public void TestFeatureAlignment()
{
const string outFilePath = @"\\protoapps\UserData\Jungkap\Lewy\aligned\promex_crosstab_temp.tsv";
//CPTAC_Intact_CR32A_24Aug15_Bane_15-02-06-RZ
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(10);
var alignment = new LcMsFeatureAlignment(new AnalysisCompRef.CompRefFeatureComparer(tolerance));
for (var i = 0; i < NdataSet; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", PbfPath, GetDataSetNames(i));
var mspFile = string.Format(@"{0}\{1}_IcTda.tsv", MsPfFolder, GetDataSetNames(i));
var mspFile2 = string.Format(@"{0}\{1}_IcTda.tsv", MsPfFolder2, GetDataSetNames(i));
var ms1FtFile = string.Format(@"{0}\{1}.ms1ft", Ms1FtFolder, GetDataSetNames(i));
Console.WriteLine(rawFile);
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var prsmList1 = prsmReader.LoadIdentificationResult(mspFile, ProteinSpectrumMatch.SearchTool.MsPathFinder);
var prsmList2 = prsmReader.LoadIdentificationResult(mspFile2, ProteinSpectrumMatch.SearchTool.MsPathFinder);
prsmList1.AddRange(prsmList2);
var prsmList = MergePrsm(prsmList1);
var features = LcMsFeatureAlignment.LoadProMexResult(i, ms1FtFile, run);
for (var j = 0; j < prsmList.Count; j++)
{
var match = prsmList[j];
match.ProteinId = match.ProteinName;
}
// tag features by PrSMs
for (var j = 0; j < features.Count; j++)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsTh(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);
}
}
}
alignment.AddDataSet(i, features, run);
}
alignment.AlignFeatures();
Console.WriteLine("{0} alignments ", alignment.CountAlignedFeatures);
for (var i = 0; i < NdataSet; i++)
{
alignment.FillMissingFeatures(i);
Console.WriteLine("{0} has been processed", GetDataSetNames(i));
}
OutputCrossTabWithId(outFilePath, alignment);
}
public void TestFeatureAlignment()
{
const string outFilePath = @"\\protoapps\UserData\Jungkap\CompRef\aligned\promex_crosstab_temp.tsv";
var runLabels = new[] { "32A", "32B", "32C", "32D", "32E", "32F", "32G", "33A", "33B", "33C", "33D", "33E", "33F", "33G" };
var nDataset = runLabels.Length;
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(10);
var alignment = new LcMsFeatureAlignment(new CompRefFeatureComparer(tolerance));
for (var i = 0; i < nDataset; i++)
{
var rawFile = string.Format(@"{0}\CPTAC_Intact_CR{1}_24Aug15_Bane_15-02-06-RZ.pbf", RawFolder, runLabels[i]);
var mspFile = string.Format(@"{0}\CPTAC_Intact_CR{1}_24Aug15_Bane_15-02-06-RZ_IcTda.tsv", MsPfFolder, runLabels[i]);
var ms1FtFile = string.Format(@"{0}\CPTAC_Intact_CR{1}_24Aug15_Bane_15-02-06-RZ.ms1ft", Ms1FtFolder, runLabels[i]);
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var features = LcMsFeatureAlignment.LoadProMexResult(i, ms1FtFile, run);
if (File.Exists(mspFile))
{
var prsmList = prsmReader.LoadIdentificationResult(mspFile, ProteinSpectrumMatch.SearchTool.MsPathFinder);
for (var j = 0; j < prsmList.Count; j++)
{
var match = prsmList[j];
match.ProteinId = match.ProteinName;
}
// 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);
}
}
}
}
alignment.AddDataSet(i, features, run);
}
alignment.AlignFeatures();
Console.WriteLine("{0} alignments ", alignment.CountAlignedFeatures);
for (var i = 0; i < nDataset; i++)
{
alignment.FillMissingFeatures(i);
Console.WriteLine("{0} has been processed", runLabels[i]);
}
OutputCrossTabWithId(outFilePath, alignment, runLabels);
}
private void AlignFeatures(List <string> datasets, string mspfFolder, string ms1ftFolder, string outFilePath)
{
var nDataset = datasets.Count;
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(12);
var alignment = new LcMsFeatureAlignment(new AnalysisCompRef.CompRefFeatureComparer(tolerance));
for (var i = 0; i < nDataset; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", PbfPath, datasets[i]);
var mspFile = string.Format(@"{0}\{1}_IcTda.tsv", mspfFolder, datasets[i]);
var ms1FtFile = string.Format(@"{0}\{1}.ms1ft", ms1ftFolder, datasets[i]);
var ms1FtFile2 = string.Format(@"{0}\{1}.seqtag.ms1ft", ms1ftFolder, datasets[i]);
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var features = LcMsFeatureAlignment.LoadProMexResult(i, ms1FtFile, run);
var features2 = LcMsFeatureAlignment.LoadProMexResult(i, ms1FtFile2, run);
features.AddRange(features2);
if (File.Exists(mspFile))
{
var prsmList = prsmReader.LoadIdentificationResult(mspFile, ProteinSpectrumMatch.SearchTool.MsPathFinder);
//var prsmFeatureMatch = new bool[prsmList.Count];
foreach (var match in prsmList)
{
match.ProteinId = match.ProteinName;
}
// tag features by PrSMs
foreach (var feature in features)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsMz(feature.Mass);
foreach (var match in prsmList)
{
if (feature.MinScanNum < match.ScanNum && match.ScanNum < feature.MaxScanNum && Math.Abs(feature.Mass - match.Mass) < massTol)
{
feature.ProteinSpectrumMatches.Add(match);
//prsmFeatureMatch[k] = true;
}
}
}
}
alignment.AddDataSet(i, features, run);
}
alignment.AlignFeatures();
Console.WriteLine("{0} alignments ", alignment.CountAlignedFeatures);
for (var i = 0; i < nDataset; i++)
{
alignment.FillMissingFeatures(i);
Console.WriteLine("{0} has been processed", datasets[i]);
}
AnalysisCompRef.OutputCrossTabWithId(outFilePath, alignment, datasets);
}
public void TestFeatureAlignment()
{
const string outFilePath = @"\\protoapps\UserData\Jungkap\Quant\aligned\promex_crosstab.tsv";
//const string outFolder = @"\\protoapps\UserData\Jungkap\CompRef\aligned";
var runLabels = new string[] { "1x1", "1x2", "1x3", "1x4", "1x5", "5x1", "5x2", "5x3", "5x4", "5x5", "10x1", "10x2", "10x3", "10x4", "10x5", };
var nDataset = runLabels.Length;
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(10);
var alignment = new LcMsFeatureAlignment(new SpikeInFeatureComparer(tolerance));
for (var i = 0; i < nDataset; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", RawFolder, datasets[i]);
var mspFile = string.Format(@"{0}\{1}_IcTda.tsv", MsPfFolder, datasets[i]);
var ms1FtFile = string.Format(@"{0}\{1}.ms1ft", Ms1FtFolder, datasets[i]);
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var prsmList = prsmReader.LoadIdentificationResult(mspFile, ProteinSpectrumMatch.SearchTool.MsPathFinder);
var features = LcMsFeatureAlignment.LoadProMexResult(i, ms1FtFile, run);
for (var j = 0; j < prsmList.Count; j++)
{
var match = prsmList[j];
match.ProteinId = match.ProteinName;
}
// tag features by PrSMs
for (var j = 0; j < features.Count; j++)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsTh(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);
}
}
}
alignment.AddDataSet(i, features, run);
}
alignment.AlignFeatures();
Console.WriteLine("{0} alignments ", alignment.CountAlignedFeatures);
/*
* for (var i = 0; i < nDataset; i++)
* {
* alignment.FillMissingFeatures(i);
* Console.WriteLine("{0} has been processed", runLabels[i]);
* }
*/
OutputCrossTabWithId(outFilePath, alignment, runLabels);
}
public static ICollection <ProteinSpectrumMatchSet> CollectTrainSet(string pbfFilePath, string idFilePath)
{
Modification.RegisterAndGetModification(Modification.Cysteinyl.Name, Modification.Cysteinyl.Composition);
var prsmReader = new ProteinSpectrumMatchReader(0.01);
var prsmList = prsmReader.LoadIdentificationResult(idFilePath);
var run = PbfLcMsRun.GetLcMsRun(pbfFilePath);
var groupedPrsmList = GroupingByPrsm(0, prsmList, new PrsmComparer(run));
var finalPrsmGroups = new List <ProteinSpectrumMatchSet>();
foreach (var prsmSet in groupedPrsmList)
{
if (prsmSet.Count < 2)
{
continue;
}
var isGood = false;
var sequence = prsmSet[0].GetSequence();
if (sequence == null)
{
continue;
}
foreach (var scan in prsmSet.Select(prsm => prsm.ScanNum))
{
var spectrum = run.GetSpectrum(scan) as ProductSpectrum;
if (spectrum == null)
{
continue;
}
if (IsGoodTarget(spectrum, sequence))
{
isGood = true;
break;
}
}
if (isGood)
{
finalPrsmGroups.Add(prsmSet);
}
}
return(finalPrsmGroups);
}
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();
}
}
public void FindMissingLcMsFeatures()
{
var mspfFolder = @"D:\MassSpecFiles\CompRef_Kelleher\Study3";
var ms1ftFolder = @"D:\MassSpecFiles\CompRef_Kelleher\Study3";
const int Nfraction1 = 3;
const int Nfraction2 = 5;
for (var frac1 = 1; frac1 <= Nfraction1; frac1++)
{
for (var frac2 = 1; frac2 <= Nfraction2; frac2++)
{
var datasets = GetDataSetNamesStudy3(frac1, frac2);
//var outFilePath = string.Format(@"D:\MassSpecFiles\CompRef_Kelleher\study3_GFrep{0}_Gfrac{1}.tsv", frac1.ToString("D2"), frac2.ToString("D2"));
var nDataset = datasets.Count;
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(12);
for (var i = 0; i < nDataset; i++)
{
var rawFile = string.Format(@"{0}\{1}.pbf", PbfPath, datasets[i]);
var mspFile = string.Format(@"{0}\{1}_IcTda.tsv", mspfFolder, datasets[i]);
var ms1FtFile = string.Format(@"{0}\{1}.ms1ft", ms1ftFolder, datasets[i]);
var outPath = string.Format(@"{0}\{1}.seqtag.ms1ft", ms1ftFolder, datasets[i]);
if (File.Exists(outPath))
{
continue;
}
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var features = LcMsFeatureAlignment.LoadProMexResult(i, ms1FtFile, run);
var prsmList = prsmReader.LoadIdentificationResult(mspFile, ProteinSpectrumMatch.SearchTool.MsPathFinder);
var prsmFeatureMatch = new bool[prsmList.Count];
for (var j = 0; j < features.Count; j++)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsTh(features[j].Mass);
for (var k = 0; k < prsmList.Count; k++)
{
var match = prsmList[k];
if (features[j].MinScanNum < match.ScanNum && match.ScanNum < features[j].MaxScanNum && Math.Abs(features[j].Mass - match.Mass) < massTol)
{
features[j].ProteinSpectrumMatches.Add(match);
prsmFeatureMatch[k] = true;
}
}
}
var missingPrsm = new List <ProteinSpectrumMatch>();
for (var k = 0; k < prsmList.Count; k++)
{
if (!prsmFeatureMatch[k])
{
missingPrsm.Add(prsmList[k]);
}
}
FeatureFind(missingPrsm, run, outPath);
Console.WriteLine(outPath);
}
}
}
}
public void TestIMERFeatureAlignment()
{
const string outFilePath = @"D:\MassSpecFiles\IMER\promex_crosstab.tsv";
const string rawFolder = @"D:\MassSpecFiles\IMER";
var runLabels = new string[] { "1", "2", "3", "4", "5", "6" };
var nDataset = runLabels.Length;
//CPTAC_Intact_CR32A_24Aug15_Bane_15-02-06-RZ
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(10);
var alignment = new LcMsFeatureAlignment(new CompRefFeatureComparer(tolerance));
for (var i = 0; i < nDataset; i++)
{
var k = runLabels[i].Equals("2") || runLabels[i].Equals("3") ? 14 : 13;
var rawFile = string.Format(@"{0}\Diabetes_iPSC_Beta_{1}_IMER_{2}May14_Alder_14-01-33.pbf", rawFolder, runLabels[i], k);
var mspFile = string.Format(@"{0}\Diabetes_iPSC_Beta_{1}_IMER_{2}May14_Alder_14-01-33_msgfdb_syn.txt", rawFolder, runLabels[i], k);
var ms1FtFile = string.Format(@"{0}\Diabetes_iPSC_Beta_{1}_IMER_{2}May14_Alder_14-01-33.ms1ft", rawFolder, runLabels[i], k);
Console.WriteLine(rawFile);
Console.WriteLine(File.Exists(rawFile));
var run = PbfLcMsRun.GetLcMsRun(rawFile);
var features = LcMsFeatureAlignment.LoadProMexResult(i, ms1FtFile, run, 500, 15000);
if (File.Exists(mspFile))
{
var prsmList = prsmReader.LoadIdentificationResult(mspFile, ProteinSpectrumMatch.SearchTool.MsGfPlus);
for (var j = 0; j < prsmList.Count; j++)
{
var match = prsmList[j];
match.ProteinId = match.ProteinName;
}
// tag features by PrSMs
for (var j = 0; j < features.Count; j++)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsTh(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);
}
}
}
}
alignment.AddDataSet(i, features, run);
}
alignment.AlignFeatures();
Console.WriteLine("{0} alignments ", alignment.CountAlignedFeatures);
for (var i = 0; i < nDataset; i++)
{
alignment.FillMissingFeatures(i);
Console.WriteLine("{0} has been processed", runLabels[i]);
}
OutputCrossTabWithId(outFilePath, alignment, runLabels);
}
public static void Main(string[] args)
{
// Parse file
var inputFilePath = args[0];
var datasets = DatasetInfo.ParseDatasetInfoFile(inputFilePath);
var fileName = Path.GetFileNameWithoutExtension(inputFilePath);
var directory = Path.GetDirectoryName(inputFilePath);
var outputfilePath = Path.Combine(directory, string.Format("{0}_crosstab.tsv", fileName));
int nDataset = datasets.Count;
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(100);
var alignment = new LcMsFeatureAlignment(new CompRefFeatureComparer(tolerance));
int dataId = 0;
foreach (var dataset in datasets)
{
var run = PbfLcMsRun.GetLcMsRun(dataset.RawFilePath, 0, 0);
var features = LcMsFeatureAlignment.LoadProMexResult(dataId++, dataset.Ms1FtFilePath, run);
if (File.Exists(dataset.MsPfIdFilePath))
{
var prsmList = prsmReader.LoadIdentificationResult(dataset.MsPfIdFilePath, ProteinSpectrumMatch.SearchTool.MsPathFinder);
foreach (var match in prsmList)
{
match.ProteinId = match.ProteinName;
}
// tag features by PrSMs
foreach (LcMsFeature feature in features)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsMz(feature.Mass);
foreach (var match in prsmList)
{
if (feature.MinScanNum < match.ScanNum && match.ScanNum < feature.MaxScanNum && Math.Abs(feature.Mass - match.Mass) < massTol)
{
feature.ProteinSpectrumMatches.Add(match);
}
}
}
}
alignment.AddDataSet(dataId, features, run);
}
alignment.AlignFeatures();
Console.WriteLine("{0} alignments ", alignment.CountAlignedFeatures);
for (var i = 0; i < nDataset; i++)
{
alignment.FillMissingFeatures(i);
Console.WriteLine("{0} has been processed", datasets[i].Label);
}
OutputCrossTabWithId(outputfilePath, alignment, datasets.Select(ds => ds.Label).ToArray());
}
public static void Main(string[] args)
{
if (args.Length == 0)
{
ShowSyntax();
return;
}
// Parse file
var inputFilePath = args[0];
if (!File.Exists(inputFilePath))
{
ConsoleMsgUtils.ShowError("File not found: " + inputFilePath);
return;
}
var datasets = DatasetInfo.ParseDatasetInfoFile(inputFilePath);
if (datasets.Count == 0)
{
ConsoleMsgUtils.ShowError("No valid data found in the dataset info file");
ShowSyntax();
return;
}
var fileName = Path.GetFileNameWithoutExtension(inputFilePath);
var directory = Path.GetDirectoryName(inputFilePath);
var crosstabFilename = string.Format("{0}_crosstab.tsv", fileName);
string outputfilePath;
if (string.IsNullOrWhiteSpace(directory))
{
outputfilePath = crosstabFilename;
}
else
{
outputfilePath = Path.Combine(directory, crosstabFilename);
}
var nDataset = datasets.Count;
var prsmReader = new ProteinSpectrumMatchReader();
var tolerance = new Tolerance(100);
var alignment = new LcMsFeatureAlignment(new CompRefFeatureComparer(tolerance));
var dataId = 0;
foreach (var dataset in datasets)
{
if (!File.Exists(dataset.RawFilePath))
{
ConsoleMsgUtils.ShowError("Instrument file not found: " + dataset.RawFilePath);
continue;
}
if (!File.Exists(dataset.Ms1FtFilePath))
{
ConsoleMsgUtils.ShowError("ProMex results file not found: " + dataset.Ms1FtFilePath);
continue;
}
Console.WriteLine("Opening " + dataset.RawFilePath);
var run = PbfLcMsRun.GetLcMsRun(dataset.RawFilePath, 0, 0);
Console.WriteLine("Opening " + dataset.Ms1FtFilePath);
var features = LcMsFeatureAlignment.LoadProMexResult(dataId++, dataset.Ms1FtFilePath, run);
if (!string.IsNullOrWhiteSpace(dataset.MsPfIdFilePath) && File.Exists(dataset.MsPfIdFilePath))
{
Console.WriteLine("Opening " + dataset.MsPfIdFilePath);
var prsmList = prsmReader.LoadIdentificationResult(dataset.MsPfIdFilePath, ProteinSpectrumMatch.SearchTool.MsPathFinder);
foreach (var match in prsmList)
{
match.ProteinId = match.ProteinName;
}
// tag features by PrSMs
foreach (var feature in features)
{
//features[j].ProteinSpectrumMatches = new ProteinSpectrumMatchSet(i);
var massTol = tolerance.GetToleranceAsMz(feature.Mass);
foreach (var match in prsmList)
{
if (feature.MinScanNum < match.ScanNum && match.ScanNum < feature.MaxScanNum && Math.Abs(feature.Mass - match.Mass) < massTol)
{
feature.ProteinSpectrumMatches.Add(match);
}
}
}
}
alignment.AddDataSet(dataId, features, run);
}
alignment.AlignFeatures();
Console.WriteLine("{0} alignments ", alignment.CountAlignedFeatures);
var validResults = 0;
for (var datasetIndex = 0; datasetIndex < nDataset; datasetIndex++)
{
if (datasetIndex >= alignment.CountDatasets)
{
ConsoleMsgUtils.ShowWarning(string.Format("Could not align {0}; features not found", datasets[datasetIndex].Label));
continue;
}
alignment.FillMissingFeatures(datasetIndex);
Console.WriteLine("{0} has been processed", datasets[datasetIndex].Label);
validResults++;
}
if (validResults > 0)
{
OutputCrossTabWithId(outputfilePath, alignment, datasets.Select(ds => ds.Label).ToArray());
}
}
public void TestGenerateFrequencyData()
{
var methodName = MethodBase.GetCurrentMethod().Name;
Utils.ShowStarting(methodName);
const string idFileFolder = @"D:\MassSpecFiles\training\IdScoring\MSPF_trainset";
const string outFileFolder = @"D:\MassSpecFiles\training\IdScoring";
if (!Directory.Exists(idFileFolder))
{
Assert.Ignore(@"Skipping test {0} since folder not found: {1}", methodName, idFileFolder);
}
Modification.RegisterAndGetModification(Modification.Cysteinyl.Name, Modification.Cysteinyl.Composition);
Modification.RegisterAndGetModification(Modification.Phosphorylation.Name, Modification.Phosphorylation.Composition);
Modification.RegisterAndGetModification(Modification.Methylation.Name, Modification.Methylation.Composition);
Modification.RegisterAndGetModification(Modification.DiMethylation.Name, Modification.DiMethylation.Composition);
Modification.RegisterAndGetModification(Modification.TriMethylation.Name, Modification.TriMethylation.Composition);
Modification.RegisterAndGetModification("Trioxidation", new Composition(0, 0, 0, 3, 0));
// var aaSet = new AminoAcidSet(@"D:\MassSpecFiles\training\Mods.txt");
foreach (var dataset in TrainSetFileLists)
{
var dataname = Path.GetFileNameWithoutExtension(dataset);
var idFile = string.Format(@"{0}\{1}_IcTda.tsv", idFileFolder, dataname);
var decoyFile = string.Format(@"{0}\{1}_IcDecoy.tsv", idFileFolder, dataname);
// var targetFile = string.Format(@"{0}\{1}_IcTarget.tsv", idFileFolder, dataname);
if (!File.Exists(idFile))
{
continue;
}
var prsmReader = new ProteinSpectrumMatchReader(0.01);
var prsmList = prsmReader.LoadIdentificationResult(idFile);
var minScore = prsmList.Last().Score;
var decoyMatches = prsmReader.ReadMsPathFinderResult(decoyFile, int.MaxValue, 1, Math.Max(minScore - 5, 10));
var run = PbfLcMsRun.GetLcMsRun(dataset);
var spectrumMatchSet = LcMsFeatureTrain.CollectTrainSet(dataset, idFile);
Console.WriteLine(spectrumMatchSet.Count);
var writer = new StreamWriter(string.Format(@"{0}\{1}_target.tsv", outFileFolder, dataname));
foreach (var matches in spectrumMatchSet)
{
foreach (var match in matches)
{
var spec = run.GetSpectrum(match.ScanNum) as ProductSpectrum;
GetMatchStatistics(spec, match.GetSequence(), match.Charge, writer);
}
}
writer.Close();
writer = new StreamWriter(string.Format(@"{0}\{1}_decoy.tsv", outFileFolder, dataname));
foreach (var match in decoyMatches)
{
var sequence = match.GetSequence();
var spec = run.GetSpectrum(match.ScanNum) as ProductSpectrum;
GetMatchStatistics(spec, sequence, match.Charge, writer);
}
writer.Close();
}
}
