public static MSUltraLight RichTextBox2MS(string textBoxInput, double minimumIntensity)
{
MSUltraLight myMS = new MSUltraLight();
string[] theLines = Regex.Split(textBoxInput, @"\n");
List <Tuple <float, float> > myIons = new List <Tuple <float, float> >();
foreach (string line in theLines)
{
if (Regex.IsMatch(line, "^[0-9]"))
{
string[] entries = Regex.Split(line, @" |\t");
float intensity = float.Parse(entries[1]);
if (intensity >= minimumIntensity)
{
float thisMZ = float.Parse(entries[0]);
myIons.Add(new Tuple <float, float>(thisMZ, intensity));
}
}
}
myMS.UpdateIons(myIons);
Tuple <float, short> precursor = new Tuple <float, short>(myMS.Ions.Last().Item1, 2);
myMS.Precursors = new List <Tuple <float, short> >()
{
precursor
};
return(myMS);
}
private void ShowSpectrum(int scanNumber)
{
MSUltraLight theMS = MyMS2.Find(a => a.ScanNumber == scanNumber);
List <Ion> ions = theMS.Ions.Select(a => new Ion(a.Item1, a.Item2, 0, scanNumber)).ToList();
SpectrumEye1.Plot(ions, ions[0].MZ, ions.Last().MZ, 0);
//MyMSViewer.Modifications = new List<ModificationItem>();
//MyMSViewer.PeptideSequence = "P";
//MyMSViewer.MyMS = new MSFull(theMS);
//MyMSViewer.FuncPrintMS(theMS.Ions.Min(a => a.Item1), theMS.Ions.Max(a => a.Item1));
TextBoxMSPeaks.Text = PatternTools.MSParserLight.MSUltraLightPrinter.PrintSpectrum(theMS);
}
public MSFull(MSUltraLight ms2)
{
MSData = ms2.Ions.Select(a => new Ion(a.Item1, a.Item2, ms2.CromatographyRetentionTime, ms2.ScanNumber)).ToList();
isMS2 = true;
ActivationType = ms2.ActivationType.ToString();
ChargedPrecursor = ms2.Precursors[0].Item1;
//These lines are not working for velos data
Charges = new List <int>()
{
ms2.Precursors[0].Item2
};
//---------------------------------------------------
CromatographyRetentionTime = ms2.CromatographyRetentionTime;
InstrumentType = ms2.InstrumentType.ToString();
ScanNumber = ms2.ScanNumber;
ZLines = ms2.GetZLines();
}
/// <summary>
/// Parsers an MS2 mgf file
/// </summary>
/// <param name="file"></param>
/// <returns></returns>
public static List <MSUltraLight> ParseMGFFile(string file)
{
System.IO.StreamReader sr = new System.IO.StreamReader(file);
List <MSUltraLight> theMS2 = new List <MSUltraLight>();
string line = "";
MSUltraLight ms = new MSUltraLight();
List <double> charges = new List <double>();
while ((line = sr.ReadLine()) != null)
{
//take care of the header
if (line.StartsWith("#") || line.StartsWith("_") || line.Equals(""))
{
//This is a header line
//msFile.Header += line;
}
else if (line.Equals("BEGIN IONS"))
{
//Prepare Z lines
theMS2.Add(ms);
ms = new MSUltraLight();
ms.MSLevel = 2;
ms.ActivationType = -1;
ms.InstrumentType = -1;
}
else if (line.StartsWith("CHARGE"))
{
string[] cols = Regex.Split(line, "=");
string c = cols[1].Replace(@"+", "");
if (c.Equals("2,3"))
{
charges = new List <double>()
{
2, 3
};
}
else
{
int charge = int.Parse(c);
charges = new List <double>()
{
charge
};
}
}
else if (line.StartsWith("PEPMASS"))
{
string[] cols = Regex.Split(line, "=");
string[] nums = Regex.Split(cols[1], " ");
float chargedPrecursor;
if (nums[0].Equals(""))
{
chargedPrecursor = float.Parse(nums[1]);
}
else
{
chargedPrecursor = float.Parse(nums[0]);
}
foreach (short z in charges)
{
float mh = (chargedPrecursor + (z * 1.007276466f)) / (float)z;
ms.Precursors.Add(new Tuple <float, short>(mh, z));
}
}
else if (line.StartsWith("RTINSECONDS"))
{
string[] cols = Regex.Split(line, "=");
float retTime;
if (cols[1].Contains("-"))
{
string[] nums = Regex.Split(cols[1], "-");
retTime = float.Parse(nums[0]);
}
else
{
retTime = float.Parse(cols[1]);
}
ms.CromatographyRetentionTime = retTime;
}
else if (line.StartsWith("SCANS"))
{
string[] cols = Regex.Split(line, "=");
double scan;
if (cols[1].Contains("-"))
{
string[] nums = Regex.Split(cols[1], "-");
scan = double.Parse(nums[0]);
}
else
{
scan = double.Parse(cols[1]);
}
ms.ScanNumber = (int)scan;
}
else if (line.StartsWith("TITLE"))
{
if (line.Contains("Fragmentation:hcd"))
{
ms.ActivationType = 3;
ms.InstrumentType = -1;
}
}
else if (Regex.IsMatch(line, "^[0-9]+"))
{
//If the line begins with a number it is an ion line
string[] cols = Regex.Split(line, "\t| ");
try
{
Tuple <float, float> i = new Tuple <float, float>(float.Parse(cols[0]), float.Parse(cols[1]));
ms.Ions.Add(i);
}
catch
{
throw new Exception("An inconsistency has been found in file: " + file + "\nThe line reads:\n" + line);
}
}
}
theMS2.Add(ms);
//The first one is always bogus!
if (theMS2.Count > 0)
{
theMS2.RemoveAt(0);
}
sr.Close();
return(theMS2);
}
private void buttonGo_Click(object sender, EventArgs e)
{
//Verify write permission to directory
if (!Directory.Exists(textBoxOutputDirectory.Text))
{
MessageBox.Show("Please specify a valid output directory");
return;
}
if (!Regex.IsMatch(textBoxIsobaricMasses.Text, "[0-9]+ [0-9]+"))
{
MessageBox.Show("Please fill out the masses of the isobaric tags.");
return;
}
if (!PatternTools.pTools.HasWriteAccessToFolder(textBoxOutputDirectory.Text))
{
MessageBox.Show("Please specify a valid output directory");
return;
}
//Obtain class labels
if (textBoxClassLabels.Text.Length == 0)
{
MessageBox.Show("Please input the class labels (eg., for iTRAQ 1,2,3,4");
return;
}
List <int> labels = Regex.Split(textBoxClassLabels.Text, " ").Select(a => int.Parse(a)).ToList();
//Obtain the isobaric masses
string[] im = Regex.Split(textBoxIsobaricMasses.Text, " ");
List <double> isobaricMasses = im.Select(a => double.Parse(a)).ToList();
if (labels.Count != isobaricMasses.Count)
{
MessageBox.Show("Please make sure that the class labels and isobaric masses match");
return;
}
buttonGo.Text = "Working...";
this.Update();
richTextBoxLog.Clear();
//--------------------------------------------
//Get signal from all
signalAllNormalizationDictionary = new Dictionary <string, double[]>();
//if (false)
FileInfo fi = new FileInfo(textBoxitraqSEPro.Text);
bool extractSignal = false;
ResultPackage rp = null;
if (checkBoxNormalizationChannelSignal.Checked)
{
//We should get the MS infor and merge it the the sepro package
if (fi.Extension.Equals(".sepr"))
{
rp = ResultPackage.Load(textBoxitraqSEPro.Text);
extractSignal = true;
}
List <FileInfo> rawFiles = fi.Directory.GetFiles("*.RAW").ToList();
foreach (FileInfo rawFile in rawFiles)
{
Console.WriteLine("Extracting data for " + rawFile.Name);
PatternTools.RawReader.RawReaderParams rParams = new PatternTools.RawReader.RawReaderParams();
rParams.ExtractMS1 = false;
rParams.ExtractMS2 = true;
rParams.ExtractMS3 = false;
PatternTools.RawReader.Reader reader = new PatternTools.RawReader.Reader(rParams);
List <MSLight> theMS2 = reader.GetSpectra(rawFile.FullName, new List <int>(), false);
theMS2.RemoveAll(a => a.Ions == null);
double [] totalSignal = new double[isobaricMasses.Count];
List <SQTScan> theScans = null;
//Update the sepro result package with the signal
if (extractSignal)
{
//Get all the scans from this file
string rawName = rawFile.Name.Substring(0, rawFile.Name.Length - 4);
theScans = rp.MyProteins.AllSQTScans.FindAll(a => a.FileName.Substring(0, a.FileName.Length - 4).Equals(rawName));
}
foreach (MSLight ms in theMS2)
{
double[] thisQuantitation = GetIsobaricSignal(ms.Ions, isobaricMasses);
if (extractSignal)
{
SQTScan scn = theScans.Find(a => a.ScanNumber == ms.ScanNumber);
if (scn != null)
{
scn.MSLight = ms;
scn.MSLight.Ions.RemoveAll(a => a.MZ > 400);
}
}
for (int i = 0; i < thisQuantitation.Length; i++)
{
totalSignal[i] += thisQuantitation[i];
}
}
string theName = rawFile.Name.Substring(0, rawFile.Name.Length - 3);
theName += "sqt";
signalAllNormalizationDictionary.Add(theName, totalSignal);
}
}
Console.WriteLine("Loading SEPro File");
if (!File.Exists(textBoxitraqSEPro.Text))
{
MessageBox.Show("Unable to find SEPro file");
return;
}
#region Load the spero or pepexplorer file
theScansToAnalyze = new List <SQTScan>();
List <FastaItem> theFastaItems = new List <FastaItem>();
if (fi.Extension.Equals(".sepr"))
{
Console.WriteLine("Loading SEPro file");
if (!extractSignal)
{
rp = ResultPackage.Load(textBoxitraqSEPro.Text);
}
rp.MyProteins.AllSQTScans.RemoveAll(a => a.MSLight == null);
theScansToAnalyze = rp.MyProteins.AllSQTScans;
Console.WriteLine("Done reading SEPro result");
theFastaItems = rp.MyProteins.MyProteinList.Select(a => new FastaItem(a.Locus, a.Sequence, a.Description)).ToList();
}
else if (fi.Extension.Equals(".mpex"))
{
Console.WriteLine("Loading PepExplorer file....");
PepExplorer2.Result2.ResultPckg2 result = PepExplorer2.Result2.ResultPckg2.DeserializeResultPackage(textBoxitraqSEPro.Text);
theFastaItems = result.MyFasta;
theScansToAnalyze = new List <SQTScan>();
foreach (PepExplorer2.Result2.AlignmentResult al in result.Alignments)
{
foreach (var dnr in al.DeNovoRegistries)
{
SQTScan sqt = new SQTScan();
sqt.ScanNumber = dnr.ScanNumber;
sqt.FileName = dnr.FileName;
sqt.PeptideSequence = dnr.PtmSequence;
theScansToAnalyze.Add(sqt);
}
}
//And now we need to retrieve the mass spectra. For this, the raw files should be inside the directory containing the mpex file
List <string> rawFiles = theScansToAnalyze.Select(a => a.FileName).Distinct().ToList();
for (int i = 0; i < rawFiles.Count; i++)
{
rawFiles[i] = rawFiles[i].Remove(rawFiles[i].Length - 3, 3);
rawFiles[i] = rawFiles[i] += "raw";
}
foreach (string fn in rawFiles)
{
Console.WriteLine("Retrieving spectra for file: " + fn);
ParserUltraLightRAW parser = new ParserUltraLightRAW();
string tmpFile = fn.Substring(0, fn.Length - 3);
List <SQTScan> scansForThisFile = theScansToAnalyze.FindAll(a => Regex.IsMatch(tmpFile, a.FileName.Substring(0, a.FileName.Length - 3), RegexOptions.IgnoreCase)).ToList();
List <int> scnNumbers = scansForThisFile.Select(a => a.ScanNumber).ToList();
FileInfo theInputFile = new FileInfo(textBoxitraqSEPro.Text);
List <MSUltraLight> theSpectra = parser.ParseFile(theInputFile.DirectoryName + "/" + fn, -1, 2, scnNumbers);
foreach (SQTScan sqt in scansForThisFile)
{
MSUltraLight spec = theSpectra.Find(a => a.ScanNumber == sqt.ScanNumber);
sqt.MSLight = new MSLight();
sqt.MSLight.MZ = spec.Ions.Select(a => (double)a.Item1).ToList();
sqt.MSLight.Intensity = spec.Ions.Select(a => (double)a.Item2).ToList();
}
Console.WriteLine("\tDone processing this file.");
}
}
else
{
throw new Exception("This file format is not supported.");
}
#endregion
//Obtaining multiplexed spectra
SEProQ.IsobaricQuant.YadaMultiplexCorrection.YMC ymc = null;
if (textBoxCorrectedYadaDirectory.Text.Length > 0)
{
Console.WriteLine("Reading Yada results");
ymc = new IsobaricQuant.YadaMultiplexCorrection.YMC(new DirectoryInfo(textBoxCorrectedYadaDirectory.Text));
Console.WriteLine("Done loading Yada results");
}
//Remove multiplexed spectra from sepro results
if (textBoxCorrectedYadaDirectory.Text.Length > 0)
{
int removedCounter = 0;
foreach (KeyValuePair <string, List <int> > kvp in ymc.fileNameScanNumberMultiplexDictionary)
{
Console.WriteLine("Removing multiplexed spectra for file :: " + kvp.Key);
richTextBoxLog.AppendText("Removing multiplexed spectra for file :: " + kvp.Key + "\n");
string cleanName = kvp.Key.Substring(0, kvp.Key.Length - 4);
cleanName += ".sqt";
foreach (int scnNo in kvp.Value)
{
int index = theScansToAnalyze.FindIndex(a => a.ScanNumber == scnNo && a.FileName.Equals(cleanName));
if (index >= 0)
{
Console.Write(theScansToAnalyze[index].ScanNumber + " ");
richTextBoxLog.AppendText(theScansToAnalyze[index].ScanNumber + " ");
removedCounter++;
theScansToAnalyze.RemoveAt(index);
}
}
Console.WriteLine("\n");
richTextBoxLog.AppendText("\n");
}
Console.WriteLine("Done removing multiplexed spectra :: " + removedCounter);
}
PatternTools.CSML.Matrix correctionMatrix = new PatternTools.CSML.Matrix();
if (checkBoxApplyPurityCorrection.Checked)
{
List <List <double> > correctionData = GetPurityCorrectionsFromForm();
correctionMatrix = IsobaricQuant.IsobaricImpurityCorrection.GenerateInverseCorrectionMatrix(correctionData);
}
//--------------------------------------------------------------------------------------------------------------------
//Prepare normalization Dictionary
signalIdentifiedNormalizationDictionary = new Dictionary <string, double[]>();
List <string> fileNames = theScansToAnalyze.Select(a => a.FileName).Distinct().ToList();
foreach (string fileName in fileNames)
{
signalIdentifiedNormalizationDictionary.Add(fileName, new double[isobaricMasses.Count]);
}
//-------------------------------------
//If necessary, correct for impurity and feed global signal dictionary
foreach (SQTScan scn in theScansToAnalyze)
{
double[] thisQuantitation = GetIsobaricSignal(scn.MSLight.Ions, isobaricMasses);
double maxSignal = thisQuantitation.Max();
//We can only correct for signal for those that have quantitation values in all places
if (checkBoxApplyPurityCorrection.Checked && (thisQuantitation.Count(a => a > maxSignal * (double)numericUpDownIonCountThreshold.Value) == isobaricMasses.Count))
{
thisQuantitation = IsobaricQuant.IsobaricImpurityCorrection.CorrectForSignal(correctionMatrix, thisQuantitation).ToArray();
}
if (checkBoxNormalizationChannelSignal.Checked)
{
for (int k = 0; k < thisQuantitation.Length; k++)
{
signalIdentifiedNormalizationDictionary[scn.FileName][k] += thisQuantitation[k];
}
}
scn.Quantitation = new List <List <double> >()
{
thisQuantitation.ToList()
};
}
//And now normalize -------------------
if (checkBoxNormalizationChannelSignal.Checked)
{
Console.WriteLine("Performing channel signal normalization for " + theScansToAnalyze.Count + " scans.");
foreach (SQTScan scn2 in theScansToAnalyze)
{
for (int m = 0; m < isobaricMasses.Count; m++)
{
scn2.Quantitation[0][m] /= signalIdentifiedNormalizationDictionary[scn2.FileName][m];
}
if (scn2.Quantitation[0].Contains(double.NaN))
{
Console.WriteLine("Problems on signal of scan " + scn2.FileNameWithScanNumberAndChargeState);
}
}
}
comboBoxSelectFileForGraphs.Items.Clear();
foreach (string file in signalIdentifiedNormalizationDictionary.Keys.ToList())
{
comboBoxSelectFileForGraphs.Items.Add(file);
}
tabControlMain.SelectedIndex = 1;
if (radioButtonAnalysisPeptideReport.Checked)
{
//Peptide Analysis
//Write Peptide Analysis
StreamWriter sw = new StreamWriter(textBoxOutputDirectory.Text + "/" + "PeptideQuantitationReport.txt");
//Eliminate problematic quants
int removed = theScansToAnalyze.RemoveAll(a => Object.ReferenceEquals(a.Quantitation, null));
Console.WriteLine("Problematic scans removed: " + removed);
var pepDic = from scn in theScansToAnalyze
group scn by scn.PeptideSequenceCleaned
into groupedSequences
select new { PeptideSequence = groupedSequences.Key, TheScans = groupedSequences.ToList() };
foreach (var pep in pepDic)
{
sw.WriteLine("Peptide:" + pep.PeptideSequence + "\tSpecCounts:" + pep.TheScans.Count);
foreach (SQTScan sqt in pep.TheScans)
{
sw.WriteLine(sqt.FileNameWithScanNumberAndChargeState + "\t" + string.Join("\t", sqt.Quantitation[0]));
}
}
//And now write the Fasta
sw.WriteLine("#Fasta Items");
foreach (FastaItem fastaItem in theFastaItems)
{
sw.WriteLine(">" + fastaItem.SequenceIdentifier + " " + fastaItem.Description);
sw.WriteLine(fastaItem.Sequence);
}
sw.Close();
}
else
{
rp = ResultPackage.Load(textBoxitraqSEPro.Text);
//Peptide Level
if (true)
{
PatternTools.SparseMatrixIndexParserV2 ip = new SparseMatrixIndexParserV2();
List <int> allDims = new List <int>();
List <PeptideResult> peptides = rp.MyProteins.MyPeptideList;
if (checkBoxOnlyUniquePeptides.Checked)
{
int removedPeptides = peptides.RemoveAll(a => a.MyMapableProteins.Count > 1);
Console.WriteLine("Removing {0} peptides for not being unique.", removedPeptides);
}
for (int i = 0; i < peptides.Count; i++)
{
SparseMatrixIndexParserV2.Index index = new SparseMatrixIndexParserV2.Index();
index.Name = peptides[i].PeptideSequence;
index.Description = string.Join(" ", peptides[i].MyMapableProteins);
index.ID = i;
ip.Add(index, true);
allDims.Add(i);
}
SparseMatrix sm = new SparseMatrix();
List <int> dims = ip.allIDs();
for (int l = 0; l < labels.Count; l++)
{
if (labels[l] < 0)
{
continue;
}
sparseMatrixRow smr = new sparseMatrixRow(labels[l]);
List <double> values = new List <double>(dims.Count);
List <int> dimsWithValues = new List <int>();
foreach (int d in dims)
{
List <SQTScan> scns = peptides[d].MyScans.FindAll(a => !object.ReferenceEquals(a.Quantitation, null));
if (scns.Count > 0)
{
double signalSum = scns.FindAll(a => !double.IsNaN(a.Quantitation[0][l])).Sum(a => a.Quantitation[0][l]);
values.Add(signalSum);
dimsWithValues.Add(d);
}
}
smr.Dims = dimsWithValues;
smr.Values = values;
smr.FileName = isobaricMasses[l].ToString();
sm.addRow(smr);
}
PatternLabProject plp = new PatternLabProject(sm, ip, "IsobaricQuant");
plp.Save(textBoxOutputDirectory.Text + "/MyPatternLabProjectPeptides.plp");
}
//Protein Level
if (true)
{
//Generate Index
PatternTools.SparseMatrixIndexParserV2 ip = new SparseMatrixIndexParserV2();
List <MyProtein> theProteins = rp.MyProteins.MyProteinList;
if (checkBoxOnlyUniquePeptides.Checked)
{
int removedProteins = theProteins.RemoveAll(a => !a.PeptideResults.Exists(b => b.NoMyMapableProteins == 1));
Console.WriteLine("{0} removed proteins for not having unique peptides", removedProteins);
}
for (int i = 0; i < theProteins.Count; i++)
{
SparseMatrixIndexParserV2.Index index = new SparseMatrixIndexParserV2.Index();
index.ID = i;
index.Name = theProteins[i].Locus;
index.Description = theProteins[i].Description;
ip.Add(index, false);
}
//SparseMatrix
SparseMatrix sm = new SparseMatrix();
List <int> dims = ip.allIDs();
for (int l = 0; l < labels.Count; l++)
{
if (labels[l] < 0)
{
continue;
}
if (!sm.ClassDescriptionDictionary.ContainsKey(labels[l]))
{
sm.ClassDescriptionDictionary.Add(labels[l], labels[l].ToString());
}
sparseMatrixRow smr = new sparseMatrixRow(labels[l]);
List <double> values = new List <double>(dims.Count);
List <int> dimsToInclude = new List <int>();
foreach (int d in dims)
{
double signalSum = 0;
List <PeptideResult> thePeptides = theProteins[d].PeptideResults;
if (checkBoxOnlyUniquePeptides.Checked)
{
thePeptides.RemoveAll(a => a.MyMapableProteins.Count > 1);
}
foreach (PeptideResult pr in thePeptides)
{
List <SQTScan> scns = pr.MyScans.FindAll(a => !object.ReferenceEquals(a.Quantitation, null));
foreach (SQTScan sqt in scns)
{
if (!double.IsNaN(sqt.Quantitation[0][l]) && !double.IsInfinity(sqt.Quantitation[0][l]))
{
signalSum += sqt.Quantitation[0][l];
}
}
}
if (signalSum > 0)
{
dimsToInclude.Add(d);
values.Add(signalSum);
}
else
{
Console.WriteLine("No signal found for " + theProteins[d].Locus + " on marker " + l);
}
}
smr.Dims = dims;
smr.Values = values;
smr.FileName = isobaricMasses[l].ToString();
sm.addRow(smr);
}
PatternLabProject plp = new PatternLabProject(sm, ip, "IsobaricQuant");
plp.Save(textBoxOutputDirectory.Text + "/MyPatternLabProjectProteins.plp");
}
}
comboBoxSelectFileForGraphs.Enabled = true;
tabControlMain.SelectedIndex = 2;
Console.WriteLine("Done");
buttonGo.Text = "Generate Report";
}
