public void TestProMexFilter()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string specFilePath = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\TopDown\ProductionQCShew\QC_Shew_Intact_26Sep14_Bane_C2Column3.raw";
if (!File.Exists(specFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, specFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(specFilePath, 0, 0);
const string ms1FtPath = @"\\proto-2\UnitTest_Files\InformedProteomics_TestFiles\TopDown\ProductionQCShew\QC_Shew_Intact_26Sep14_Bane_C2Column3.ms1ft";
if (!File.Exists(ms1FtPath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, ms1FtPath);
}
var filter = new Ms1FtFilter(run, new Tolerance(10), ms1FtPath);
// Console.WriteLine("ScanNums: {0}", string.Join("\t",filter.GetMatchingMs2ScanNums(8480.327609)));
Assert.IsTrue(filter.GetMatchingMs2ScanNums(8480.327609).Contains(5255));
}
public FeatureBasedTagSearchEngine(
LcMsRun run,
Ms1FtParser featureParser,
ProductScorerBasedOnDeconvolutedSpectra ms2Scorer,
SequenceTagParser tagParser,
FastaDatabase fastaDb,
Tolerance tolerance,
AminoAcidSet aaSet,
double maxSequenceMass = 50000.0,
int minProductIonCharge = 1,
int maxProductIonCharge = 20)
{
_run = run;
_ms2Scorer = ms2Scorer;
_featureParser = featureParser;
_ms1FtFilter = new Ms1FtFilter(run, tolerance, featureParser.Ms1FtFileName);
_tagParser = tagParser;
_fastaDb = fastaDb;
_searchableDb = new SearchableDatabase(fastaDb);
_tolerance = tolerance;
_aaSet = aaSet;
_maxSequenceMass = maxSequenceMass;
_minProductIonCharge = minProductIonCharge;
_maxProductIonCharge = maxProductIonCharge;
}
public void TestReadingProMexFile(double massToFind, string expectedScanNumbers)
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
var pbfFilePath = Utils.GetPbfTestFilePath(false);
var pbfFile = Utils.GetTestFile(methodName, pbfFilePath);
var promexFilePath = Path.Combine(Utils.DEFAULT_SPEC_FILES_FOLDER, "QC_Shew_Intact_26Sep14_Bane_C2Column3_Excerpt.ms1ft");
var promexFile = Utils.GetTestFile(methodName, promexFilePath);
var run = PbfLcMsRun.GetLcMsRun(pbfFile.FullName);
Console.Write("Reading ProMex results...");
var ms1Filter = new Ms1FtFilter(run, new Tolerance(10), promexFile.FullName);
Console.WriteLine();
var matchingScanNums = new SortedSet <int>();
foreach (var item in ms1Filter.GetMatchingMs2ScanNums(massToFind))
{
matchingScanNums.Add(item);
}
var scanNumList = string.Join(",", matchingScanNums);
Console.WriteLine("Scans with mass {0}:", massToFind);
Console.WriteLine(scanNumList);
var expectedScanNumList = expectedScanNumbers.Split(',');
var matchCount = 0;
foreach (var scanNumText in expectedScanNumList)
{
var scanNum = int.Parse(scanNumText);
if (!matchingScanNums.Contains(scanNum))
{
Assert.Fail("Did not find scan {0} for mass {1}", scanNum, massToFind);
}
matchCount++;
}
Assert.AreEqual(matchCount, matchingScanNums.Count, "Found extra matching scan nums vs. what was expected");
}
public void TestReadingProMexFile()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string rawFilePath = @"\\proto-11\MSXML_Cache\PBF_Gen_1_214\2014_3\QC_Shew_Intact_26Sep14_Bane_C2Column3.pbf";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(rawFilePath);
const string promexFileName = @"\\Proto-5\VOrbiETD02\2014_3\QC_Shew_Intact_26Sep14_Bane_C2Column3\MSP201508271107_Auto1226713\QC_Shew_Intact_26Sep14_Bane_C2Column3.ms1ft";
Console.Write("Reading ProMex results...");
var ms1Filter = new Ms1FtFilter(run, new Tolerance(10), promexFileName);
Console.WriteLine(string.Join(",", ms1Filter.GetMatchingMs2ScanNums(3016.6583)));
}
public void TestFeatureId()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string dataSet = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3";
if (!File.Exists(dataSet))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, dataSet);
}
// Feature: 5236-5286 6-12 8480.3681 5
const int minScanNum = 5236;
const int maxScanNum = 5286;
const double featureMass = 8480.3681;
//const int minScanNum = 7251;
//const int maxScanNum = 7326;
//const double featureMass = 32347.18;
// const int minScanNum = 4451;
// const int maxScanNum = 4541;
// const double featureMass = 31267.95;
var tolerance = new Tolerance(10);
var relaxedTolerance = new Tolerance(20);
const int minTagLength = 5;
const int minMergedTagLength = 7;
const int minNumTagMatches = 1;
var rawFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".raw");
var run = PbfLcMsRun.GetLcMsRun(rawFileName);
var aminoAcidSet = AminoAcidSet.GetStandardAminoAcidSet();
var featureFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".ms1ft");
var filter = new Ms1FtFilter(run, tolerance, featureFileName);
var ms2ScanNums =
filter.GetMatchingMs2ScanNums(featureMass)
.Where(scanNum => scanNum > minScanNum && scanNum < maxScanNum)
.ToArray();
const string tagFileName = dataSet + ".seqtag"; //"_MinLength3.seqtag"; //Path.ChangeExtension(dataSet, ".seqtag");
const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fastaDb = new FastaDatabase(fastaFilePath);
var searchableDb = new SearchableDatabase(fastaDb);
var tagParser = new SequenceTagParser(tagFileName, minTagLength);
var proteinsToTags = new Dictionary <string, IList <MatchedTag> >();
foreach (var ms2ScanNum in ms2ScanNums)
{
var tags = tagParser.GetSequenceTags(ms2ScanNum);
foreach (var tag in tags)
{
var matchedIndices = searchableDb.FindAllMatchedSequenceIndices(tag.Sequence).ToArray();
foreach (var index in matchedIndices)
{
var protein = fastaDb.GetProteinName(index);
var startIndex = fastaDb.GetZeroBasedPositionInProtein(index);
var matchedTag = new MatchedTag(tag, startIndex, featureMass);
IList <MatchedTag> existingTags;
if (proteinsToTags.TryGetValue(protein, out existingTags))
{
existingTags.Add(matchedTag);
}
else
{
proteinsToTags.Add(protein, new List <MatchedTag> {
matchedTag
});
}
}
}
}
foreach (var entry in proteinsToTags.OrderByDescending(e => e.Value.Count))
{
if (entry.Value.Count < minNumTagMatches)
{
break;
}
var proteinName = entry.Key;
var proteinSequence = fastaDb.GetProteinSequence(proteinName);
var protein = new Sequence(proteinSequence, aminoAcidSet);
Console.WriteLine(proteinName + "\t" + entry.Value.Count);
var matchedTagSet = new MatchedTagSet(proteinSequence, aminoAcidSet,
tolerance, relaxedTolerance);
Console.WriteLine("********** Before merging");
foreach (var matchedTag in entry.Value)
{
var seq = proteinSequence.Substring(matchedTag.StartIndex,
matchedTag.EndIndex - matchedTag.StartIndex);
var nTermMass = protein.GetMass(0, matchedTag.StartIndex);
var cTermMass = protein.GetMass(matchedTag.EndIndex, protein.Count);
Console.WriteLine("\t{0}\t{1}\t{2}\t{3}\t{4}\t{5}",
(matchedTag.NTermFlankingMass - nTermMass), seq, (matchedTag.CTermFlankingMass - cTermMass), matchedTag.StartIndex,
matchedTag.IsNTermFlankingMassReliable, matchedTag.IsCTermFlankingMassReliable);
matchedTagSet.Add(matchedTag);
}
Console.WriteLine("********** After merging");
foreach (var matchedTag in matchedTagSet.Tags)
{
if (matchedTag.Length < minMergedTagLength)
{
continue;
}
var seq = proteinSequence.Substring(matchedTag.StartIndex,
matchedTag.EndIndex - matchedTag.StartIndex);
var nTermMass = protein.GetMass(0, matchedTag.StartIndex);
var cTermMass = protein.GetMass(matchedTag.EndIndex, protein.Count);
Console.WriteLine("\t{0}\t{1}\t{2}\t{3}\t{4}\t{5}",
(matchedTag.NTermFlankingMass - nTermMass), seq, (matchedTag.CTermFlankingMass - cTermMass), matchedTag.StartIndex,
matchedTag.IsNTermFlankingMassReliable, matchedTag.IsCTermFlankingMassReliable);
}
break;
}
}
public void TestMs1Filter()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
// QC_Shew
const string specFilePath = @"D:\MassSpecFiles\training\raw\QC_Shew_Intact_26Sep14_Bane_C2Column3.pbf";
const string ms1FtFileName = @"D:\MassSpecFiles\training\raw\QC_Shew_Intact_26Sep14_Bane_C2Column3.ms1ft";
const string idFilePath = @"D:\MassSpecFiles\training\IcTda\QC_Shew_Intact_26Sep14_Bane_C2Column3_IcTda.tsv";
if (!File.Exists(specFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, specFilePath);
}
if (!File.Exists(idFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, idFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(specFilePath);
var massTolerance = new Tolerance(10);
var ms1ftFilter = new Ms1FtFilter(run, massTolerance, ms1FtFileName);
var n = 0;
var ms2ScanNums = run.GetScanNumbers(2);
foreach (var ms2ScanNum in ms2ScanNums)
{
var matchingMass = ms1ftFilter.GetMatchingMass(ms2ScanNum);
n += matchingMass.Count();
}
Console.WriteLine("{0} / {1}", n, ms2ScanNums.Count);
/*
* var tsvReader = new TsvFileParser(idFilePath);
*
* for (var i = 0; i < tsvReader.NumData; i++)
* {
* var qv = double.Parse(tsvReader.GetData("QValue")[i]);
* if (qv > 0.01) break;
*
* var scan = int.Parse(tsvReader.GetData("Scan")[i]);
* var charge = int.Parse(tsvReader.GetData("Charge")[i]);
* var mass = double.Parse(tsvReader.GetData("Mass")[i]);
*
* if (mass > 15000) continue;
*
* var seq = tsvReader.GetData("Sequence")[i];
* var mod = tsvReader.GetData("Modifications")[i];
* var nMatched = int.Parse(tsvReader.GetData("#MatchedFragments")[i]);
*
* var hit = false;
* foreach (var ms2Scan in ms1ftFilter.GetMatchingMs2ScanNums(mass))
* {
* if (ms2Scan == scan)
* {
* hit = true;
* break;
* }
* }
*
* if (!hit)
* {
* Console.WriteLine("{0}\t{1}\t{2}", scan, mass, nMatched);
* }
* }*/
}
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);
}
