/// <summary>
/// Creates a spectral comparer based on the comparison type.
/// </summary>
/// <param name="comparisonType"></param>
/// <returns></returns>
public static ISpectralComparer CreateSpectraComparer(SpectralComparison comparisonType, double percent = .4)
{
ISpectralComparer comparer = null;
switch (comparisonType)
{
case SpectralComparison.DotProduct:
comparer = new SpectralDotProductComprarer(percent);
break;
case SpectralComparison.NormalizedDotProduct:
comparer = new SpectralNormalizedDotProductComparer(percent);
break;
case SpectralComparison.BinaryDotProduct:
comparer = new BinaryDotProduct(percent);
break;
case SpectralComparison.CosineDotProduct:
comparer = new SpectralCosineComparer();
break;
case SpectralComparison.PeakCounts:
comparer = new SpectralPeakCountComparer();
break;
case SpectralComparison.SteinDotProduct:
comparer = new SteinDotProduct();
break;
default:
break;
}
return(comparer);
}
public void Create(string directory, SpectralComparison comparerType, double percent)
{
var comparer = SpectralComparerFactory.CreateSpectraComparer(comparerType, percent);
var spectra = new List <MSSpectra>();
var files = Directory.GetFiles(directory, "*.txt");
foreach (var file in files)
{
spectra.Add(ReadSpectrum(file));
}
var values = new List <double>();
for (var i = 0; i < spectra.Count; i++)
{
//for (int j = i + 1; j < spectra.Count; j++)
{
//double value = comparer.CompareSpectra(spectra[i], spectra[j]);
//values.Add(value);
Console.WriteLine("{0}", Path.GetFileNameWithoutExtension(files[i]));
// Path.GetFileNameWithoutExtension(files[j]));
DisplaySpectra(spectra[i], .15, newTitle: "MS/MS Spectra - ");
}
}
var baseName = Path.GetFileName(directory);
// Console.WriteLine("{0} - {1}", baseName, values.Average());
// values.ForEach(x => Console.WriteLine("\t{0}", x));
}
public void Create(string directory, SpectralComparison comparerType, double percent)
{
var comparer = SpectralComparerFactory.CreateSpectraComparer(comparerType, percent);
var spectra = new List<MSSpectra>();
var files = Directory.GetFiles(directory, "*.txt");
foreach (var file in files)
{
spectra.Add(ReadSpectrum(file));
}
var values = new List<double>();
for (var i = 0; i < spectra.Count; i++)
{
//for (int j = i + 1; j < spectra.Count; j++)
{
//double value = comparer.CompareSpectra(spectra[i], spectra[j]);
//values.Add(value);
Console.WriteLine("{0}", Path.GetFileNameWithoutExtension(files[i]));
// Path.GetFileNameWithoutExtension(files[j]));
DisplaySpectra(spectra[i], .15, newTitle: "MS/MS Spectra - ");
}
}
var baseName = Path.GetFileName(directory);
// Console.WriteLine("{0} - {1}", baseName, values.Average());
// values.ForEach(x => Console.WriteLine("\t{0}", x));
}
public void TestSpectralSimilarityScore(
string pathX, int scanX,
string pathY, int scanY,
SpectralComparison comparerType,
SpectraFilters filterType,
double percent,
double mzTolerance)
{
// Convert relative paths to absolute paths
pathX = GetPath(pathX);
pathY = GetPath(pathY);
var spectrumX = GetSpectrum(pathX, scanX);
var spectrumY = GetSpectrum(pathY, scanY);
var comparer = SpectralComparerFactory.CreateSpectraComparer(comparerType, percent);
var filter = SpectrumFilterFactory.CreateFilter(filterType);
spectrumX.Peaks = filter.Threshold(spectrumX.Peaks, percent);
spectrumY.Peaks = filter.Threshold(spectrumY.Peaks, percent);
spectrumX.Peaks = XYData.Bin(spectrumX.Peaks,
0,
2000,
mzTolerance);
spectrumY.Peaks = XYData.Bin(spectrumY.Peaks,
0,
2000,
mzTolerance);
var value = comparer.CompareSpectra(spectrumX, spectrumY);
var path = Path.GetDirectoryName(pathX);
var plotTitle = string.Format("comparison-{2}-{3}-{0}-{1}_{4:0.000}", scanX, scanY, comparerType, percent,
value);
var pathCompareImage = Path.Combine(path, plotTitle + ".png");
DisplayComparisonPlot(spectrumX, spectrumY, mzTolerance, pathCompareImage, plotTitle);
}
protected static SpectralAnalysis MatchDatasets(SpectralComparison comparerType,
ISpectraProvider readerX,
ISpectraProvider readerY,
SpectralOptions options,
AlignmentDataset datasetX,
AlignmentDataset datasetY,
List <string> names)
{
var peptideReader = PeptideReaderFactory.CreateReader(SequenceFileType.MSGF);
var finder = new SpectralAnchorPointFinder();
var validator = new SpectralAnchorPointValidator();
var comparer = SpectralComparerFactory.CreateSpectraComparer(comparerType);
var filter = SpectrumFilterFactory.CreateFilter(SpectraFilters.TopPercent);
var matches = finder.FindAnchorPoints(readerX,
readerY,
comparer,
filter,
options);
var peptidesX = peptideReader.Read(datasetX.PeptideFile);
var peptidesY = peptideReader.Read(datasetY.PeptideFile);
validator.ValidateMatches(matches,
peptidesX,
peptidesY,
options);
var analysis = new SpectralAnalysis
{
DatasetNames = names,
Matches = matches,
Options = options
};
return(analysis);
}
public void TestSpectralSimilarityScore(
string pathX, int scanX,
string pathY, int scanY,
SpectralComparison comparerType,
SpectraFilters filterType,
double percent,
double mzTolerance)
{
// Convert relative paths to absolute paths
pathX = GetPath(pathX);
pathY = GetPath(pathY);
var spectrumX = GetSpectrum(pathX, scanX);
var spectrumY = GetSpectrum(pathY, scanY);
var comparer = SpectralComparerFactory.CreateSpectraComparer(comparerType, percent);
var filter = SpectrumFilterFactory.CreateFilter(filterType);
spectrumX.Peaks = filter.Threshold(spectrumX.Peaks, percent);
spectrumY.Peaks = filter.Threshold(spectrumY.Peaks, percent);
spectrumX.Peaks = XYData.Bin(spectrumX.Peaks,
0,
2000,
mzTolerance);
spectrumY.Peaks = XYData.Bin(spectrumY.Peaks,
0,
2000,
mzTolerance);
var value = comparer.CompareSpectra(spectrumX, spectrumY);
var path = Path.GetDirectoryName(pathX);
var plotTitle = string.Format("comparison-{2}-{3}-{0}-{1}_{4:0.000}", scanX, scanY, comparerType, percent,
value);
var pathCompareImage = Path.Combine(path, plotTitle + ".png");
DisplayComparisonPlot(spectrumX, spectrumY, mzTolerance, pathCompareImage, plotTitle);
}
public void GenerateFigure4_MetaMatches(string directory,
SpectralComparison comparerType,
double mzBinSize,
double mzTolerance,
double netTolerance,
double similarityScoreCutoff,
double peptideScore,
double peptideFdr,
double ionPercent,
int numberOfRequiredPeaks,
string name)
{
AlignmentAnalysisWriterFactory.BasePath = @"M:\doc\papers\paperAlignment\Data\figure4";
Console.WriteLine(@"Post-Pre Tests For {0}", directory);
var cacheFiles = Directory.GetFiles(directory, "*.mscache");
Console.WriteLine(@"Building data cache");
var data = cacheFiles.Select(path => new FigureBase.PathCache {
Cache = path
}).ToList();
// The options for the analysis
var options = new SpectralOptions
{
MzBinSize = mzBinSize,
MzTolerance = mzTolerance,
NetTolerance = netTolerance,
SimilarityCutoff = similarityScoreCutoff,
TopIonPercent = ionPercent,
IdScore = peptideScore,
ComparerType = comparerType,
Fdr = peptideFdr,
RequiredPeakCount = numberOfRequiredPeaks
};
var comparison = 0;
for (var i = 0; i < data.Count; i++)
{
var cachex = data[i];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathX = ScanSummaryCache.ReadPath(cachex.Cache);
var datasetX = new AlignmentDataset(rawPathX, "", cachex.Msgf);
// create a raw file reader for the datasets
using (var readerX = new InformedProteomicsReader())
{
// wrap it in the cached object so we can load scan meta-data
var cacheReaderX = new RawLoaderCache(readerX);
var cacheDataX = ScanSummaryCache.ReadCache(cachex.Cache);
readerX.AddDataFile(rawPathX, 0);
cacheReaderX.AddCache(0, cacheDataX);
for (var j = i + 1; j < data.Count; j++)
{
var cachey = data[j];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathY = ScanSummaryCache.ReadPath(cachey.Cache);
var datasetY = new AlignmentDataset(rawPathY, "", cachey.Msgf);
// create a raw file reader for the datasets
using (var readerY = new InformedProteomicsReader())
{
// Then the writer for creating a report
var writer =
AlignmentAnalysisWriterFactory.Create(AlignmentFigureType.Figure3, name + comparison);
comparison++;
// wrap it in the cached object so we can load scan meta-data
var cacheReaderY = new RawLoaderCache(readerY);
var cacheDataY = ScanSummaryCache.ReadCache(cachey.Cache);
cacheReaderY.AddCache(0, cacheDataY);
readerY.AddDataFile(rawPathY, 0);
var names = new List <string> {
data[i].Cache, data[j].Cache
};
var analysis = MatchDatasets(comparerType,
readerX,
readerY,
options,
datasetX,
datasetY,
names);
AlignMatches(analysis, writer);
writer.Close();
}
}
}
}
}
public void GenerateFigure3_Matches(string directory,
SpectralComparison comparerType,
double mzBinSize,
double mzTolerance,
double netTolerance,
double similarityScoreCutoff,
double peptideScore,
double peptideFdr,
double ionPercent,
int numberOfRequiredPeaks)
{
AlignmentAnalysisWriterFactory.BasePath = @"M:\doc\papers\paperAlignment\Data\figure4";
Console.WriteLine(@"Post-Pre Tests For {0}", directory);
var cacheFiles = Directory.GetFiles(directory, "*.mscache");
var msgfFiles = Directory.GetFiles(directory, "*_msgfdb_fht.txt");
Console.WriteLine(@"Building data cache");
var map = cacheFiles.ToDictionary<string, string, FigureBase.PathCache>(path => path.ToLower(), path => null);
var data = (from path in msgfFiles
let name = path.ToLower().Replace("_msgfdb_fht.txt", ".mscache")
let newName = Path.Combine(directory, name)
let features = Path.Combine(directory, name)
where map.ContainsKey(newName)
select new FigureBase.PathCache { Cache = newName, Msgf = path, Features = features }).ToList();
// The options for the analysis
var options = new SpectralOptions
{
MzBinSize = mzBinSize,
MzTolerance = mzTolerance,
NetTolerance = netTolerance,
SimilarityCutoff = similarityScoreCutoff,
TopIonPercent = ionPercent,
IdScore = peptideScore,
ComparerType = comparerType,
Fdr = peptideFdr,
RequiredPeakCount = numberOfRequiredPeaks
};
Console.WriteLine(@"{0}", data.Count);
var comparison = 0;
for (var i = 0; i < data.Count; i++)
{
var cachex = data[i];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathX = ScanSummaryCache.ReadPath(cachex.Cache);
var datasetX = new AlignmentDataset(rawPathX, "", cachex.Msgf);
// create a raw file reader for the datasets
using (var readerX = RawLoaderFactory.CreateFileReader(datasetX.RawFile))
{
// wrap it in the cached object so we can load scan meta-data
var cacheReaderX = new RawLoaderCache(readerX);
var cacheDataX = ScanSummaryCache.ReadCache(cachex.Cache);
readerX.AddDataFile(rawPathX, 0);
cacheReaderX.AddCache(0, cacheDataX);
for (var j = i + 1; j < data.Count; j++)
{
// Then the writer for creating a report
var writer =
AlignmentAnalysisWriterFactory.Create(AlignmentFigureType.Figure3,
"results-figure3-largeScale" + comparison);
comparison++;
var cachey = data[j];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathY = ScanSummaryCache.ReadPath(cachey.Cache);
var datasetY = new AlignmentDataset(rawPathY, "", cachey.Msgf);
// create a raw file reader for the datasets
using (var readerY = RawLoaderFactory.CreateFileReader(datasetY.RawFile))
{
// wrap it in the cached object so we can load scan meta-data
var cacheReaderY = new RawLoaderCache(readerY);
var cacheDataY = ScanSummaryCache.ReadCache(cachey.Cache);
cacheReaderY.AddCache(0, cacheDataY);
readerY.AddDataFile(rawPathY, 0);
var names = new List<string> { data[i].Cache, data[j].Cache };
// Write the results
var analysis = MatchDatasets(comparerType,
cacheReaderX,
cacheReaderY,
options,
datasetX,
datasetY,
names);
AlignMatches(analysis, writer);
}
}
}
}
}
protected static SpectralAnalysis MatchDatasets(SpectralComparison comparerType,
ISpectraProvider readerX,
ISpectraProvider readerY,
SpectralOptions options,
AlignmentDataset datasetX,
AlignmentDataset datasetY,
List<string> names)
{
var peptideReader = PeptideReaderFactory.CreateReader(SequenceFileType.MSGF);
var finder = new SpectralAnchorPointFinder();
var validator = new SpectralAnchorPointValidator();
var comparer = SpectralComparerFactory.CreateSpectraComparer(comparerType);
var filter = SpectrumFilterFactory.CreateFilter(SpectraFilters.TopPercent);
var matches = finder.FindAnchorPoints(readerX,
readerY,
comparer,
filter,
options);
var peptidesX = peptideReader.Read(datasetX.PeptideFile);
var peptidesY = peptideReader.Read(datasetY.PeptideFile);
validator.ValidateMatches(matches,
peptidesX,
peptidesY,
options);
var analysis = new SpectralAnalysis
{
DatasetNames = names,
Matches = matches,
Options = options
};
return analysis;
}
public void GenerateFigure3_Matches(string directory,
SpectralComparison comparerType,
double mzBinSize,
double mzTolerance,
double netTolerance,
double similarityScoreCutoff,
double peptideScore,
double peptideFdr,
double ionPercent,
int numberOfRequiredPeaks)
{
AlignmentAnalysisWriterFactory.BasePath = @"M:\doc\papers\paperAlignment\Data\figure4";
Console.WriteLine(@"Post-Pre Tests For {0}", directory);
var cacheFiles = Directory.GetFiles(directory, "*.mscache");
var msgfFiles = Directory.GetFiles(directory, "*_msgfdb_fht.txt");
Console.WriteLine(@"Building data cache");
var map = cacheFiles.ToDictionary <string, string, FigureBase.PathCache>(path => path.ToLower(), path => null);
var data = (from path in msgfFiles
let name = path.ToLower().Replace("_msgfdb_fht.txt", ".mscache")
let newName = Path.Combine(directory, name)
let features = Path.Combine(directory, name)
where map.ContainsKey(newName)
select new FigureBase.PathCache {
Cache = newName, Msgf = path, Features = features
}).ToList();
// The options for the analysis
var options = new SpectralOptions
{
MzBinSize = mzBinSize,
MzTolerance = mzTolerance,
NetTolerance = netTolerance,
SimilarityCutoff = similarityScoreCutoff,
TopIonPercent = ionPercent,
IdScore = peptideScore,
ComparerType = comparerType,
Fdr = peptideFdr,
RequiredPeakCount = numberOfRequiredPeaks
};
Console.WriteLine(@"{0}", data.Count);
var comparison = 0;
for (var i = 0; i < data.Count; i++)
{
var cachex = data[i];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathX = ScanSummaryCache.ReadPath(cachex.Cache);
var datasetX = new AlignmentDataset(rawPathX, "", cachex.Msgf);
// create a raw file reader for the datasets
using (var readerX = new InformedProteomicsReader())
{
// wrap it in the cached object so we can load scan meta-data
var cacheReaderX = new RawLoaderCache(readerX);
var cacheDataX = ScanSummaryCache.ReadCache(cachex.Cache);
readerX.AddDataFile(rawPathX, 0);
cacheReaderX.AddCache(0, cacheDataX);
for (var j = i + 1; j < data.Count; j++)
{
// Then the writer for creating a report
var writer =
AlignmentAnalysisWriterFactory.Create(AlignmentFigureType.Figure3,
"results-figure3-largeScale" + comparison);
comparison++;
var cachey = data[j];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathY = ScanSummaryCache.ReadPath(cachey.Cache);
var datasetY = new AlignmentDataset(rawPathY, "", cachey.Msgf);
// create a raw file reader for the datasets
using (var readerY = new InformedProteomicsReader())
{
// wrap it in the cached object so we can load scan meta-data
var cacheReaderY = new RawLoaderCache(readerY);
var cacheDataY = ScanSummaryCache.ReadCache(cachey.Cache);
cacheReaderY.AddCache(0, cacheDataY);
readerY.AddDataFile(rawPathY, 0);
var names = new List <string> {
data[i].Cache, data[j].Cache
};
// Write the results
var analysis = MatchDatasets(comparerType,
cacheReaderX,
cacheReaderY,
options,
datasetX,
datasetY,
names);
AlignMatches(analysis, writer);
}
}
}
}
}
public void GenerateFigure4_MetaMatches(string directory,
SpectralComparison comparerType,
double mzBinSize,
double mzTolerance,
double netTolerance,
double similarityScoreCutoff,
double peptideScore,
double peptideFdr,
double ionPercent,
int numberOfRequiredPeaks,
string name)
{
AlignmentAnalysisWriterFactory.BasePath = @"M:\doc\papers\paperAlignment\Data\figure4";
Console.WriteLine(@"Post-Pre Tests For {0}", directory);
var cacheFiles = Directory.GetFiles(directory, "*.mscache");
Console.WriteLine(@"Building data cache");
var data = cacheFiles.Select(path => new FigureBase.PathCache { Cache = path }).ToList();
// The options for the analysis
var options = new SpectralOptions
{
MzBinSize = mzBinSize,
MzTolerance = mzTolerance,
NetTolerance = netTolerance,
SimilarityCutoff = similarityScoreCutoff,
TopIonPercent = ionPercent,
IdScore = peptideScore,
ComparerType = comparerType,
Fdr = peptideFdr,
RequiredPeakCount = numberOfRequiredPeaks
};
var comparison = 0;
for (var i = 0; i < data.Count; i++)
{
var cachex = data[i];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathX = ScanSummaryCache.ReadPath(cachex.Cache);
var datasetX = new AlignmentDataset(rawPathX, "", cachex.Msgf);
// create a raw file reader for the datasets
using (var readerX = RawLoaderFactory.CreateFileReader(datasetX.RawFile))
{
// wrap it in the cached object so we can load scan meta-data
var cacheReaderX = new RawLoaderCache(readerX);
var cacheDataX = ScanSummaryCache.ReadCache(cachex.Cache);
readerX.AddDataFile(rawPathX, 0);
cacheReaderX.AddCache(0, cacheDataX);
for (var j = i + 1; j < data.Count; j++)
{
var cachey = data[j];
// Get the raw path stored in the cache file...
// then get the dataset object
var rawPathY = ScanSummaryCache.ReadPath(cachey.Cache);
var datasetY = new AlignmentDataset(rawPathY, "", cachey.Msgf);
// create a raw file reader for the datasets
using (var readerY = RawLoaderFactory.CreateFileReader(datasetY.RawFile))
{
// Then the writer for creating a report
var writer =
AlignmentAnalysisWriterFactory.Create(AlignmentFigureType.Figure3, name + comparison);
comparison++;
// wrap it in the cached object so we can load scan meta-data
var cacheReaderY = new RawLoaderCache(readerY);
var cacheDataY = ScanSummaryCache.ReadCache(cachey.Cache);
cacheReaderY.AddCache(0, cacheDataY);
readerY.AddDataFile(rawPathY, 0);
var names = new List<string> { data[i].Cache, data[j].Cache };
var analysis = MatchDatasets(comparerType,
readerX,
readerY,
options,
datasetX,
datasetY,
names);
AlignMatches(analysis, writer);
writer.Close();
}
}
}
}
}
