private void Run()
{
richTextBox1.Clear();
string rawFolder = txtRawFolder.Text;
List <string> csvFilepaths = new List <string>(lstOmssaCsvFiles.Items.Cast <string>());
List <Modification> fixedModifications = new List <Modification>(from ListViewItem checkedItem in lstSelectedFixedModifications.Items select(Modification) checkedItem.Tag);
double maxFalseDiscoveryRate = (double)numMaximumFalseDiscoveryRate.Value;
double maxPPM = (double)maximumPPMUD.Value;
bool isBatched = checkBox1.Checked;
bool is2DFDR = twoDCB.Checked;
bool includeFixedMods = checkBox2.Checked;
UniquePeptideType uniquePeptideType = (UniquePeptideType)comboBox1.SelectedValue;
string outputFolder = txtOutputFolder.Text;
if (string.IsNullOrEmpty(outputFolder))
{
MessageBox.Show("Output folder must be specified");
return;
}
Directory.CreateDirectory(outputFolder);
FdrOptimizer fdrOptimizer = new FdrOptimizer(csvFilepaths, rawFolder,
fixedModifications,
maxFalseDiscoveryRate, maxPPM, uniquePeptideType,
outputFolder, isBatched, is2DFDR, includeFixedMods);
fdrOptimizer.UpdateProgress += HandleUpdateProgress;
fdrOptimizer.Finished += fdrOptimizer_Finished;
fdrOptimizer.UpdateLog += fdrOptimizer_UpdateLog;
prgProgress.Value = prgProgress.Minimum;
btnOK.Enabled = false;
Thread thread = new Thread(fdrOptimizer.Optimize)
{
IsBackground = true
};
thread.Start();
}
public FdrOptimizer(IList<string> csvFilepaths, string rawFolder,
IList<Modification> fixedModifications,
double maximumFalseDiscoveryRate,
double maximumPPMError,
UniquePeptideType uniquePeptideType,
string outputFolder, bool isBatched = false, bool is2DFDR = true, bool includeFixedMods = false, double evalueThresholdForPPMError = 1e-2)
{
_csvFilepaths = csvFilepaths;
_rawFolder = rawFolder;
_fixedModifications = fixedModifications;
_maximumFalseDiscoveryRate = maximumFalseDiscoveryRate/100.0;
_maximumPPMError = maximumPPMError;
_uniquePeptideType = uniquePeptideType;
_outputFolder = outputFolder;
_outputPsmFolder = Path.Combine(outputFolder, "psms");
_outputPeptideFolder = Path.Combine(outputFolder, "peptides");
_outputScansFolder = Path.Combine(outputFolder, "scans");
_isBatched = isBatched;
_is2DFDR = is2DFDR;
_includeFixedMods = includeFixedMods;
_evalueThresholdPPMError = evalueThresholdForPPMError;
}
public FdrOptimizer(IList <string> csvFilepaths, string rawFolder,
IList <Modification> fixedModifications,
double maximumFalseDiscoveryRate,
double maximumPPMError,
UniquePeptideType uniquePeptideType,
string outputFolder, bool isBatched = false, bool is2DFDR = true, bool includeFixedMods = false, double evalueThresholdForPPMError = 1e-2)
{
_csvFilepaths = csvFilepaths;
_rawFolder = rawFolder;
_fixedModifications = fixedModifications;
_maximumFalseDiscoveryRate = maximumFalseDiscoveryRate / 100.0;
_maximumPPMError = maximumPPMError;
_uniquePeptideType = uniquePeptideType;
_outputFolder = outputFolder;
_outputPsmFolder = Path.Combine(outputFolder, "psms");
_outputPeptideFolder = Path.Combine(outputFolder, "peptides");
_outputScansFolder = Path.Combine(outputFolder, "scans");
_isBatched = isBatched;
_is2DFDR = is2DFDR;
_includeFixedMods = includeFixedMods;
_evalueThresholdPPMError = evalueThresholdForPPMError;
}
private void ReducePsms(IList<InputFile> csvFiles, UniquePeptideType uniquePeptideType, bool isBatched = false)
{
string msg = "Converting PSMs into unique peptides based ";
IEqualityComparer<Peptide> comparer;
switch (uniquePeptideType)
{
default:
msg += "on sequence only";
comparer = new SequenceComparer();
break;
case UniquePeptideType.Mass:
msg += "on mass";
comparer = new MassComparer();
break;
case UniquePeptideType.SequenceAndModifications:
msg += "on sequence and positional modifications";
comparer = new SequenceModComparer();
break;
case UniquePeptideType.SequenceAndModLocations:
msg += "on sequence and modification locations";
comparer = new SequenceAndModPositionComparer();
break;
case UniquePeptideType.SequenceAndMass:
msg += "on sequence and mass";
comparer = new SequenceMassComparer();
break;
case UniquePeptideType.Nothing:
msg += "on nothing (no reduction)";
comparer = new IdentityComparer<Peptide>();
break;
}
Log(msg);
foreach (InputFile csvFile in csvFiles)
{
csvFile.ReducePsms(comparer);
Log(string.Format("{0:N0} unique peptides remain from {1:N0} PSMs from {2}", csvFile.Peptides.Count, csvFile.PeptideSpectralMatches.Count, csvFile.Name));
}
if (!isBatched)
return;
Dictionary<Peptide, Peptide> peptideDictionary = new Dictionary<Peptide, Peptide>(comparer);
foreach (Peptide peptide in csvFiles.SelectMany(csvFile => csvFile.Peptides))
{
Peptide realPeptide;
if (peptideDictionary.TryGetValue(peptide, out realPeptide))
{
foreach (PSM psm in peptide.PSMs)
{
realPeptide.AddPsm(psm);
}
}
else
{
Peptide newPeptide = new Peptide(peptide);
peptideDictionary.Add(newPeptide, newPeptide);
}
}
_allPeptides = peptideDictionary.Values.ToList();
Log(string.Format("{0:N0} unique peptides from all files [Batched]", _allPeptides.Count));
}
private void ReducePsms(IList <InputFile> csvFiles, UniquePeptideType uniquePeptideType, bool isBatched = false)
{
string msg = "Converting PSMs into unique peptides based ";
IEqualityComparer <Peptide> comparer;
switch (uniquePeptideType)
{
default:
msg += "on sequence only";
comparer = new SequenceComparer();
break;
case UniquePeptideType.Mass:
msg += "on mass";
comparer = new MassComparer();
break;
case UniquePeptideType.SequenceAndModifications:
msg += "on sequence and positional modifications";
comparer = new SequenceModComparer();
break;
case UniquePeptideType.SequenceAndModLocations:
msg += "on sequence and modification locations";
comparer = new SequenceAndModPositionComparer();
break;
case UniquePeptideType.SequenceAndMass:
msg += "on sequence and mass";
comparer = new SequenceMassComparer();
break;
case UniquePeptideType.Nothing:
msg += "on nothing (no reduction)";
comparer = new IdentityComparer <Peptide>();
break;
}
Log(msg);
foreach (InputFile csvFile in csvFiles)
{
csvFile.ReducePsms(comparer);
Log(string.Format("{0:N0} unique peptides remain from {1:N0} PSMs from {2}", csvFile.Peptides.Count, csvFile.PeptideSpectralMatches.Count, csvFile.Name));
}
if (!isBatched)
{
return;
}
Dictionary <Peptide, Peptide> peptideDictionary = new Dictionary <Peptide, Peptide>(comparer);
foreach (Peptide peptide in csvFiles.SelectMany(csvFile => csvFile.Peptides))
{
Peptide realPeptide;
if (peptideDictionary.TryGetValue(peptide, out realPeptide))
{
foreach (PSM psm in peptide.PSMs)
{
realPeptide.AddPsm(psm);
}
}
else
{
Peptide newPeptide = new Peptide(peptide);
peptideDictionary.Add(newPeptide, newPeptide);
}
}
_allPeptides = peptideDictionary.Values.ToList();
Log(string.Format("{0:N0} unique peptides from all files [Batched]", _allPeptides.Count));
}
