public void FeatureFind(List <ProteinSpectrumMatch> prsms, LcMsRun run, string outTsvFilePath)
{
var featureFinder = new LcMsPeakMatrix(run, new LcMsFeatureLikelihood());
// write result files
var tsvWriter = new StreamWriter(outTsvFilePath);
tsvWriter.WriteLine(LcMsFeatureFinderLauncher.GetHeaderString(false));
var featureId = 1;
foreach (var match in prsms)
{
var minScan = run.GetPrevScanNum(match.ScanNum, 1);
var maxScan = run.GetNextScanNum(match.ScanNum, 1);
var feature = featureFinder.GetLcMsPeakCluster(match.Mass, match.Charge, minScan, maxScan);
if (feature == null)
{
continue;
}
tsvWriter.WriteLine("{0}\t{1}", featureId, LcMsFeatureFinderLauncher.GetString(feature, false));
featureId++;
}
tsvWriter.Close();
}
public void TestQuantifyIdedProteoforms()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
const string rawFolder = @"\\proto-11\MSXML_Cache\PBF_Gen_1_193\2015_2";
const string promexOutFolder = @"D:\MassSpecFiles\UTEX\MSAlign";
const string msAlignResultFolder = @"D:\MassSpecFiles\UTEX\MSAlign";
if (!Directory.Exists(rawFolder))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, rawFolder);
}
var nDataset = 32;
var dataset = new string[nDataset];
for (var i = 0; i < nDataset; i++)
{
dataset[i] = string.Format("Syn_utex2973_Top_{0,2:D2}_TopDown_7May15_Bane_14-09-01RZ", i + 1);
//var rawFile = string.Format(@"{0}\{1}.pbf", rawFolder, dataset[i]);
}
var prsmReader = new ProteinSpectrumMatchReader(0.01);
var filesProcessed = 0;
var tolerance = new Tolerance(10);
for (var i = 0; i < dataset.Length; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", rawFolder, dataset[i]);
if (!File.Exists(rawFile))
{
Console.WriteLine(@"Warning: Skipping file not found: {0}", rawFile);
continue;
}
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var path = string.Format(@"{0}\{1}_MSAlign_ResultTable.txt", msAlignResultFolder, dataset[i]);
if (!File.Exists(path))
{
Console.WriteLine(@"Warning: Skipping file not found: {0}", path);
continue;
}
var prsmList = prsmReader.LoadIdentificationResult(path, ProteinSpectrumMatch.SearchTool.MsAlign);
filesProcessed++;
for (var j = 0; j < prsmList.Count; j++)
{
var match = prsmList[j];
match.ProteinId = match.ProteinName.Substring(match.ProteinName.IndexOf(ProteinNamePrefix) + ProteinNamePrefix.Length, 5);
}
// PrSM To Feature
var prsmToFeatureIdMap = new int[prsmList.Count];
for (var k = 0; k < prsmToFeatureIdMap.Length; k++)
{
prsmToFeatureIdMap[k] = -1;
}
// Feature To PrSM
var featureToPrsm = new List <ProteinSpectrumMatchSet>();
var featureFinder = new LcMsPeakMatrix(run, new LcMsFeatureLikelihood());
var featureList = new List <LcMsPeakCluster>();
var featureId = 0;
for (var j = 0; j < prsmList.Count; j++)
{
if (prsmToFeatureIdMap[j] >= 0)
{
continue;
}
var match = prsmList[j];
var minScanNum = match.ScanNum;
var maxScanNum = match.ScanNum;
var mass = match.Mass;
var charge = match.Charge;
var massTh = tolerance.GetToleranceAsMz(mass);
var id1 = match.ProteinId;
var feature = featureFinder.GetLcMsPeakCluster(mass, charge, minScanNum, maxScanNum);
var prsmSet = new ProteinSpectrumMatchSet(i)
{
match
};
if (feature == null)
{
feature = featureFinder.GetLcMsPeaksFromNoisePeaks(mass, charge, minScanNum, maxScanNum, charge, charge);
prsmToFeatureIdMap[j] = featureId;
}
else
{
prsmToFeatureIdMap[j] = featureId;
var etTol = Math.Max(run.GetElutionTime(run.MaxLcScan) * 0.005, feature.ElutionLength * 0.2);
for (var k = j + 1; k < prsmList.Count; k++)
{
var otherMatch = prsmList[k];
var id2 = otherMatch.ProteinId;
var et2 = run.GetElutionTime(otherMatch.ScanNum);
if (id1.Equals(id2) &&
feature.MinElutionTime - etTol < et2 && et2 < feature.MaxElutionTime - etTol &&
Math.Abs(otherMatch.Mass - mass) < massTh)
{
prsmToFeatureIdMap[k] = featureId;
prsmSet.Add(otherMatch);
}
}
}
featureId++;
feature.Flag = 1;
featureList.Add(feature);
featureToPrsm.Add(prsmSet);
}
// Overlap between features???
for (var j = 0; j < featureList.Count; j++)
{
var f1 = featureList[j];
if (f1.Flag < 1)
{
continue;
}
var prsm1 = featureToPrsm[j];
for (var k = j + 1; k < featureList.Count; k++)
{
var f2 = featureList[k];
if (f2.Flag < 1)
{
continue;
}
var prsm2 = featureToPrsm[k];
if (Math.Abs(f1.Mass - f2.Mass) > tolerance.GetToleranceAsMz(f1.Mass))
{
continue;
}
if (!f1.CoElutedByNet(f2, 0.005))
{
continue;
}
if (!prsm1.ShareProteinId(prsm2))
{
continue;
}
// let us merge!!
if (f1.ScanLength > f2.ScanLength)
{
prsm1.AddRange(prsm2);
prsm2.Clear();
f2.Flag = 0;
}
else
{
prsm2.AddRange(prsm1);
prsm1.Clear();
f1.Flag = 0;
}
}
}
// now output results!!
var ms1ftFilePath = string.Format(@"{0}\{1}.ms1ft", promexOutFolder, dataset[i]);
var writer = new StreamWriter(ms1ftFilePath);
writer.WriteLine(LcMsFeatureFinderLauncher.GetHeaderString());
for (var j = 0; j < featureList.Count; j++)
{
var f1 = featureList[j];
if (f1.Flag < 1)
{
continue;
}
var prsm1 = featureToPrsm[j];
var minScanNum = run.GetPrevScanNum(prsm1.MinScanNum, 1);
var maxScanNum = run.GetNextScanNum(prsm1.MaxScanNum, 1);
f1.ExpandScanRange(minScanNum, maxScanNum);
writer.Write("{0}\t", j + 1);
writer.WriteLine(LcMsFeatureFinderLauncher.GetString(f1));
}
writer.Close();
Console.WriteLine(ms1ftFilePath);
}
if (filesProcessed == 0)
{
Assert.Ignore("Skipped since data files not found");
}
}
public void TestAlignFeatures()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
const string rawFolder = @"\\proto-11\MSXML_Cache\PBF_Gen_1_193\2015_2";
const string promexOutFolder = @"D:\MassSpecFiles\UTEX\MSAlign";
const string msAlignResultFolder = @"D:\MassSpecFiles\UTEX\MSAlign";
if (!Directory.Exists(rawFolder))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, rawFolder);
}
var nDataset = 32;
var dataset = new string[nDataset];
for (var i = 0; i < nDataset; i++)
{
dataset[i] = string.Format("Syn_utex2973_Top_{0,2:D2}_TopDown_7May15_Bane_14-09-01RZ", i + 1);
//var rawFile = string.Format(@"{0}\{1}.pbf", rawFolder, dataset[i]);
}
var tolerance = new Tolerance(10);
var ftComparer = new UtexFeatureComparer(tolerance);
var align = new LcMsFeatureAlignment(ftComparer);
var prsmReader = new ProteinSpectrumMatchReader(0.01);
var filesProcessed = 0;
for (var i = 0; i < dataset.Length; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", rawFolder, dataset[i]);
if (!File.Exists(rawFile))
{
Console.WriteLine(@"Warning: Skipping file not found: {0}", rawFile);
continue;
}
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var path = string.Format(@"{0}\{1}_MSAlign_ResultTable.txt", msAlignResultFolder, dataset[i]);
if (!File.Exists(path))
{
Console.WriteLine(@"Warning: Skipping file not found: {0}", path);
continue;
}
var ms1ftPath = string.Format(@"{0}\{1}.ms1ft", promexOutFolder, dataset[i]);
if (!File.Exists(ms1ftPath))
{
Console.WriteLine(@"Warning: Skipping file not found: {0}", ms1ftPath);
continue;
}
filesProcessed++;
//var map = new ProteinSpectrumMathMap(run, i, dataset[i]);
//map.LoadIdentificationResult(path, ProteinSpectrumMatch.SearchTool.MsAlign);
var prsmList = prsmReader.LoadIdentificationResult(path, ProteinSpectrumMatch.SearchTool.MsAlign);
for (var j = 0; j < prsmList.Count; j++)
{
var match = prsmList[j];
match.ProteinId =
match.ProteinName.Substring(
match.ProteinName.IndexOf(ProteinNamePrefix) + ProteinNamePrefix.Length, 5);
}
var features = LcMsFeatureAlignment.LoadProMexResult(i, ms1ftPath, run);
// tag features by PrSMs
for (var j = 0; j < features.Count; j++)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsMz(features[j].Mass);
foreach (var match in prsmList)
{
if (features[j].MinScanNum < match.ScanNum && match.ScanNum < features[j].MaxScanNum && Math.Abs(features[j].Mass - match.Mass) < massTol)
{
features[j].ProteinSpectrumMatches.Add(match);
}
}
}
align.AddDataSet(i, features, run);
}
if (filesProcessed == 0)
{
Assert.Ignore("Skipped since input files not found");
}
align.AlignFeatures();
Console.WriteLine("{0} alignments ", align.CountAlignedFeatures);
align.RefineAbundance();
var alignedFeatureList = align.GetAlignedFeatures();
for (var i = 0; i < nDataset; i++)
{
var ms1ftPath = string.Format(@"{0}\{1}_aligned.ms1ft", promexOutFolder, dataset[i]);
var writer = new StreamWriter(ms1ftPath);
writer.Write(LcMsFeatureFinderLauncher.GetHeaderString());
writer.WriteLine("\tIdedMs2ScanNums");
for (var j = 0; j < alignedFeatureList.Count; j++)
{
writer.Write(j + 1);
writer.Write("\t");
if (alignedFeatureList[j][i] == null)
{
for (var k = 0; k < 14; k++)
{
writer.Write("0\t");
}
writer.Write("0\n");
}
else
{
writer.Write(LcMsFeatureFinderLauncher.GetString(alignedFeatureList[j][i]));
writer.Write("\t");
if (alignedFeatureList[j][i].ProteinSpectrumMatches == null)
{
writer.Write("");
}
else
{
var scanNums = string.Join(";", alignedFeatureList[j][i].ProteinSpectrumMatches.Select(prsm => prsm.ScanNum));
writer.Write(scanNums);
}
writer.Write("\n");
}
}
writer.Close();
}
}
private void OutputAlignmentResult(LcMsFeatureAlignment align, string outFilePath, IReadOnlyList <string> rawFiles, bool isTemp = true)
{
var alignedFeatureList = align.GetAlignedFeatures();
var writer = new StreamWriter(outFilePath);
writer.Write("MonoMass\tMinElutionTime\tMaxElutionTime");
for (var i = 0; i < align.CountDatasets; i++)
{
var dataSetName = Path.GetFileNameWithoutExtension(rawFiles[i]);
writer.Write("\t{0}", dataSetName);
}
for (var i = 0; i < align.CountDatasets; i++)
{
//var dataSetName = Path.GetFileNameWithoutExtension(align.RawFileList[i]);
writer.Write("\t{0}_Score", i);
}
/*
* for (var i = 0; i < align.CountDatasets; i++)
* {
* //var dataSetName = Path.GetFileNameWithoutExtension(align.RawFileList[i]);
* writer.Write("\t{0}_Net", i);
* }*/
writer.Write("\n");
for (var i = 0; i < align.CountAlignedFeatures; i++)
{
var features = alignedFeatureList[i];
var minMaxNet = GetMinMaxNet(features);
writer.Write(@"{0}\t{1:0.00000}\t{2:0.00000}", minMaxNet.Item1, minMaxNet.Item3, minMaxNet.Item4);
for (var j = 0; j < align.CountDatasets; j++)
{
var feature = features[j];
writer.Write("\t");
writer.Write(feature?.Abundance ?? 0d);
}
for (var j = 0; j < align.CountDatasets; j++)
{
var feature = features[j];
writer.Write("\t");
writer.Write(feature?.Score ?? 0d);
}
/*
* for (var j = 0; j < align.CountDatasets; j++)
* {
* var feature = features[j];
* writer.Write("\t");
* if (feature != null) writer.Write("{0:0.00000}", feature.MinNet);
* else writer.Write(0);
* }
*
* for (var j = 0; j < align.CountDatasets; j++)
* {
* var feature = features[j];
* writer.Write("\t");
* if (feature != null) writer.Write("{0:0.00000}", feature.MaxNet);
* else writer.Write(0);
* }*/
writer.Write("\n");
}
writer.Close();
if (isTemp)
{
return;
}
var outDirectory = Path.GetDirectoryName(Path.GetFullPath(outFilePath));
for (var i = 0; i < align.CountDatasets; i++)
{
var dataSetName = Path.GetFileNameWithoutExtension(rawFiles[i]);
//writer.Write("\t{0}", dataSetName);
// now output results!!
var ms1ftFilePath = string.Format(@"{0}\{1}.aligned.ms1ft", outDirectory, dataSetName);
var writer2 = new StreamWriter(ms1ftFilePath);
writer2.WriteLine(LcMsFeatureFinderLauncher.GetHeaderString());
for (var j = 0; j < align.CountAlignedFeatures; j++)
{
var f1 = alignedFeatureList[j][i];
writer2.Write("{0}\t", j + 1);
writer2.WriteLine(LcMsFeatureFinderLauncher.GetString(f1));
}
writer2.Close();
}
}