private void buttonGenerateSummary_Click(object sender, EventArgs e)
{
SEProSummary ss = new SEProSummary();
ss.MyRichTextBox.AppendText("\tFileName\t\tMass Spectra\t(unique/decoy/total)Peptides\tProteins\tMax Parsimony Proteins\n\n");
foreach (DirectoryClassDescription cdd in multipleDirectorySelector1.MyDirectoryDescriptionDictionary)
{
ss.MyRichTextBox.AppendText(new DirectoryInfo(cdd.MyDirectoryFullName).Name + "\t" + cdd.Description + "\n");
Console.WriteLine("SEPro files found in directory: " + new DirectoryInfo(cdd.MyDirectoryFullName).Name);
List <FileInfo> files = new DirectoryInfo(cdd.MyDirectoryFullName).GetFiles("*.sepr", SearchOption.AllDirectories).ToList();
foreach (FileInfo fi in files)
{
Console.WriteLine("\tLoading " + fi.Name);
ResultPackage sepro = ResultPackage.Load(fi.FullName);
ss.MyRichTextBox.AppendText("\t" + fi.Name + "\t\t" + sepro.MyFDRResult.SpectraFDRLabel + "\t" + sepro.MyProteins.MyPeptideList.Count(a => a.MyMapableProteins.Count == 1) + "/" + sepro.MyFDRResult.PeptideFDRLabel + "\t" + sepro.MyFDRResult.ProteinFDRLabel + "\t" + sepro.MaxParsimonyList().Count + "\n");
}
}
ss.ShowDialog();
}
public Core35(List <DirectoryClassDescription> dcds, XQuantClusteringParameters myClusteringParameters)
{
MyClusterParams = myClusteringParameters;
IdentifiedSequencesInFullDirDict = new Dictionary <string, List <string> >();
MyFastaItems = new List <FastaItem>();
SEProFiles = new List <SEProFileInfo>();
//First we will need to load all SEPros and generate a list of peptides with their respectve quantitations
foreach (DirectoryClassDescription cdc in dcds)
{
FileInfo[] files = new DirectoryInfo(cdc.MyDirectoryFullName).GetFiles("*.sepr", SearchOption.AllDirectories);
SEProFiles.Add(new SEProFileInfo(cdc.MyDirectoryFullName, cdc.ClassLabel, cdc.Description, files.Select(a => a.FullName).ToList()));
foreach (FileInfo fi in files)
{
//Make sure we only have 1 sepro file per directory
if (fi.Directory.GetFiles("*.sepr").Count() != 1)
{
throw new Exception("There can be only one SEPro file per directory; error in directory:\n" + fi.DirectoryName);
}
Console.WriteLine("Loading " + fi.FullName);
ResultPackage rp = ResultPackage.Load(fi.FullName);
//Verify if all equivalent ms1 or raw files are in directory
List <string> filesInSEPro = RemoveExtensions((from sqt in rp.MyProteins.AllSQTScans
select sqt.FileName).Distinct().ToList());
List <string> ms1OrRawOrmzMLFiles = fi.Directory.GetFiles("*.ms1").Select(a => a.Name).Concat(fi.Directory.GetFiles("*.RAW").Select(a => a.Name)).Concat(fi.Directory.GetFiles("*.mzML").Select(a => a.Name)).ToList();
ms1OrRawOrmzMLFiles = RemoveExtensions(ms1OrRawOrmzMLFiles);
//Lets store the fasta items
List <MyProtein> proteinsToAnalyze = new List <MyProtein>();
if (myClusteringParameters.MaxParsimony)
{
proteinsToAnalyze = rp.MaxParsimonyList();
}
else
{
proteinsToAnalyze = rp.MyProteins.MyProteinList;
}
IdentifiedSequencesInFullDirDict.Add(fi.Directory.FullName, proteinsToAnalyze.Select(a => a.Locus).ToList());
foreach (MyProtein p in proteinsToAnalyze)
{
if (!MyFastaItems.Exists(a => a.SequenceIdentifier.Equals(p.Locus)))
{
MyFastaItems.Add(new FastaItem(p.Locus, p.Sequence, p.Description));
}
}
//End storing fasta stuff
foreach (string f in filesInSEPro)
{
if (!ms1OrRawOrmzMLFiles.Contains(f))
{
throw new Exception("All .ms1, .mzML, or Thermo .RAW files must be placed in each corresponding SEPro directory. Error in directory:\n" + fi.DirectoryName + "\nfor file:" + f);
}
}
}
}
Process();
PreparePeptideProteinDictionary();
if (myClusteringParameters.RetainOptimal)
{
RetainOptimalSignal();
}
Console.WriteLine("Done");
}
/// <summary>
/// This eliminates decoys and builds the index
/// </summary>
public void ProcessParsedData()
{
//Generate a SEPro Fusion---------------------
ResultPackage seprofusion = FusionSEPro(MyResultPackages.Select(a => a.MyResultPackage).ToList());
///-------------------------------------------
//If we wish to eliminate decoy proteins
if (MyParameters.EliminateDecoys)
{
seprofusion.MyProteins.RemoveDecoyProteins(MyParameters.DecoyTag);
}
//we do this in two steps... first we build the index, then, we build the sparse matrix rows
List <MyProtein> myProteins = seprofusion.MyProteins.MyProteinList;
if (UseMaxParsimony)
{
myProteins = seprofusion.MaxParsimonyList();
}
List <TMPProt> result = (from prot in myProteins
select new TMPProt(prot.Locus, prot.Description, prot.ContainsUniquePeptide)).ToList();
//If we only want proteins of unique peptides
if (MyParameters.MyProteinType == ProteinOutputType.SpecCountsOfUniquePeptides)
{
result = (from r in result
where r.UniquePeptides > 0
select r).ToList();
}
theIndex = new List <ProteinIndexStruct>(result.Count());
for (int i = 0; i < result.Count; i++)
{
ProteinIndexStruct p = new ProteinIndexStruct();
p.Locus = result[i].Locus;
p.Description = result[i].Description;
p.Index = i;
if (!theIndex.Exists(a => a.Locus.Equals(result[i].Locus)))
{
theIndex.Add(p);
}
else
{
throw new Exception("Problems generating SEPro fusion file");
}
}
//int counter = 0;
//foreach (TMPProt r in result)
//{
// counter++;
// ProteinIndexStruct p = new ProteinIndexStruct();
// p.Locus = r.Locus;
// p.Description = r.Description;
// p.Index = counter;
// if (!theIndex.Exists(a => a.Locus.Equals(r.Locus)))
// {
// theIndex.Add(p);
// }
// else
// {
// counter--;
// }
//}
}