public void TestPathUtils()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
const string rawFilePath = @"C:\cygwin\home\kims336\Data\TopDownJia\raw\Synocho_D1_1.raw";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
Console.WriteLine(MassSpecDataReaderFactory.RemoveExtension(rawFilePath) + "_Target.tsv");
Console.WriteLine(Path.GetDirectoryName(rawFilePath));
Console.WriteLine(Path.Combine(Path.GetDirectoryName(rawFilePath), Path.GetFileNameWithoutExtension(rawFilePath) + "_IcTarget.tsv"));
var outputDir = @"C:\cygwin\home\kims336\Data\TopDownJia\raw\L1_1_Mode2\Synocho_L1_1_IcTarget.tsv";
if (!Directory.Exists(outputDir))
{
if (!File.GetAttributes(outputDir).HasFlag(FileAttributes.Directory))
{
throw new Exception(outputDir + " is not a directory!");
}
Directory.CreateDirectory(outputDir);
}
Console.WriteLine(outputDir);
}
/// <summary>
/// Process the given file or folder
/// </summary>
/// <returns>0 if success, otherwise an error code</returns>
/// <remarks>Folder could either be a supported mass spec data folder, or a normal directory with several supported data files</remarks>
public int Run()
{
// Normalize the input path. Only affects paths to a file/folder in a folder-type dataset
Parameters.InputPath = MassSpecDataReaderFactory.NormalizeDatasetPath(Parameters.InputPath);
var attr = File.GetAttributes(Parameters.InputPath);
int errorCode;
if ((attr & FileAttributes.Directory) == FileAttributes.Directory &&
!MassSpecDataReaderFactory.IsADirectoryDataset(Parameters.InputPath))
{
errorCode = ProcessDirectory(Parameters.InputPath);
}
else
{
if (!MsRawFile(Parameters.InputPath) && !MsPbfFile(Parameters.InputPath))
{
ShowErrorMessage(@"File extension not supported, " + Parameters.InputPath);
return(-1);
}
errorCode = ProcessFile(Parameters.InputPath);
}
return(errorCode);
}
public bool Validate()
{
// Check for folder-type datasets, and replace specFilePath with the directory name if it is.
SourcePath = MassSpecDataReaderFactory.GetDatasetName(SourcePath);
var isDirectoryDataset = MassSpecDataReaderFactory.IsADirectoryDataset(SourcePath);
// True if specFilePath is a directory that is NOT a supported folder-type dataset.
var specPathIsDirectory = Directory.Exists(SourcePath) && !isDirectoryDataset;
if (!File.Exists(SourcePath) && !specPathIsDirectory && !isDirectoryDataset)
{
PrintError("File not found: " + SourcePath);
return(false);
}
var types = MassSpecDataReaderFactory.MassSpecDataTypeFilterList;
types.Remove(".pbf");
if (!specPathIsDirectory && !(types.Select(ext => SourcePath.ToLower().EndsWith(ext)).Any()))
{
PrintError("Invalid file extension: (" + Path.GetExtension(SourcePath) + ") " + SourcePath);
return(false);
}
if (string.IsNullOrWhiteSpace(OutputDir))
{
// Must use "Path.GetFullPath" to return the absolute path when the source file is in the working directory
// But, it could cause problems with too-long paths.
OutputDir = specPathIsDirectory ? SourcePath : Path.GetDirectoryName(Path.GetFullPath(SourcePath));
}
if (string.IsNullOrWhiteSpace(OutputDir))
{
PrintError("Invalid output file directory: " + OutputDir);
return(false);
}
if (EndScan < StartScan && EndScan > -1)
{
PrintError($"End scan ({EndScan}) must not be less than start scan ({StartScan})!");
return(false);
}
if (!Directory.Exists(OutputDir))
{
if (File.Exists(OutputDir) && !File.GetAttributes(OutputDir).HasFlag(FileAttributes.Directory))
{
PrintError("OutputDir \"" + OutputDir + "\" is not a directory!");
return(false);
}
Directory.CreateDirectory(OutputDir);
}
return(true);
}
public void TestTagBasedSearch()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
// const string rawFilePath = @"H:\Research\Lewy\raw\Lewy_intact_01.raw";
// const string rawFilePath = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3.raw";
// const string rawFilePath = @"H:\Research\Yufeng\TopDownYufeng\raw\yufeng_column_test2.raw";
// const string rawFilePath = @"H:\Research\Weijun_TopDown\raw\UC4_Intact_plasmaTest_90_6May15_Bane_14-09-01RZ.raw";
// const string rawFilePath = @"H:\Research\Charles\TopDown\raw\SBEP_STM_001_02272012_Aragon.raw";
const string rawFilePath = @"D:\MassSpecFiles\60k\Yufeng_SampleTest1_150614113438.pbf";
//const string rawFilePath = @"D:\MassSpecFiles\60k\NCR_50K_Test_24Jun15_Bane_15-02-02RZ.pbf";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(rawFilePath);
//const int minTagLength = 5;
var tagFilePath = MassSpecDataReaderFactory.ChangeExtension(rawFilePath, ".seqtag");
//var tagParser = new SequenceTagParser(tagFilePath, minTagLength, 100);
const string fastaFilePath = @"D:\MassSpecFiles\60k\ID_003836_DA9CC1E4.fasta";
//const string fastaFilePath = @"D:\MassSpecFiles\60k\ID_004973_9BA6912F.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fastaDb = new FastaDatabase(fastaFilePath);
var tolerance = new Tolerance(10);
// var modsFilePath = @"H:\Research\QCShew_TopDown\Production\Mods_Methyl.txt";
var modsFilePath = @"D:\MassSpecFiles\60k\Mods.txt";
// var modsFilePath = "";
if (!File.Exists(modsFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, modsFilePath);
}
var aaSet = new AminoAcidSet(modsFilePath);
TestTagBasedSearch(run, fastaDb, tolerance, aaSet);
}
public void TestTagBasedSearchCompRef()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string dataSetPath = @"D:\MassSpecFiles\CompRef";
const string fastaFilePath = @"D:\MassSpecFiles\CompRef\ID_003278_4B4B3CB1.fasta";
const string modsFilePath = @"D:\MassSpecFiles\CompRef\Mods.txt";
if (!Directory.Exists(dataSetPath))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, dataSetPath);
}
if (!File.Exists(modsFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, modsFilePath);
}
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fileEntries = Directory.GetFiles(dataSetPath);
var dataset = (from fileName in fileEntries where fileName.EndsWith("pbf") select Path.GetFileNameWithoutExtension(fileName)).ToList();
dataset.Sort();
var fastaDb = new FastaDatabase(fastaFilePath);
var tolerance = new Tolerance(10);
var aaSet = new AminoAcidSet(modsFilePath);
for (var i = 0; i < dataset.Count; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", dataSetPath, dataset[i]);
var ms1File = string.Format(@"{0}\{1}.ms1ft", dataSetPath, dataset[i]);
var tagFilePath = MassSpecDataReaderFactory.ChangeExtension(rawFile, ".seqtag");
var run = PbfLcMsRun.GetLcMsRun(rawFile);
const int minTagLength = 5;
//var tagParser = new SequenceTagParser(tagFilePath, minTagLength, 100);
Console.WriteLine("-----------------{0}--------------------", rawFile);
TestTagBasedSearch(run, fastaDb, tolerance, aaSet);
Console.WriteLine("-----------------------------------------------------------------------");
}
}
/// <summary>
/// Initialize this data set by reading the raw file asynchronously and initializing child view models
/// </summary>
/// <param name="filePath">The raw File Path.</param>
/// <returns>
/// The <see cref="Task"/>.
/// </returns>
public async Task InitializeAsync(string filePath)
{
filePath = MassSpecDataReaderFactory.NormalizeDatasetPath(filePath);
this.IsLoading = true; // Show animated loading screen
this.Title = Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(filePath));
this.rawFilePath = filePath;
this.LoadProgressPercent = 0.0;
this.LoadProgressStatus = "Loading...";
var progress = new Progress <ProgressData>(progressData =>
{
progressData.UpdateFrequencySeconds = 2;
if (progressData.ShouldUpdate())
{
this.LoadProgressPercent = progressData.Percent;
this.LoadProgressStatus = progressData.Status;
}
});
// load raw file
this.LcMs = await Task.Run(() => PbfLcMsRun.GetLcMsRun(filePath, 0, 0, progress));
// Now that we have an LcMsRun, initialize viewmodels that require it
this.XicViewModel = new XicViewModel(this.dialogService, this.LcMs);
this.SpectrumViewModel = new SpectrumViewModel(this.dialogService, this.LcMs);
this.FeatureMapViewModel = new FeatureViewerViewModel((LcMsRun)this.LcMs, this.dialogService);
// When the selected scan changes in the xic plots, the selected scan for the prsm should update
this.XicViewModel.SelectedScanUpdated().Subscribe(scan => this.SelectedPrSm.Scan = scan);
// When an ID is selected on FeatureMap, update selectedPrSm
this.FeatureMapViewModel.FeatureMapViewModel.WhenAnyValue(x => x.SelectedPrSm).Where(prsm => prsm != null).Subscribe(prsm => this.SelectedPrSm = prsm);
// Create prsms for scan numbers (unidentified)
await this.LoadScans();
////await this.ScanViewModel.ToggleShowInstrumentDataAsync(IcParameters.Instance.ShowInstrumentData, (PbfLcMsRun)this.LcMs);
this.SelectedPrSm.LcMs = this.LcMs; // For the selected PrSm, we should always use the LcMsRun for this dataset.
this.IsLoading = false; // Hide animated loading screen
}
public void TestTagMatchingSingleSpec()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string dataSet = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3";
const int scanNum = 4533;
// Parse sequence tags
var tagFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".seqtag");
const int minTagLength = 8;
const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.fasta";
if (!File.Exists(tagFileName))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, tagFileName);
}
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 tags = tagParser.GetSequenceTags(scanNum);
foreach (var tag in tags)
{
var matchedProteins = searchableDb.FindAllMatchedSequenceIndices(tag.Sequence)
.Select(index => fastaDb.GetProteinName(index)).ToArray();
if (matchedProteins.Any())
{
Console.WriteLine("{0}\t{1}\t{2}\t{3}", tag.Sequence, tag.IsPrefix, tag.FlankingMass, string.Join("\t", matchedProteins));
}
}
}
public void TestTagBasedSearchForLewy()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
const string rawFilePath = @"D:\MassSpecFiles\Lewy\Lewy_AT_AD1_21May15_Bane_14-09-01RZ.pbf";
if (!File.Exists(rawFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFilePath);
}
var run = PbfLcMsRun.GetLcMsRun(rawFilePath);
//const int minTagLength = 4;
var tagFilePath = MassSpecDataReaderFactory.ChangeExtension(rawFilePath, ".seqtag");
//var tagParser = new SequenceTagParser(tagFilePath, minTagLength, 10000);
const string fastaFilePath = @"D:\MassSpecFiles\Lewy\a4_human.fasta";
if (!File.Exists(fastaFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, fastaFilePath);
}
var fastaDb = new FastaDatabase(fastaFilePath);
var tolerance = new Tolerance(10);
var modsFilePath = @"D:\MassSpecFiles\Lewy\Mods.txt";
if (!File.Exists(modsFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, modsFilePath);
}
var aaSet = new AminoAcidSet(modsFilePath);
TestTagBasedSearch(run, fastaDb, tolerance, aaSet);
}
/// <summary>
/// Create a PNG image of previously found MS1 features
/// </summary>
/// <param name="pbfFilePath">.pbf file path</param>
/// <param name="ms1FeaturesFilePath">.ms1ft file path</param>
/// <returns>0 if success, otherwise an error code</returns>
/// <remarks>
/// If ms1FeaturesFilePath is an empty string, it is auto-determined based on the .pbf file name
/// Mass range is determined using Parameters.MinSearchMass and Parameters.MaxSearchMass
/// </remarks>
public int CreateFeatureMapImage(string pbfFilePath, string ms1FeaturesFilePath)
{
if (!File.Exists(pbfFilePath))
{
Console.WriteLine(@"Error: Data file not found: " + pbfFilePath);
return(-1);
}
var outDirectory = GetOutputDirectory(pbfFilePath);
if (string.IsNullOrEmpty(outDirectory))
{
return(-2);
}
var baseName = Path.GetFileName(MassSpecDataReaderFactory.RemoveExtension(pbfFilePath));
if (string.IsNullOrEmpty(ms1FeaturesFilePath) || string.Equals(ms1FeaturesFilePath, "."))
{
ms1FeaturesFilePath = Path.Combine(outDirectory, baseName + "." + FileExtension);
}
var pngFilePath = Path.Combine(outDirectory, baseName + "_" + FileExtension + ".png");
if (!File.Exists(ms1FeaturesFilePath))
{
Console.WriteLine(@"Error: MS1 features file not found: " + ms1FeaturesFilePath);
return(-3);
}
Console.WriteLine(@"Start loading MS1 data from {0}", pbfFilePath);
var run = PbfLcMsRun.GetLcMsRun(pbfFilePath);
CreateFeatureMapImage(run, ms1FeaturesFilePath, pngFilePath);
return(0);
}
/// <summary>
/// Find features in the data file
/// </summary>
/// <param name="rawFile">Data file (either a pbf file or a file type from which a pbf file can be auto-created)</param>
/// <returns>0 if success; negative number on error</returns>
private int ProcessFile(string rawFile)
{
var outDirectory = GetOutputDirectory(rawFile);
if (string.IsNullOrEmpty(outDirectory))
{
return(-1);
}
var baseName = Path.GetFileName(MassSpecDataReaderFactory.RemoveExtension(rawFile));
var ms1FeaturesFilePath = Path.Combine(outDirectory, baseName + "." + FileExtension);
var outCsvFilePath = Path.Combine(outDirectory, baseName + "_" + FileExtension + ".csv");
var pngFilePath = Path.Combine(outDirectory, baseName + "_" + FileExtension + ".png");
if (File.Exists(ms1FeaturesFilePath))
{
Console.WriteLine(@"ProMex output already exists: {0}", ms1FeaturesFilePath);
return(-2);
}
if (!File.Exists(rawFile))
{
ShowErrorMessage(@"Cannot find input file: " + rawFile);
return(-3);
}
var stopwatch = Stopwatch.StartNew();
Console.WriteLine(@"Start loading MS1 data from {0}", rawFile);
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var featureFinder = new LcMsPeakMatrix(run, _likelihoodScorer, 1, 60, Parameters.MaxThreads);
Console.WriteLine(@"Complete loading MS1 data. Elapsed Time = {0:0.000} sec", (stopwatch.ElapsedMilliseconds) / 1000.0d);
if (run.GetMs1ScanVector().Length == 0)
{
ShowErrorMessage(@"Data file has no MS1 spectra: " + Path.GetFileName(rawFile));
return(-4);
}
var comparer = featureFinder.Comparer;
var container = new LcMsFeatureContainer(featureFinder.Ms1Spectra, _likelihoodScorer, new LcMsFeatureMergeComparer(new Tolerance(10)));
var minSearchMassBin = comparer.GetBinNumber(Parameters.MinSearchMass);
var maxSearchMassBin = comparer.GetBinNumber(Parameters.MaxSearchMass);
double totalMassBin = maxSearchMassBin - minSearchMassBin + 1;
Console.WriteLine(@"Start MS1 feature extraction.");
stopwatch.Restart();
for (var binNum = minSearchMassBin; binNum <= maxSearchMassBin; binNum++)
{
var clusters = featureFinder.FindFeatures(binNum);
container.Add(clusters);
if (binNum > minSearchMassBin && (binNum - minSearchMassBin) % 1000 == 0)
{
var elapsed = (stopwatch.ElapsedMilliseconds) / 1000.0d;
var processedBins = binNum - minSearchMassBin;
var processedPercentage = ((double)processedBins / totalMassBin) * 100;
Console.WriteLine(@"Processing {0:0.0}% of mass bins ({1:0.0} Da); elapsed time = {2:0.000} sec; # of features = {3}",
processedPercentage, featureFinder.Comparer.GetMzEnd(binNum), elapsed,
container.NumberOfFeatures);
}
}
Console.WriteLine(@"Complete MS1 feature extraction.");
Console.WriteLine(@" - Elapsed time = {0:0.000} sec", (stopwatch.ElapsedMilliseconds) / 1000.0d);
Console.WriteLine(@" - Number of extracted features = {0}", container.NumberOfFeatures);
Console.WriteLine(@"Start selecting mutually independent features from feature network graph");
stopwatch.Restart();
// write result files
var tsvWriter = new StreamWriter(ms1FeaturesFilePath);
tsvWriter.WriteLine(GetHeaderString(Parameters.ScoreReport));
StreamWriter csvWriter = null;
if (Parameters.CsvOutput)
{
csvWriter = new StreamWriter(outCsvFilePath);
csvWriter.WriteLine("scan_num,charge,abundance,mz,fit,monoisotopic_mw,FeatureID");
}
var filteredFeatures = container.GetFilteredFeatures(featureFinder);
var featureId = 0;
foreach (var feature in filteredFeatures)
{
featureId++;
tsvWriter.WriteLine("{0}\t{1}", featureId, GetString(feature, Parameters.ScoreReport));
var mostAbuIdx = feature.TheoreticalEnvelope.IndexOrderByRanking[0];
if (csvWriter != null)
{
foreach (var envelope in feature.EnumerateEnvelopes())
{
//var mostAbuIsotopeInternalIndex = cluster.IsotopeList.SortedIndexByIntensity[0];
var mostAbuPeak = envelope.Peaks[mostAbuIdx];
if (mostAbuPeak == null || !mostAbuPeak.Active)
{
continue;
}
var fitscore = 1.0 - feature.BestCorrelationScore;
csvWriter.WriteLine(string.Format("{0},{1},{2},{3},{4},{5},{6}", envelope.ScanNum, envelope.Charge, envelope.Abundance, mostAbuPeak.Mz, fitscore, envelope.MonoMass, featureId));
}
}
}
tsvWriter.Close();
Console.WriteLine(@"Complete feature filtration");
Console.WriteLine(@" - Elapsed time = {0:0.000} sec", (stopwatch.ElapsedMilliseconds) / 1000.0d);
Console.WriteLine(@" - Number of filtered features = {0}", featureId);
Console.WriteLine(@" - ProMex output: {0}", ms1FeaturesFilePath);
if (csvWriter != null)
{
csvWriter.Close();
Console.WriteLine(@" - ProMex output in ICR2LS format: {0}", outCsvFilePath);
}
if (Parameters.FeatureMapImage)
{
CreateFeatureMapImage(run, ms1FeaturesFilePath, pngFilePath);
}
return(0);
}
private bool HasExistingPbfFile(string path)
{
return(File.Exists(MassSpecDataReaderFactory.ChangeExtension(path, ".pbf")));
}
/// <summary>
/// Look for the default .pbf testing file in the default test file folder,
/// plus also in the UnitTest_Files directory in this project or solution
/// </summary>
/// <param name="createIfMissing">If true and the .mzML file is ofund, create the .pbf file</param>
/// <returns>Path to the file if found, otherwise the default path on Proto-2</returns>
public static string GetPbfTestFilePath(bool createIfMissing)
{
var methodName = MethodBase.GetCurrentMethod().Name;
var pbfFilePath = Path.Combine(DEFAULT_SPEC_FILES_FOLDER, PBF_TEST_FILE);
var pbfFileInfo = GetTestFile("PbfTestFilePath", pbfFilePath, false);
if (pbfFileInfo != null)
{
// Check for a lock file, which would indicate another process is creating the .pbf file
var existingLockFile = new FileInfo(Path.Combine(pbfFileInfo.FullName + ".lock"));
WaitForLockFile(existingLockFile, false);
pbfFileInfo.Refresh();
if (pbfFileInfo.Exists)
{
DeleteLockFile(existingLockFile);
return(pbfFileInfo.FullName);
}
}
if (!createIfMissing)
{
// Not creating the file if missing; return the default path, despite the fact that the file does not exist
// The calling method will discover that the file is missing and act accordingly
return(pbfFilePath);
}
// Create the missing file
var mzmlFilePath = Path.Combine(DEFAULT_SPEC_FILES_FOLDER, MZML_TEST_FILE);
var mzmlFileInfo = GetTestFile("MzmlTestFilePath", mzmlFilePath, false);
if (mzmlFileInfo?.DirectoryName == null)
{
// Unable to create the pbf file; return the default path, despite the fact that the file does not exist
return(pbfFilePath);
}
ShowMessage(methodName, string.Format("Creating {0} using {1}", PBF_TEST_FILE, mzmlFileInfo.FullName));
var lockFile = new FileInfo(Path.Combine(mzmlFileInfo.DirectoryName, Path.GetFileNameWithoutExtension(mzmlFileInfo.Name) + ".pbf.lock"));
var newPbfFilePath = Path.Combine(mzmlFileInfo.DirectoryName, PBF_TEST_FILE);
try
{
// Create a new lock file so other processes know this thread is creating the .pbf file
WaitForLockFile(lockFile, true);
mLastStatus = string.Empty;
var startTime = DateTime.UtcNow;
var reader = MassSpecDataReaderFactory.GetMassSpecDataReader(mzmlFileInfo.FullName);
var progress = new Progress <ProgressData>(p =>
{
p.UpdateFrequencySeconds = 2;
if (p.Percent < 100 && (p.Percent % 25).Equals(0) || p.ShouldUpdate())
{
var statusMessage = string.Format("{0}, {1:00.0}% complete ", p.Status, p.Percent);
if (string.Equals(mLastStatus, statusMessage))
{
return;
}
mLastStatus = statusMessage;
Console.Write("\r{0}, {1:00.0}% complete ", p.Status, p.Percent);
}
});
var run = new PbfLcMsRun(mzmlFileInfo.FullName, reader, newPbfFilePath, 0, 0, progress);
Console.WriteLine();
ShowMessage(methodName, string.Format("Created {0} in {1:F0} seconds", run.PbfFilePath, DateTime.UtcNow.Subtract(startTime).TotalSeconds));
DeleteLockFile(lockFile);
return(run.PbfFilePath);
}
catch (Exception ex)
{
ShowMessage(methodName, string.Format("Exception creating {0} using {1}: {2}", PBF_TEST_FILE, mzmlFileInfo.FullName, ex.Message));
try
{
var incompletePbfFile = new FileInfo(newPbfFilePath);
if (incompletePbfFile.Exists)
{
incompletePbfFile.Delete();
}
DeleteLockFile(lockFile);
}
catch
{
// Ignore errors here
}
// Return the default path, despite the fact that the file does not exist
return(pbfFilePath);
}
}
public void TestThermoRawAvailable()
{
Assert.AreEqual(true, MassSpecDataReaderFactory.IsThermoRawAvailable());
}
public void TestPwizAvailable()
{
Assert.AreEqual(true, MassSpecDataReaderFactory.IsPwizAvailable());
}
public static int Main(string[] args)
{
PbfGenInputParameters options;
try
{
var osVersionInfo = new clsOSVersionInfo();
if (osVersionInfo.GetOSVersion().ToLower().Contains("windows"))
{
var handle = Process.GetCurrentProcess().MainWindowHandle;
SetConsoleMode(handle, EnableExtendedFlags);
}
var exeName = System.Reflection.Assembly.GetEntryAssembly().GetName().Name;
var parser = new CommandLineParser <PbfGenInputParameters>(exeName, Version);
parser.UsageExamples.Add($"Using -start and -end to limit the scan range to include in the .pbf file\n\t{exeName}.exe -s Dataset.raw -start 2000 -end 3000");
var results = parser.ParseArgs(args);
if (!results.Success)
{
// Wait for 1.5 seconds
System.Threading.Thread.Sleep(1500);
return(-1);
}
if (!results.ParsedResults.Validate())
{
parser.PrintHelp();
// Wait for 1.5 seconds
System.Threading.Thread.Sleep(1500);
return(-1);
}
options = results.ParsedResults;
}
catch (Exception ex)
{
Console.WriteLine("Exception while parsing the command line parameters: " + ex.Message);
return(-5);
}
#if (!DEBUG)
try
#endif
{
var specFilePaths = new[] { options.SourcePath };
if (Directory.Exists(options.SourcePath) && !MassSpecDataReaderFactory.IsADirectoryDataset(options.SourcePath))
{
specFilePaths = Directory.GetFiles(options.SourcePath, "*.raw"); // TODO: Support folders with other formats in them too...
}
foreach (var rawFilePath in specFilePaths)
{
var pbfFileName = MassSpecDataReaderFactory.ChangeExtension(rawFilePath, PbfLcMsRun.FileExtensionConst);
var pbfFilePath = Path.Combine(options.OutputDir, Path.GetFileName(pbfFileName));
if (File.Exists(pbfFilePath) && PbfLcMsRun.CheckFileFormatVersion(pbfFilePath, out var isCurrent) && isCurrent)
{
Console.WriteLine("{0} already exists.", pbfFilePath);
continue;
}
Console.WriteLine("Creating {0} from {1}", pbfFilePath, rawFilePath);
if (options.StartScan > 0 && options.EndScan > 0)
{
Console.WriteLine("Only including scans {0} to {1}", options.StartScan, options.EndScan);
}
else if (options.StartScan > 0)
{
Console.WriteLine("Only including scans {0} to the end", options.StartScan);
}
else if (options.EndScan > 0)
{
Console.WriteLine("Only including scans 1 to {0}", options.EndScan);
}
var reader = MassSpecDataReaderFactory.GetMassSpecDataReader(rawFilePath);
var progress = new Progress <ProgressData>(p =>
{
p.UpdateFrequencySeconds = 2;
if ((p.Percent % 25).Equals(0) || p.ShouldUpdate())
{
Console.Write("\r{0}, {1:00.0}% complete ", p.Status, p.Percent);
}
});
var run = new PbfLcMsRun(rawFilePath, reader, pbfFilePath, 0, 0, progress, false, options.StartScan, options.EndScan);
Console.WriteLine();
}
Console.WriteLine("PbfFormatVersion: {0}", PbfLcMsRun.FileFormatId);
return(0);
}
#if (!DEBUG)
catch (Exception ex)
{
// NOTE: The DMS Analysis Manager looks for this text; do not change it
Console.WriteLine("Exception while processing: " + ex.Message);
Console.WriteLine(ex.StackTrace);
var errorCode = -Math.Abs(ex.Message.GetHashCode());
if (errorCode == 0)
{
return(-1);
}
else
{
return(errorCode);
}
}
#endif
}
public void TestFeatureIdMatching()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
const string resultFilePath = @"H:\Research\QCShew_TopDown\Production\M1_V092\QC_Shew_Intact_26Sep14_Bane_C2Column3_IcTda.tsv";
if (!File.Exists(resultFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, resultFilePath);
}
var resultParser = new MsPathFinderParser(resultFilePath);
const double qValueThreshold = 0.01;
const double tolerancePpm = 13;
const string dataSet = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3";
var rawFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".raw");
if (!File.Exists(rawFileName))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFileName);
}
var run = PbfLcMsRun.GetLcMsRun(rawFileName);
var idList =
resultParser.GetIdList().TakeWhile(id => id.QValue <= qValueThreshold).OrderBy(id => id.Mass).ToList();
var idMassList = idList.Select(id => id.Mass).ToList();
var idFlag = new bool[idList.Count];
// Parse sequence tags
var tagFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".seqtag");
const int minTagLength = 6;
const int numProtMatches = 4;
// const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.fasta";
const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.icsfldecoy.fasta";
if (!File.Exists(tagFileName))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, tagFileName);
}
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 featureFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".ms1ft");
var featureParser = new TsvFileParser(featureFileName);
var minScan = featureParser.GetData("MinScan").Select(s => Convert.ToInt32(s)).ToArray();
var maxScan = featureParser.GetData("MaxScan").Select(s => Convert.ToInt32(s)).ToArray();
var minCharge = featureParser.GetData("MinCharge").Select(s => Convert.ToInt32(s)).ToArray();
var maxCharge = featureParser.GetData("MaxCharge").Select(s => Convert.ToInt32(s)).ToArray();
var monoMass = featureParser.GetData("MonoMass").Select(Convert.ToDouble).ToArray();
var numFeaturesWithId = 0;
var numFeaturesWithMs2 = 0;
var numFeaturesWithTags = 0;
var numFeaturesWithMatchingTags = 0;
var numFeaturesWithTwoOrMoreMatchingTags = 0;
var numFeaturesWithNoIdAndMatchingTags = 0;
for (var i = 0; i < featureParser.NumData; i++)
{
var mass = monoMass[i];
// Find Id
var tolDa = new Tolerance(tolerancePpm).GetToleranceAsDa(mass, 1);
var minMass = mass - tolDa;
var maxMass = mass + tolDa;
var index = idMassList.BinarySearch(mass);
if (index < 0)
{
index = ~index;
}
var matchedId = new List <MsPathFinderId>();
// go down
var curIndex = index - 1;
while (curIndex >= 0)
{
var curId = idList[curIndex];
if (curId.Mass < minMass)
{
break;
}
if (curId.Scan > minScan[i] && curId.Scan < maxScan[i] &&
curId.Charge >= minCharge[i] && curId.Charge <= maxCharge[i])
{
matchedId.Add(curId);
idFlag[curIndex] = true;
}
--curIndex;
}
// go up
curIndex = index;
while (curIndex < idList.Count)
{
var curId = idList[curIndex];
if (curId.Mass > maxMass)
{
break;
}
if (curId.Scan >= minScan[i] && curId.Scan <= maxScan[i] &&
curId.Charge >= minCharge[i] && curId.Charge <= maxCharge[i])
{
matchedId.Add(curId);
idFlag[curIndex] = true;
}
++curIndex;
}
var hasId = false;
if (matchedId.Any())
{
++numFeaturesWithId;
hasId = true;
}
// Find MS2 scans
// var numMs2Scans = 0;
var tags = new List <SequenceTag>();
var hasMs2 = false;
for (var scanNum = minScan[i]; scanNum <= maxScan[i]; scanNum++)
{
var isolationWindow = run.GetIsolationWindow(scanNum);
if (isolationWindow == null)
{
continue;
}
var isolationWindowTargetMz = isolationWindow.IsolationWindowTargetMz;
var charge = (int)Math.Round(mass / isolationWindowTargetMz);
if (charge < minCharge[i] || charge > maxCharge[i])
{
continue;
}
var mz = Ion.GetIsotopeMz(mass, charge,
Averagine.GetIsotopomerEnvelope(mass).MostAbundantIsotopeIndex);
if (isolationWindow.Contains(mz))
{
// ++numMs2Scans;
tags.AddRange(tagParser.GetSequenceTags(scanNum));
hasMs2 = true;
}
}
if (hasMs2)
{
++numFeaturesWithMs2;
}
if (tags.Any())
{
++numFeaturesWithTags;
}
var protHist = new Dictionary <string, int>();
var hasMatchedTag = false;
foreach (var tag in tags)
{
var matchedProteins = searchableDb.FindAllMatchedSequenceIndices(tag.Sequence).Select(idx => fastaDb.GetProteinName(idx)).ToArray();
if (matchedProteins.Any())
{
hasMatchedTag = true;
foreach (var protein in matchedProteins)
{
int num;
if (protHist.TryGetValue(protein, out num))
{
protHist[protein] = num + 1;
}
else
{
protHist[protein] = 1;
}
}
}
}
if (hasMatchedTag)
{
++numFeaturesWithMatchingTags;
if (!hasId)
{
++numFeaturesWithNoIdAndMatchingTags;
}
}
if (protHist.Any())
{
var maxOcc = protHist.Values.Max();
if (maxOcc >= numProtMatches)
{
++numFeaturesWithTwoOrMoreMatchingTags;
}
}
}
Console.WriteLine("NumFeatures: {0}", featureParser.NumData);
Console.WriteLine("NumId: {0}", idList.Count);
Console.WriteLine("NumFeaturesWithId: {0} ({1})", numFeaturesWithId, numFeaturesWithId / (float)featureParser.NumData);
Console.WriteLine("NumFeaturesWithMs2: {0} ({1})", numFeaturesWithMs2, numFeaturesWithMs2 / (float)featureParser.NumData);
Console.WriteLine("NumFeaturesWithTag: {0} ({1})", numFeaturesWithTags, numFeaturesWithTags / (float)featureParser.NumData);
Console.WriteLine("NumFeaturesWithMatchedTag: {0} ({1})", numFeaturesWithMatchingTags, numFeaturesWithMatchingTags / (float)featureParser.NumData);
Console.WriteLine("NumFeaturesWithMoreThanOneMatchedTag: {0} ({1})", numFeaturesWithTwoOrMoreMatchingTags, numFeaturesWithTwoOrMoreMatchingTags / (float)featureParser.NumData);
Console.WriteLine("NumFeaturesWithNoIdAndMatchedTag: {0} ({1})", numFeaturesWithNoIdAndMatchingTags, numFeaturesWithNoIdAndMatchingTags / (float)featureParser.NumData);
for (var i = 0; i < idFlag.Length; i++)
{
if (!idFlag[i])
{
Console.WriteLine(idList[i].Scan);
}
}
// Console.WriteLine(string.Join(",", filter.GetMatchingMs2ScanNums(8115.973001)));
//
// Console.WriteLine(featureFileName);
}
/// <summary>
/// Process spectra and identifications.
/// </summary>
/// <param name="rawFilePath">Full file path to raw file.</param>
/// <param name="idFilePath">Full file path to identification file.</param>
/// <param name="cancellationToken">For notification of cancellation.</param>
/// <param name="progress">Progress reporter.</param>
/// <returns>List of processed IDs.</returns>
public List <ProcessedResult> Process(string rawFilePath, string idFilePath, CancellationToken cancellationToken, IProgress <ProgressData> progress = null)
{
// Set up progress reporter
progress = progress ?? new Progress <ProgressData>();
var progressData = new ProgressData(progress);
// Show initial loading message
progressData.Report(0.1, "Loading...");
// Read mzid file
var mzidReader = new SimpleMZIdentMLReader();
var identifications = mzidReader.Read(idFilePath, cancellationToken);
// Check to make sure raw and MZID file match.
var rawFileName = Path.GetFileNameWithoutExtension(rawFilePath);
var spectrumFileFromId = Path.GetFileNameWithoutExtension(identifications.SpectrumFile);
var dtaIndex = spectrumFileFromId.LastIndexOf("_dta");
if (dtaIndex >= 0)
{
spectrumFileFromId = spectrumFileFromId.Substring(0, dtaIndex);
}
if (rawFileName != spectrumFileFromId)
{
throw new ArgumentException($"Mismatch between spectrum file ({rawFileName}) and id file ({spectrumFileFromId}).");
}
// Group IDs into a hash by scan number
var idMap = identifications.Identifications.GroupBy(id => id.ScanNum).ToDictionary(scan => scan.Key, ids => ids);
var processedResults = new ConcurrentBag <ProcessedResult>();
// Load raw file
using (var lcms = MassSpecDataReaderFactory.GetMassSpecDataReader(rawFilePath))
{
int count = 0;
Parallel.ForEach(
lcms.ReadAllSpectra(),
spectrum =>
{
if (cancellationToken.IsCancellationRequested)
{ // Cancel if necessary
return;
}
// Report completion percentage and current scan number
if (count % (int)Math.Max(0.01 * lcms.NumSpectra, 1) == 0)
{
progressData.Report(count, lcms.NumSpectra, $"{Math.Round(100.0*count / lcms.NumSpectra)}%");
}
Interlocked.Increment(ref count);
// Skip spectrum if it isn't MS2
var productSpectrum = spectrum as ProductSpectrum;
if (productSpectrum == null || !idMap.ContainsKey(spectrum.ScanNum))
{
return;
}
var specResults = idMap[spectrum.ScanNum];
var results = from specResult in specResults
let sequence = specResult.Peptide.GetIpSequence()
let coverage = this.CalculateSequenceCoverage(productSpectrum, sequence, specResult.Charge)
select new ProcessedResult
{
ScanNum = spectrum.ScanNum,
Sequence = sequence,
Charge = specResult.Charge,
PrecursorMz = specResult.CalMz,
DeNovoScore = specResult.DeNovoScore,
SpecEValue = specResult.SpecEv,
EValue = specResult.EValue,
QValue = specResult.QValue,
PepQValue = specResult.PepQValue,
FragMethod = productSpectrum.ActivationMethod,
IsotopeError = specResult.IsoError,
SequenceCoverage = Math.Round(coverage),
};
foreach (var result in results)
{
processedResults.Add(result);
}
});
}
// Sort spectra by SpecEValue
return(processedResults.OrderBy(pr => pr.SpecEValue).ToList());
}
public bool Validate()
{
// Spec file path validation
if (string.IsNullOrWhiteSpace(SpecFilePath))
{
PrintError("Missing parameter for spectrum file path");
return(false);
}
// Check for folder-type datasets, and replace specFilePath with the directory name if it is.
SpecFilePath = MassSpecDataReaderFactory.GetDatasetName(SpecFilePath);
var isDirectoryDataset = MassSpecDataReaderFactory.IsADirectoryDataset(SpecFilePath);
// True if specFilePath is a directory that is NOT a supported folder-type dataset.
var specPathIsDirectory = Directory.Exists(SpecFilePath) && !isDirectoryDataset;
if (!File.Exists(SpecFilePath) && !specPathIsDirectory && !isDirectoryDataset)
{
PrintError("File not found: " + SpecFilePath);
return(false);
}
var types = MassSpecDataReaderFactory.MassSpecDataTypeFilterList;
if (!specPathIsDirectory && !(types.Select(ext => SpecFilePath.ToLower().EndsWith(ext)).Any()))
{
PrintError("Invalid file extension for spectrum file: (" + Path.GetExtension(SpecFilePath) + ") " + SpecFilePath);
return(false);
}
// TODO: Handle non-.raw files in the subfolder
SpecFilePaths = Directory.Exists(SpecFilePath) && !MassSpecDataReaderFactory.IsADirectoryDataset(SpecFilePath) ? Directory.GetFiles(SpecFilePath, "*.raw") : new[] { SpecFilePath };
// Database path validation
if (string.IsNullOrWhiteSpace(DatabaseFilePath))
{
PrintError("Missing parameter for database file path");
return(false);
}
if (!File.Exists(DatabaseFilePath))
{
PrintError("File not found: " + DatabaseFilePath);
return(false);
}
var dbExtension = Path.GetExtension(DatabaseFilePath).ToLower();
if (!dbExtension.Equals(".fa") &&
!dbExtension.Equals(".fasta"))
{
PrintError("Invalid extension for the database file path (" + dbExtension + ")");
return(false);
}
// Output directory validation
if (string.IsNullOrWhiteSpace(OutputDir))
{
// Must use "Path.GetFullPath" to return the absolute path when the source file is in the working directory
// But, it could cause problems with too-long paths.
OutputDir = specPathIsDirectory ? SpecFilePath : Path.GetDirectoryName(Path.GetFullPath(SpecFilePath));
}
if (string.IsNullOrWhiteSpace(OutputDir))
{
PrintError("Invalid output file directory: " + OutputDir);
return(false);
}
if (!Directory.Exists(OutputDir))
{
if (File.Exists(OutputDir) && !File.GetAttributes(OutputDir).HasFlag(FileAttributes.Directory))
{
PrintError("OutputDir \"" + OutputDir + "\" is not a directory!");
return(false);
}
Directory.CreateDirectory(OutputDir);
}
// Mods file validation
if (!string.IsNullOrWhiteSpace(ModsFilePath) && !File.Exists(ModsFilePath))
{
PrintError("Modifications file not found: " + ModsFilePath);
return(false);
}
try
{
var errorMessage = LoadModsFile(ModsFilePath);
if (!string.IsNullOrWhiteSpace(errorMessage))
{
PrintError(errorMessage);
return(false);
}
}
catch (Exception ex)
{
PrintError("Exception parsing the file for parameter -mod: " + ex.Message);
return(false);
}
// Scans file validation
if (!string.IsNullOrWhiteSpace(ScansFilePath) && !File.Exists(ScansFilePath))
{
PrintError("Scans File file not found: " + ScansFilePath);
return(false);
}
try
{
var errorMessage = LoadScansFile(ScansFilePath);
if (!string.IsNullOrWhiteSpace(errorMessage))
{
PrintError(errorMessage);
return(false);
}
}
catch (Exception ex)
{
PrintError("Exception parsing the file for parameter -scansFile: " + ex.Message);
return(false);
}
// Feature file validation
if (!string.IsNullOrWhiteSpace(FeatureFilePath) && !File.Exists(FeatureFilePath))
{
PrintError("Feature File not found: " + FeatureFilePath);
return(false);
}
if (!string.IsNullOrWhiteSpace(FeatureFilePath) &&
!Path.GetExtension(FeatureFilePath).ToLower().Equals(".csv") &&
!Path.GetExtension(FeatureFilePath).ToLower().Equals(".ms1ft") &&
!Path.GetExtension(FeatureFilePath).ToLower().Equals(".msalign"))
{
PrintError("Invalid extension for the Feature file path (" + Path.GetExtension(FeatureFilePath) + ")");
return(false);
}
// MinX/MaxX validation
if (MinSequenceLength > MaxSequenceLength)
{
PrintError("MinPrecursorCharge (" + MinPrecursorIonCharge + ") is larger than MaxPrecursorCharge (" + MaxPrecursorIonCharge + ")!");
return(false);
}
if (MinProductIonCharge > MaxProductIonCharge)
{
PrintError("MinFragmentCharge (" + MinProductIonCharge + ") is larger than MaxFragmentCharge (" + MaxProductIonCharge + ")!");
return(false);
}
if (MinSequenceMass > MaxSequenceMass)
{
PrintError("MinSequenceMassInDa (" + MinSequenceMass + ") is larger than MaxSequenceMassInDa (" + MaxSequenceMass + ")!");
return(false);
}
MaxNumThreads = GetOptimalMaxThreads(MaxNumThreads);
return(true);
}
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);
}
public void Write()
{
foreach (var specFilePath in SpecFilePaths)
{
var outputFilePath = Path.Combine(OutputDir, Path.GetFileNameWithoutExtension(specFilePath) + ParameterFileExtension);
using (var writer = new StreamWriter(outputFilePath))
{
writer.WriteLine("SpecFile\t" + Path.GetFileName(specFilePath));
writer.WriteLine("DatabaseFile\t" + Path.GetFileName(DatabaseFilePath));
writer.WriteLine("FeatureFile\t{0}", FeatureFilePath != null ? Path.GetFileName(FeatureFilePath) : Path.GetFileName(MassSpecDataReaderFactory.ChangeExtension(specFilePath, ".ms1ft")));
writer.WriteLine("SearchMode\t" + SearchModeInt);
writer.WriteLine("Tag-based search\t" + TagBasedSearch);
writer.WriteLine("Tda\t" + (TdaBool == null ? "Decoy" : (bool)TdaBool ? "Target+Decoy" : "Target"));
writer.WriteLine("PrecursorIonTolerancePpm\t" + PrecursorIonTolerancePpm);
writer.WriteLine("ProductIonTolerancePpm\t" + ProductIonTolerancePpm);
writer.WriteLine("MinSequenceLength\t" + MinSequenceLength);
writer.WriteLine("MaxSequenceLength\t" + MaxSequenceLength);
writer.WriteLine("MinPrecursorIonCharge\t" + MinPrecursorIonCharge);
writer.WriteLine("MaxPrecursorIonCharge\t" + MaxPrecursorIonCharge);
writer.WriteLine("MinProductIonCharge\t" + MinProductIonCharge);
writer.WriteLine("MaxProductIonCharge\t" + MaxProductIonCharge);
writer.WriteLine("MinSequenceMass\t" + MinSequenceMass);
writer.WriteLine("MaxSequenceMass\t" + MaxSequenceMass);
writer.WriteLine("MaxDynamicModificationsPerSequence\t" + _maxNumDynModsPerSequence);
foreach (var searchMod in _searchModifications)
{
writer.WriteLine("Modification\t" + searchMod);
}
}
}
}
public void Write()
{
var outputFilePath = Path.Combine(OutputDir, Path.GetFileNameWithoutExtension(SpecFilePath) + ParameterFileExtension);
using (var writer = new StreamWriter(outputFilePath))
{
writer.WriteLine("SpecFile\t" + Path.GetFileName(SpecFilePath));
writer.WriteLine("DatabaseFile\t" + Path.GetFileName(DatabaseFilePath));
writer.WriteLine("FeatureFile\t{0}", !string.IsNullOrWhiteSpace(FeatureFilePath) ? Path.GetFileName(FeatureFilePath) : Path.GetFileName(MassSpecDataReaderFactory.ChangeExtension(SpecFilePath, ".ms1ft")));
writer.WriteLine("InternalCleavageMode\t" + InternalCleavageMode);
writer.WriteLine("Tag-based search\t" + TagBasedSearch);
writer.WriteLine("Tda\t" + (TargetDecoySearchMode == DatabaseSearchMode.Both ? "Target+Decoy" : TargetDecoySearchMode.ToString()));
writer.WriteLine("PrecursorIonTolerancePpm\t" + PrecursorIonTolerancePpm);
writer.WriteLine("ProductIonTolerancePpm\t" + ProductIonTolerancePpm);
writer.WriteLine("MinSequenceLength\t" + MinSequenceLength);
writer.WriteLine("MaxSequenceLength\t" + MaxSequenceLength);
writer.WriteLine("MinPrecursorIonCharge\t" + MinPrecursorIonCharge);
writer.WriteLine("MaxPrecursorIonCharge\t" + MaxPrecursorIonCharge);
writer.WriteLine("MinProductIonCharge\t" + MinProductIonCharge);
writer.WriteLine("MaxProductIonCharge\t" + MaxProductIonCharge);
writer.WriteLine("MinSequenceMass\t" + MinSequenceMass);
writer.WriteLine("MaxSequenceMass\t" + MaxSequenceMass);
writer.WriteLine("MaxDynamicModificationsPerSequence\t" + MaxDynamicModificationsPerSequence);
foreach (var searchMod in Modifications)
{
writer.WriteLine("Modification\t" + searchMod);
}
}
}
public void TestReadingPbfFile()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
var pbfFilePath = Path.Combine(Utils.DEFAULT_TEST_FILE_FOLDER, @"TopDown\ProductionQCShew\QC_Shew_13_04_A_17Feb14_Samwise_13-07-28.pbf");
if (!File.Exists(pbfFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, pbfFilePath);
}
var pbfRun = new PbfLcMsRun(pbfFilePath);
var checksum = pbfRun.PbfFileChecksum;
var specFilePath = Path.ChangeExtension(pbfFilePath, "raw");
if (!File.Exists(specFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, specFilePath);
}
Console.WriteLine(@"Loading .pbf into memory");
//var run = InMemoryLcMsRun.GetLcMsRun(specFilePath);
var run = new InMemoryLcMsRun(MassSpecDataReaderFactory.GetMassSpecDataReader(specFilePath), 0, 0);
Console.WriteLine(@"Comparing spectra between .pbf and in-memory spectra");
// spectrum comparison
//for (var scanNum = run.MinLcScan; scanNum <= run.MaxLcScan; scanNum++)
foreach (var scanNum in run.AllScanNumbers)
{
var spec1 = run.GetSpectrum(scanNum);
var spec2 = pbfRun.GetSpectrum(scanNum);
Assert.IsTrue(spec1.Peaks.Length == spec2.Peaks.Length);
for (var i = 0; i < spec1.Peaks.Length; i++)
{
var p1 = spec1.Peaks[i];
var p2 = spec2.Peaks[i];
Assert.True(p1.Equals(p2));
Assert.True(Math.Abs(p1.Mz - p2.Mz) < 1e-8);
Assert.True(Math.Abs(p1.Intensity - p2.Intensity) < 0.001);
}
}
Console.WriteLine(@"Comparing XICs");
// chromatogram comparison
const double targetMz = 655.01;
var tolerance = new Tolerance(10);
var xic1 = run.GetFullPrecursorIonExtractedIonChromatogram(targetMz, tolerance);
var xic2 = pbfRun.GetFullPrecursorIonExtractedIonChromatogram(targetMz, tolerance);
Assert.True(xic1.Count == xic2.Count);
for (var i = 0; i < xic1.Count; i++)
{
if (!xic1[i].Equals(xic2[i]))
{
Console.WriteLine(@"{0} {1} {2}", i, xic1[i], xic2[i]);
}
Assert.True(xic1[i].Equals(xic2[i]));
}
Console.WriteLine(@"Done");
}
public void TestTagMatching()
{
var methodName = MethodBase.GetCurrentMethod().Name;
TestUtils.ShowStarting(methodName);
// Parse sequence tags
const string dataSet = @"H:\Research\QCShew_TopDown\Production\QC_Shew_Intact_26Sep14_Bane_C2Column3";
const int minTagLength = 8;
var tagFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".seqtag");
if (!File.Exists(tagFileName))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, tagFileName);
}
var tagParser = new SequenceTagParser(tagFileName, minTagLength);
// Parse raw file
var rawFileName = MassSpecDataReaderFactory.ChangeExtension(dataSet, ".raw");
if (!File.Exists(rawFileName))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, rawFileName);
}
var run = PbfLcMsRun.GetLcMsRun(rawFileName);
// Parse ID file
const string resultFilePath = @"H:\Research\QCShew_TopDown\Production\M1_V092\QC_Shew_Intact_26Sep14_Bane_C2Column3_IcTda.tsv";
if (!File.Exists(resultFilePath))
{
Assert.Ignore(@"Skipping test {0} since file not found: {1}", methodName, resultFilePath);
}
var resultParser = new MsPathFinderParser(resultFilePath);
const double qValueThreshold = 0.01;
var idList = resultParser.GetIdWithQValuesNoLargerThan(qValueThreshold);
var idFlag = new bool[run.MaxLcScan + 1];
foreach (var id in idList)
{
idFlag[id.Scan] = true;
}
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);
}
// const string fastaFilePath = @"H:\Research\QCShew_TopDown\Production\ID_002216_235ACCEA.icsfldecoy.fasta";
var fastaDb = new FastaDatabase(fastaFilePath);
var searchableDb = new SearchableDatabase(fastaDb);
var numMs2Spectra = 0;
var numSpectraWithTag = 0;
var numSpectraWithMatchingTag = 0;
var numSpectraWithMatchedTagNoId = 0;
foreach (var ms2ScanNum in run.GetScanNumbers(2))
{
++numMs2Spectra;
var tags = tagParser.GetSequenceTags(ms2ScanNum);
if (tags != null)
{
++numSpectraWithTag;
foreach (var tag in tags)
{
if (searchableDb.Search(tag.Sequence) >= 0)
{
//Console.WriteLine(tag.Sequence);
++numSpectraWithMatchingTag;
if (!idFlag[ms2ScanNum])
{
++numSpectraWithMatchedTagNoId;
}
break;
}
}
}
}
Console.WriteLine("Tag length: {0}", minTagLength);
Console.WriteLine("NumMs2Spectra: {0}", numMs2Spectra);
Console.WriteLine("NumMs2SpectraWithTags: {0} ({1})", numSpectraWithTag, numSpectraWithTag / (float)numMs2Spectra);
Console.WriteLine("NumMs2SpectraWithMatchingTags: {0} ({1})", numSpectraWithMatchingTag, numSpectraWithMatchingTag / (float)numMs2Spectra);
Console.WriteLine("NumMs2SpectraWithMatchingTagsWithNoId: {0} ({1})", numSpectraWithMatchedTagNoId, numSpectraWithMatchedTagNoId / (float)numMs2Spectra);
}
public static int Main(string[] args)
{
string specFilePath;
string outputDir;
try
{
var handle = Process.GetCurrentProcess().MainWindowHandle;
SetConsoleMode(handle, EnableExtendedFlags);
if (args.Length == 0)
{
PrintUsageInfo();
return(-1);
}
var paramDic = new Dictionary <string, string>
{
{ "-s", null },
{ "-o", null }
};
for (var i = 0; i < args.Length / 2; i++)
{
var key = args[2 * i];
var value = args[2 * i + 1];
if (!paramDic.ContainsKey(key))
{
PrintUsageInfo("Invalid parameter: " + key);
return(-1);
}
paramDic[key] = value;
}
// Parse command line parameters
specFilePath = paramDic["-s"];
if (specFilePath == null)
{
PrintUsageInfo("Missing required parameter -s!");
return(-1);
}
// Check for folder-type datasets, and replace specFilePath with the directory name if it is.
specFilePath = MassSpecDataReaderFactory.GetDatasetName(specFilePath);
var isDirectoryDataset = MassSpecDataReaderFactory.IsADirectoryDataset(specFilePath);
// True if specFilePath is a directory that is NOT a supported folder-type dataset.
var specPathIsDirectory = Directory.Exists(specFilePath) && !isDirectoryDataset;
if (!File.Exists(specFilePath) && !specPathIsDirectory && !isDirectoryDataset)
{
PrintUsageInfo("File not found: " + specFilePath);
return(-1);
}
var types = MassSpecDataReaderFactory.MassSpecDataTypeFilterList;
types.Remove(".pbf");
if (!specPathIsDirectory && !(types.Select(ext => specFilePath.ToLower().EndsWith(ext)).Any()))
{
PrintUsageInfo("Invalid file extension: (" + Path.GetExtension(specFilePath) + ") " + specFilePath);
return(-1);
}
// Must use "Path.GetFullPath" to return the absolute path when the source file is in the working directory
// But, it could cause problems with too-long paths.
outputDir = paramDic["-o"] ?? (specPathIsDirectory ? specFilePath : Path.GetDirectoryName(Path.GetFullPath(specFilePath)));
if (outputDir == null)
{
PrintUsageInfo("Invalid output file directory: " + specFilePath);
return(-1);
}
if (!Directory.Exists(outputDir))
{
if (File.Exists(outputDir) && !File.GetAttributes(outputDir).HasFlag(FileAttributes.Directory))
{
PrintUsageInfo("OutputDir " + outputDir + " is not a directory!");
return(-1);
}
Directory.CreateDirectory(outputDir);
}
}
catch (Exception ex)
{
Console.WriteLine("Exception while parsing the command line parameters: " + ex.Message);
return(-5);
}
#if (!DEBUG)
try
{
#endif
string[] specFilePaths = new[] { specFilePath };
if (Directory.Exists(specFilePath) && !MassSpecDataReaderFactory.IsADirectoryDataset(specFilePath))
{
specFilePaths = Directory.GetFiles(specFilePath, "*.raw");
}
foreach (var rawFilePath in specFilePaths)
{
var pbfFilePath = Path.Combine(outputDir, Path.GetFileNameWithoutExtension(rawFilePath) + PbfLcMsRun.FileExtension);
bool isCurrent;
if (File.Exists(pbfFilePath) && PbfLcMsRun.CheckFileFormatVersion(pbfFilePath, out isCurrent) && isCurrent)
{
Console.WriteLine("{0} already exists.", pbfFilePath);
continue;
}
Console.WriteLine("Creating {0} from {1}", pbfFilePath, rawFilePath);
IMassSpecDataReader reader = MassSpecDataReaderFactory.GetMassSpecDataReader(rawFilePath);
var progress = new Progress <ProgressData>(p =>
{
p.UpdateFrequencySeconds = 2;
if ((p.Percent % 25).Equals(0) || p.ShouldUpdate())
{
Console.Write("\r{0}, {1:00.0}% complete ", p.Status, p.Percent);
}
});
var run = new PbfLcMsRun(rawFilePath, reader, pbfFilePath, 0, 0, progress);
Console.WriteLine();
}
Console.WriteLine("PbfFormatVersion: {0}", PbfLcMsRun.FileFormatId);
return(0);
#if (!DEBUG)
}
catch (Exception ex)
{
// NOTE: The DMS Analysis Manager looks for this text; do not change it
Console.WriteLine("Exception while processing: " + ex.Message);
Console.WriteLine(ex.StackTrace);
var errorCode = -Math.Abs(ex.Message.GetHashCode());
if (errorCode == 0)
{
return(-1);
}
else
{
return(errorCode);
}
}
#endif
}
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;
}
}
/// <summary>
/// Find features in the data file
/// </summary>
/// <param name="rawFile">Data file (either a pbf file or a file type from which a pbf file can be auto-created)</param>
/// <returns>0 if success; negative number on error</returns>
private int ProcessFile(string rawFile)
{
var outDirectory = GetOutputDirectory(rawFile);
if (string.IsNullOrEmpty(outDirectory))
{
return(-1);
}
var baseName = Path.GetFileName(MassSpecDataReaderFactory.RemoveExtension(rawFile));
var ms1FeaturesFilePath = Path.Combine(outDirectory, baseName + "." + FileExtension);
var outCsvFilePath = Path.Combine(outDirectory, baseName + "_" + FileExtension + ".csv");
var pngFilePath = Path.Combine(outDirectory, baseName + "_" + FileExtension + ".png");
if (File.Exists(ms1FeaturesFilePath))
{
ShowErrorMessage("ProMex output already exists: " + ms1FeaturesFilePath);
return(-2);
}
if (!File.Exists(rawFile))
{
ShowErrorMessage("Cannot find input file: " + rawFile);
return(-3);
}
var stopwatch = Stopwatch.StartNew();
Console.WriteLine("Start loading MS1 data from {0}", rawFile);
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var featureFinder = new LcMsPeakMatrix(run, _likelihoodScorer, 1, 60, Parameters.MaxThreads);
Console.WriteLine("Complete loading MS1 data. Elapsed Time = {0:0.000} sec", (stopwatch.ElapsedMilliseconds) / 1000.0d);
if (run.GetMs1ScanVector().Length == 0)
{
ShowErrorMessage(@"Data file has no MS1 spectra: " + Path.GetFileName(rawFile));
return(-4);
}
if (featureFinder.Ms1PeakCount == 0)
{
ShowErrorMessage(@"Data file has no MS1 peaks: " + Path.GetFileName(rawFile));
return(-5);
}
var comparer = featureFinder.Comparer;
var container = new LcMsFeatureContainer(featureFinder.Ms1Spectra, _likelihoodScorer, new LcMsFeatureMergeComparer(new Tolerance(10)));
var minSearchMassBin = comparer.GetBinNumber(Parameters.MinSearchMass);
var maxSearchMassBin = comparer.GetBinNumber(Parameters.MaxSearchMass);
double totalMassBin = maxSearchMassBin - minSearchMassBin + 1;
Console.WriteLine("Start MS1 feature extraction.");
stopwatch.Restart();
for (var binNum = minSearchMassBin; binNum <= maxSearchMassBin; binNum++)
{
var clusters = featureFinder.FindFeatures(binNum);
container.Add(clusters);
if (binNum > minSearchMassBin && (binNum - minSearchMassBin) % 1000 == 0)
{
var elapsed = (stopwatch.ElapsedMilliseconds) / 1000.0d;
var processedBins = binNum - minSearchMassBin;
var processedPercentage = processedBins / totalMassBin * 100;
Console.WriteLine("Processing {0:0.0}% of mass bins ({1:0.0} Da); elapsed time = {2:0.000} sec; # of features = {3}",
processedPercentage, featureFinder.Comparer.GetMzEnd(binNum), elapsed,
container.NumberOfFeatures);
}
}
Console.WriteLine("Complete MS1 feature extraction.");
Console.WriteLine(" - Elapsed time = {0:0.000} sec", (stopwatch.ElapsedMilliseconds) / 1000.0d);
Console.WriteLine(" - Number of extracted features = {0}", container.NumberOfFeatures);
Console.WriteLine("Start selecting mutually independent features from feature network graph");
stopwatch.Restart();
var featureId = FilterAndOutputFeatures(container, featureFinder, outCsvFilePath, ms1FeaturesFilePath);
Console.WriteLine("Complete feature filtration");
Console.WriteLine(" - Elapsed time = {0:0.000} sec", (stopwatch.ElapsedMilliseconds) / 1000.0d);
Console.WriteLine(" - Number of filtered features = {0}", featureId);
Console.WriteLine(" - ProMex output: {0}", ms1FeaturesFilePath);
if (Parameters.CsvOutput)
{
Console.WriteLine(" - ProMex output in ICR2LS format: {0}", outCsvFilePath);
}
if (Parameters.FeatureMapImage)
{
CreateFeatureMapImage(run, ms1FeaturesFilePath, pngFilePath);
}
return(0);
}
