public static void FilterForConfidentlyIdentifiedProteinOnly()
{
List <String> ppr = ProteinProphetEvaluator.extractIdentifiedProteinNames(InputFileOrganizer.OriginalProtXMLFile);
String excludedSpectraPerProteinAll = Path.Combine("ProteinTotalSpectraAgainstExcludedSpectra.txt");
String outputFile = Path.Combine(InputFileOrganizer.OutputFolderOfTheRun, "FilteredProteinWithExcludedSpectraCount.txt");
StreamReader sr = new StreamReader(excludedSpectraPerProteinAll);
StreamWriter sw = new StreamWriter(outputFile);
String line = sr.ReadLine();
sw.WriteLine(line);
line = sr.ReadLine();
int count = 0;
while (line != null)
{
String protName = line.Split("\t".ToCharArray())[0];
if (ppr.Contains(protName))
{
sw.WriteLine(line);
count++;
}
line = sr.ReadLine();
}
sw.Close();
Console.WriteLine(count);
}
private static void writeFeatures(List <IdentificationFeatures> idf)
{
log.Info("Classifying positive and negative sets");
// Extract which proteins were confidently identified at 0.01 FDR with protein prophet
List <String> identifiedProteins = ProteinProphetEvaluator.extractIdentifiedProteinNames(InputFileOrganizer.OriginalProtXMLFile);
// Extract which proteins were not confidently identified, with a specified FDR
// threshold
List <String> negativeTrainingSetProteins = ProteinProphetEvaluator.extractNegativeTrainingSetProteinNames(InputFileOrganizer.OriginalProtXMLFile, 0.25);
// 2019-05-23 FOUND IT! Here is where we filter the negative training set with
// above 20% FDR
// Proteins identified with a 0.01 FDR with protein prophet
List <IdentificationFeatures> positiveTrainingSet = new List <IdentificationFeatures>();
// Proteins not identified with a 0.01 FDR protein prophet
List <IdentificationFeatures> negativeTrainingSet = new List <IdentificationFeatures>();
List <IdentificationFeatures> nonPositiveTrainingSet = new List <IdentificationFeatures>();
// Determine which features are in positive or negative training set
foreach (IdentificationFeatures i in idf)
{
String accession = i.getAccession();
if (i.getCardinality() > 0)
{
if (!accession.StartsWith(GlobalVar.DecoyPrefix))
{
//if this is a real protein
if (identifiedProteins.Contains(accession))
{
positiveTrainingSet.Add(i);
}
else
{
nonPositiveTrainingSet.Add(i);
}
if (negativeTrainingSetProteins.Contains(accession))
{
negativeTrainingSet.Add(i);
}
}
else
{
//if it's a decoy protein
negativeTrainingSet.Add(i);
nonPositiveTrainingSet.Add(i);
}
}
}
WriteIdentificationFeaturesFile(OutputFile_PositiveAndNegative, positiveTrainingSet, negativeTrainingSet);
WriteIdentificationFeaturesFile(OutputFile_PositiveAndNonPositive, positiveTrainingSet, nonPositiveTrainingSet);
List <IdentificationFeatures> positiveSetNoDecoy = new List <IdentificationFeatures>();
List <IdentificationFeatures> nonPositiveSetNoDecoy = new List <IdentificationFeatures>();
foreach (IdentificationFeatures i in positiveTrainingSet)
{
if (!i.getAccession().Contains(GlobalVar.DecoyPrefix))
{
positiveSetNoDecoy.Add(i);
}
}
foreach (IdentificationFeatures i in nonPositiveTrainingSet)
{
if (!i.getAccession().Contains(GlobalVar.DecoyPrefix))
{
nonPositiveSetNoDecoy.Add(i);
}
}
WriteIdentificationFeaturesFile(OutputFile_PositiveAndNonPositive_NoDecoy, positiveSetNoDecoy, nonPositiveSetNoDecoy);
}
