//This is used by the Feature Class to generate Features
public Feature obtainFeatures(OpenFileDialog FileLinks, List<CompositionHypothesisEntry> comhyp)
{
List<Double> Ini = new List<Double>();
List<Double> nCS = new List<Double>();
List<Double> SD = new List<Double>();
List<Double> nMS = new List<Double>();
List<Double> tV = new List<Double>();
List<Double> EA = new List<Double>();
List<Double> CS = new List<Double>();
List<Double> NS = new List<Double>();
List<Double> SN = new List<Double>();
//Each data file is treated separately, hence the for loop.
foreach (String filename in FileLinks.FileNames)
{
//Get the Parameters.
ParametersForm parameter = new ParametersForm();
ParametersForm.ParameterSettings paradata = parameter.GetParameters();
//Perform the First and second grouping, matching and getting data for the features by the Grouping function.
Double Mas = new Double();
Mas = adductMass(comhyp);
List<ResultsGroup> LRLR = new List<ResultsGroup>();
LRLR = Groupings(filename, paradata, Mas, comhyp);
//Error prevention
if (LRLR.Count == 1)
{
MessageBox.Show("There is no match between the hypothesis and the data. Unable to generate results from the file:" + filename);
continue;
}
//##############Logistic Regression####################
//Perform logistic regression to get the Parameters
Feature featureData = new Feature();
featureData = FitLogisticRegression(LRLR);
Ini.Add(featureData.Initial);
nCS.Add(featureData.numChargeStates);
SD.Add(featureData.ScanDensity);
nMS.Add(featureData.numModiStates);
tV.Add(featureData.totalVolume);
EA.Add(featureData.ExpectedA);
CS.Add(featureData.CentroidScan);
NS.Add(featureData.numOfScan);
SN.Add(featureData.avgSigNoise);
}
// Get the average of all features.
Feature finalans = new Feature();
finalans.Initial = Ini.Average();
finalans.numChargeStates = nCS.Average();
finalans.ScanDensity = SD.Average();
finalans.numModiStates = nMS.Average();
finalans.totalVolume = tV.Average();
finalans.ExpectedA = EA.Average();
finalans.CentroidScan = CS.Average();
finalans.numOfScan = NS.Average();
finalans.avgSigNoise = SN.Average();
return finalans;
}
public List<ResultsGroup>[] run(OpenFileDialog FileLinks)
{
List<ResultsGroup>[] AllFinalResults = new List<ResultsGroup>[Convert.ToInt32(FileLinks.FileNames.Count())];
Int32 Count = 0;
//Each data file is treated separately, hence the for loop.
foreach (String filename in FileLinks.FileNames)
{
//Get the Parameters.
ParametersForm parameter = new ParametersForm();
ParametersForm.ParameterSettings paradata = parameter.GetParameters();
//Perform the First and second grouping and getting data for the features by the Grouping function.
List<ResultsGroup> LRLR = new List<ResultsGroup>();
LRLR = Groupings(filename, paradata);
//##############Logistic Regression####################
Features fe = new Features();
//current features
Feature featureData = fe.readFeature();
//default features
String defaultpath = Application.StartupPath + "\\FeatureDefault.fea";
Feature defaultData = fe.readFeature(defaultpath);
//Features that will be used
Feature finalfeatureData = new Feature();
//Here are the beta values in logistic regression.
finalfeatureData.Initial = featureData.Initial * 0.5 + defaultData.Initial * 0.5;
finalfeatureData.numChargeStates = featureData.numChargeStates * 0.5 + defaultData.numChargeStates * 0.5;
finalfeatureData.ScanDensity = featureData.ScanDensity * 0.5 + defaultData.ScanDensity * 0.5;
finalfeatureData.numModiStates = featureData.numModiStates * 0.5 + defaultData.numModiStates * 0.5;
finalfeatureData.totalVolume = featureData.totalVolume * 0.5 + defaultData.totalVolume * 0.5;
finalfeatureData.ExpectedA = featureData.ExpectedA * 0.5 + defaultData.ExpectedA * 0.5;
finalfeatureData.CentroidScan = featureData.CentroidScan * 0.5 + defaultData.CentroidScan * 0.5;
finalfeatureData.numOfScan = featureData.numOfScan * 0.5 + defaultData.numOfScan * 0.5;
finalfeatureData.avgSigNoise = featureData.avgSigNoise * 0.5 + defaultData.avgSigNoise * 0.5;
//Generate scores.
SupervisedLearner sl = new SupervisedLearner();
AllFinalResults[Count] = sl.Scorings(LRLR, finalfeatureData, paradata);
Count++;
}
return AllFinalResults;
}
public List<ResultsGroup>[] evaluate(OpenFileDialog FileLinks, Feature featureData)
{
List<ResultsGroup>[] AllFinalResults = new List<ResultsGroup>[Convert.ToInt32(FileLinks.FileNames.Count())];
Int32 Count = 0;
//Each data file is treated separately, hence the for loop.
foreach (String filename in FileLinks.FileNames)
{
//Get the Parameters.
ParametersForm parameterForm = new ParametersForm();
ParametersForm.ParameterSettings parameters = parameterForm.GetParameters();
//Perform the First and second grouping and getting data for the features by the Grouping function.
List<ResultsGroup> LRLR = new List<ResultsGroup>();
LRLR = Groupings(filename, parameters);
//Generate scores.
SupervisedLearner sl = new SupervisedLearner();
AllFinalResults[Count] = sl.Scorings(LRLR, featureData, parameters);
Count++;
}
return AllFinalResults;
}
//This is used by the Features class to evaluate the features
public List<ResultsGroup>[] EvaluateFeature(OpenFileDialog FileLinks, List<CompositionHypothesisEntry> comhyp, Feature dfeatureData)
{
//Initialize storage variables.
List<ResultsGroup>[] AllFinalResults = new List<ResultsGroup>[Convert.ToInt32(FileLinks.FileNames.Count())];
Int32 Count = 0;
//Each data file is treated separately, hence the for loop.
foreach (String filename in FileLinks.FileNames)
{
//Get the Parameters.
ParametersForm parameter = new ParametersForm();
ParametersForm.ParameterSettings paradata = parameter.GetParameters();
//Perform the First and second grouping, matching and getting data for the features by the Grouping function.
Double Mas = new Double();
Mas = adductMass(comhyp);
List<ResultsGroup> LRLR = new List<ResultsGroup>();
LRLR = Groupings(filename, paradata, Mas, comhyp);
//Error prevention
if (LRLR.Count == 1)
{
MessageBox.Show("There is no match between the hypothesis and the data. Unable to generate results from the file:" + filename);
List<ResultsGroup> FinalResult = LRLR;
AllFinalResults[Count] = FinalResult;
Count++;
continue;
}
//##############Logistic Regression####################
Feature featureData = FitLogisticRegression(LRLR);
//Generate scores.
AllFinalResults[Count] = Scorings(LRLR, featureData, paradata);
Count++;
}
return AllFinalResults;
}
public static StreamWriter WriteResultsToStream(StreamWriter writer, List<ResultsGroup> results, List<String> elementNames, List<String> moleculeNames)
{
String header = "Score,MassSpec MW,Compound Key,PeptideSequence,PPM Error,#ofAdduct,#ofCharges,#ofScans,ScanDensity,Avg A:A+2 Error,A:A+2 Ratio,Total Volume,Signal to Noise Ratio,Centroid Scan Error,Centroid Scan,MaxScanNumber,MinScanNumber";
foreach(var element in elementNames){
header += "," + element;
Console.WriteLine(element);
}
header += ",Hypothesis MW";
foreach (var name in moleculeNames)
{
header += ("," + name);
Console.WriteLine(name);
}
header += ",Adduct/Replacement,Adduct Amount,PeptideModification,PeptideMissedCleavage#,#ofGlycanAttachmentToPeptide,StartAA,EndAA,ProteinID";
ParametersForm pr = new ParametersForm();
ParametersForm.ParameterSettings parameterInfo = pr.GetParameters();
writer.WriteLine(header);
for (int i = 0; i < results.Count; i++)
{
ResultsGroup result = results[i];
DeconRow observed = result.DeconRow;
CompositionHypothesisEntry hypothesis = result.PredictedComposition;
Console.WriteLine(hypothesis);
//If this is a prediction, emit in one format
if (hypothesis.MassWeight != 0)
{
double ppmError = ((observed.MonoisotopicMassWeight - hypothesis.MassWeight) / observed.MonoisotopicMassWeight);
ppmError *= 1000000;
writer.Write(result.Score + "," + observed.MonoisotopicMassWeight + "," +
hypothesis.CompoundComposition + "," + hypothesis.PepSequence + ","
+ ppmError + "," + result.NumModiStates + "," + result.NumChargeStates + ","
+ result.NumOfScan + "," + result.ScanDensity + "," + result.ExpectedA + ","
+ (observed.MonoisotopicAbundance / (observed.MonoisotopicPlus2Abundance + 1)) + ","
+ result.TotalVolume + "," + observed.SignalNoiseRatio + ","
+ result.CentroidScan + "," + observed.ScanNum + ","
+ result.MaxScanNum + "," + result.MinScanNum
);
for (int j = 0; j < elementNames.Count; j++)
{
writer.Write("," + hypothesis.ElementAmount[j]);
}
writer.Write("," + hypothesis.MassWeight);
for (int j = 0; j < moleculeNames.Count; j++)
{
writer.Write("," + hypothesis.eqCounts[j]);
}
writer.WriteLine("," + hypothesis.AddRep + "," + hypothesis.AdductNum + ","
+ hypothesis.PepModification + "," + hypothesis.MissedCleavages + ","
+ hypothesis.NumGlycosylations + "," + hypothesis.StartAA + ","
+ hypothesis.EndAA + "," + hypothesis.ProteinID);
}
else
{
writer.Write(result.Score + "," + result.DeconRow.MonoisotopicMassWeight + "," + 0 + "" + "," + "," + 0 + "," + result.NumModiStates + "," + result.NumChargeStates + "," + result.NumOfScan + "," + result.ScanDensity + "," + result.ExpectedA + "," + (result.DeconRow.MonoisotopicAbundance / (result.DeconRow.MonoisotopicPlus2Abundance + 1)) + "," + result.TotalVolume + "," + result.DeconRow.SignalNoiseRatio + "," + result.CentroidScan + "," + result.DeconRow.ScanNum + "," + result.MaxScanNum + "," + result.MinScanNum);
for (int s = 0; s < elementNames.Count(); s++)
{
writer.Write("," + 0);
}
writer.Write("," + 0);
for (int s = 0; s < moleculeNames.Count(); s++)
{
writer.Write("," + 0);
}
writer.WriteLine("," + "N/A" + "," + 0 + "," + "" + "," + 0 + "," + 0 + "," + 0 + "," + 0);
}
}
writer.Flush();
return writer;
}
public List<DeconRow> getdata(String DeconData)
{
ParametersForm parad = new ParametersForm();
ParametersForm.ParameterSettings paradata = parad.GetParameters();
//try
// {
FileStream fileinput = new FileStream(DeconData, FileMode.Open, FileAccess.Read);
StreamReader readdata = new StreamReader(fileinput);
//The first line in the file contains the column names, we don't need it.
readdata.ReadLine();
while (readdata.Peek() >= 0)
{
DeconRow Row = new DeconRow();
String Line = readdata.ReadLine();
String[] column = Line.Split(',');
Row.ScanNum = Convert.ToInt32(column[0]);
Row.charge = Convert.ToInt32(column[1]);
Row.abundance = Convert.ToInt32(column[2]);
Row.mz = Convert.ToDouble(column[3]);
Row.fit = Convert.ToDouble(column[4]);
Row.average_mw = Convert.ToDouble(column[5]);
Row.MonoisotopicMassWeight = Convert.ToDouble(column[6]);
Row.mostabundant_mw = Convert.ToDouble(column[7]);
Row.fwhm = Convert.ToDouble(column[8]);
Row.SignalNoiseRatio = Convert.ToDouble(column[9]);
Row.MonoisotopicAbundance = Convert.ToInt32(column[10]);
Double mp2a = Convert.ToDouble(column[11]);
Row.MonoisotopicPlus2Abundance = Convert.ToInt32(mp2a);
//Flag maybe empty, so, special treatment.
if (column[12] == "")
{ Row.flag = 0; }
else
{ Row.flag = Convert.ToInt32(column[12]); }
if (Convert.ToInt32(column.Count()) == 14)
Row.interference_sore = Convert.ToDouble(column[13]);
else
Row.interference_sore = 0;
//Check if the data are within the boundaries of the Parameters
if (Row.abundance >= paradata.DataNoiseTheshold)
{
if (Row.MonoisotopicMassWeight <= paradata.MolecularWeightUpperBound)
{
if (Row.MonoisotopicMassWeight >= paradata.MolecularWeightLowerBound)
{
DeconDATA.Add(Row);
}
}
}
}
fileinput.Close();
// }
// catch(Exception ex)
// {
// MessageBox.Show("Error: Could not read DeconTools Data file from disk. Original error: " + ex.Message);
// }
return DeconDATA;
}