// Checks whether SpecFileName.ms1ft exists
var ms1FtFilePath = MassSpecDataReaderFactory.ChangeExtension(SpecFilePath, LcMsFeatureFinderLauncher.FileExtension);
if (!File.Exists(ms1FtFilePath))
{
Console.WriteLine(@"Running ProMex...");
sw.Start();
var param = new LcMsFeatureFinderInputParameter
{
InputPath = SpecFilePath,
MinSearchMass = MinSequenceMass,
MaxSearchMass = MaxSequenceMass,
MinSearchCharge = MinPrecursorIonCharge,
MaxSearchCharge = MaxPrecursorIonCharge,
CsvOutput = false,
ScoreReport = false,
LikelihoodScoreThreshold = -10
};
var featureFinder = new LcMsFeatureFinderLauncher(param);
featureFinder.Run();
}
sw.Reset();
sw.Start();
Console.Write(@"Reading ProMex results...");
ms1Filter = new Ms1FtFilter(_run, PrecursorIonTolerance, ms1FtFilePath, -10);
}
else
{
sw.Reset();
sw.Start();
var extension = Path.GetExtension(FeatureFilePath);
if (extension.ToLower().Equals(".csv"))
{
Console.Write(@"Reading ICR2LS/Decon2LS results...");
ms1Filter = new IsosFilter(_run, PrecursorIonTolerance, FeatureFilePath);
}
else if (extension.ToLower().Equals(".ms1ft"))
{
Console.Write(@"Reading ProMex results...");
ms1Filter = new Ms1FtFilter(_run, PrecursorIonTolerance, FeatureFilePath, -10);
}
else if (extension.ToLower().Equals(".msalign"))
{
Console.Write(@"Reading MS-Align+ results...");
ms1Filter = new MsDeconvFilter(_run, PrecursorIonTolerance, FeatureFilePath);
}
else
{
ms1Filter = null; //new Ms1FeatureMatrix(_run);
}
}
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
// pre-generate deconvoluted spectra for scoring
_massBinComparer = new FilteredProteinMassBinning(AminoAcidSet, MaxSequenceMass + 1000);
_ms2ScorerFactory2 = new CompositeScorerFactory(_run, _massBinComparer, AminoAcidSet,
MinProductIonCharge, MaxProductIonCharge, ProductIonTolerance);
sw.Reset();
Console.WriteLine(@"Generating deconvoluted spectra for MS/MS spectra...");
sw.Start();
var pfeOptions = new ParallelOptions
{
MaxDegreeOfParallelism = MaxNumThreads,
CancellationToken = cancellationToken ?? CancellationToken.None
};
Parallel.ForEach(_ms2ScanNums, pfeOptions, ms2ScanNum =>
{
_ms2ScorerFactory2.DeconvonluteProductSpectrum(ms2ScanNum);
});
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
progData.StepRange(10.0);
progData.Status = "Reading Fasta File";
// Target database
var targetDb = new FastaDatabase(DatabaseFilePath);
targetDb.Read();
// Generate sequence tags for all MS/MS spectra
if (TagBasedSearch)
{
progData.StepRange(25.0);
progData.Status = "Generating Sequence Tags";
sw.Reset();
Console.WriteLine(@"Generating sequence tags for MS/MS spectra...");
sw.Start();
var seqTagGen = GetSequenceTagGenerator();
_tagMs2ScanNum = seqTagGen.GetMs2ScanNumsContainingTags().ToArray();
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
_tagSearchEngine = new ScanBasedTagSearchEngine(_run, seqTagGen, new LcMsPeakMatrix(_run, ms1Filter), targetDb, ProductIonTolerance, AminoAcidSet,
_ms2ScorerFactory2,
ScanBasedTagSearchEngine.DefaultMinMatchedTagLength,
MaxSequenceMass, MinProductIonCharge, MaxProductIonCharge);
}
var specFileName = MassSpecDataReaderFactory.RemoveExtension(Path.GetFileName(SpecFilePath));
var targetOutputFilePath = Path.Combine(OutputDir, specFileName + TargetFileNameEnding);
var decoyOutputFilePath = Path.Combine(OutputDir, specFileName + DecoyFileNameEnding);
var tdaOutputFilePath = Path.Combine(OutputDir, specFileName + TdaFileNameEnding);
progData.StepRange(60.0);
progData.Status = "Running Target search";
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Target))
{
sw.Reset();
Console.Write(@"Reading the target database...");
sw.Start();
targetDb.Read();
sw.Stop();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
var targetMatches = new SortedSet <DatabaseSequenceSpectrumMatch> [_run.MaxLcScan + 1];
progData.MaxPercentage = 42.5;
if (TagBasedSearch)
{
sw.Reset();
Console.WriteLine(@"Tag-based searching the target database");
sw.Start();
RunTagBasedSearch(targetMatches, targetDb, null, prog);
Console.WriteLine(@"Target database tag-based search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.MaxPercentage = 60.0;
sw.Reset();
Console.WriteLine(@"Searching the target database");
sw.Start();
RunSearch(targetMatches, targetDb, ms1Filter, null, prog);
Console.WriteLine(@"Target database search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
// calculate spectral e-value usign generating function
sw.Reset();
Console.WriteLine(@"Calculating spectral E-values for target-spectrum matches");
sw.Start();
var bestTargetMatches = RunGeneratingFunction(targetMatches);
WriteResultsToFile(bestTargetMatches, targetOutputFilePath, targetDb);
sw.Stop();
Console.WriteLine(@"Target-spectrum match E-value calculation elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.StepRange(95.0); // total to 95%
progData.Status = "Running Decoy search";
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Decoy))
{
// Decoy database
sw.Reset();
sw.Start();
var decoyDb = targetDb.Decoy(null, true);
Console.Write(@"Reading the decoy database...");
decoyDb.Read();
Console.WriteLine(@"Elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
progData.MaxPercentage = 77.5;
var decoyMatches = new SortedSet <DatabaseSequenceSpectrumMatch> [_run.MaxLcScan + 1];
if (TagBasedSearch)
{
sw.Reset();
Console.WriteLine(@"Tag-based searching the decoy database");
sw.Start();
RunTagBasedSearch(decoyMatches, decoyDb, null, prog);
Console.WriteLine(@"Decoy database tag-based search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.MaxPercentage = 95.0;
sw.Reset();
Console.WriteLine(@"Searching the decoy database");
sw.Start();
RunSearch(decoyMatches, decoyDb, ms1Filter, null, prog);
Console.WriteLine(@"Decoy database search elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
// calculate spectral e-value usign generating function
sw.Reset();
Console.WriteLine(@"Calculating spectral E-values for decoy-spectrum matches");
sw.Start();
var bestDecoyMatches = RunGeneratingFunction(decoyMatches);
WriteResultsToFile(bestDecoyMatches, decoyOutputFilePath, decoyDb);
sw.Stop();
Console.WriteLine(@"Decoy-spectrum match E-value calculation elapsed Time: {0:f1} sec", sw.Elapsed.TotalSeconds);
}
progData.StepRange(100.0);
progData.Status = "Writing combined results file";
if (RunTargetDecoyAnalysis.HasFlag(DatabaseSearchMode.Both))
{
// Add "Qvalue" and "PepQValue"
var fdrCalculator = new FdrCalculator(targetOutputFilePath, decoyOutputFilePath);
if (fdrCalculator.HasError())
{
ErrorMessage = fdrCalculator.ErrorMessage;
Console.WriteLine(@"Error computing FDR: " + fdrCalculator.ErrorMessage);
return(false);
}
fdrCalculator.WriteTo(tdaOutputFilePath);
}
progData.Report(100.0);
Console.WriteLine(@"Done.");
swAll.Stop();
Console.WriteLine(@"Total elapsed time for search: {0:f1} sec ({1:f2} min)", swAll.Elapsed.TotalSeconds, swAll.Elapsed.TotalMinutes);
return(true);
}
