public static AnnotatedSpectrum TestNeo4j(IConSol console)
{
DBOptions dbOptions = GetDBOptions(false, false, console);
Samples ProjectRatios = new Samples(@"C:\_IRIC\DATA\NB\ProjectTest_MonoAce_Spiked_19Oct.csv", 0, dbOptions);//Group 2 (all)
Result rez = Propheus.Start(dbOptions, ProjectRatios, false, false, false);
Database.Neo4j.ResultsToNeo4j.Export(rez);
PeptideSpectrumMatch bestPsm = null;
foreach (Query query in rez.queries)
{
foreach (PeptideSpectrumMatch psm in query.psms)
{
if (psm.Peptide.Sequence.CompareTo("GK(acetylation of K)GGK(propionylation of K)GLGK(propionylation of K)GGAK(propionylation of K)R") == 0 &&
(bestPsm == null || bestPsm.ProbabilityScore() < query.psms[0].ProbabilityScore()))
{
bestPsm = psm;
}
}
}
AnnotatedSpectrum aSpec = new AnnotatedSpectrum(ProjectRatios[0], bestPsm.Query.spectrum, bestPsm.Peptide);
return(aSpec);
}
public static void Launch()//Trinity.UnitTest.SettePeptideSample.Launch()
{
try
{
string projectFile = @"C:\_IRIC\DATA\MHC\project.csv"; //Patient C
//@"G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\JUL03_2013\ProjectFile_SETTEpep_OneRAW.csv";//Sette Peptides
//C:\_IRIC\DATA\MHC\project.csv";//MHC M One sample
DBOptions dbOptions = CreateOptions(@"C:\_IRIC\DATA\Test\testMhc\");
Samples Project = new Samples(projectFile, 0, dbOptions);
//ClusterOptions clusterOptions = new ClusterOptions(Project, outputDir, 5, true, 90, true);//TODO validate its in seconds for all file types
/*
* Propheus propheus = new Propheus(dbOptions, Project);
* propheus.PrepareForSearch();
* //foreach (Sample s in propheus.AllSpectras.Keys)
* // propheus.AllSpectras[s].tracks.Export(dbOptions.outputFolder + vsCSV.GetFileName(s.sSDF) + "_Tracks.csv");
* //Result tmp = propheus.Search(propheus.AllQueries, 1.0, false, false, null);
*
* //Save file
* Result result = new Result();
* result.dbOptions = dbOptions;
* result.queries = propheus.AllQueries;
* result.Save();
* //*/
//Load file
Result result = Result.Import(dbOptions.OutputFolder + "State.GraphML");
Propheus propheus = new Propheus(result.dbOptions, Project);
propheus.Load(result);
//*/
Result tmp = propheus.SearchLatestVersion(result.queries, true);//, 1.0, false, false, null);
tmp.WriteInfoToCsv(true);
tmp.Export(0.05, "05_");
tmp.Export(0.02, "02_");
tmp.Export(1.0, "All_");
UnitTest.Tests.MatchAllFragments(tmp);
//tmp.WriteInfoToConsole();
//tmp.Save();
/*
* Optimizer op = new Optimizer(propheus);
* op.LaunchBestPSMOptimization(tmp);
* op.LaunchPrecursorScoreOptimization(tmp);
* /*
* propheus.Align(tmp);
*
* Result tmp2 = propheus.Search(1.0, false, null, propheus.CreateQueries(propheus.AllSpectras));
* tmp2.Export(0.05, "Aligned_05_");
* tmp2.Export(0.02, "Aligned_02_");
* tmp2.Export(double.MaxValue, "Aligned_All_");
* MSSearcher.Export(dbOptions.outputFolder + "Aligned_5PercentOptimized_precursors.csv", Optimizer.PrecursorOptimizer(tmp2.precursors, 0.05));
* tmp2.WriteInfoToConsole();//*/
}
catch (Exception ex)
{
Console.WriteLine("Error in SettePeptideSample : " + ex.Message);
Console.WriteLine(ex.StackTrace);
}
}
public static void Launch(bool restrain = false)
{
//@"G:\Thibault\Olivier\MnR\Databases\BD_RefGenome_WithReverse_2012-06-20.fasta";
//Trypsin
string outputDir = @"C:\_IRIC\DATA\Test\testPK\";
string fastaFile = @"C:\_IRIC\DATA\Yeast\Yeast_SwissProt.fasta";//Yeast
//@"G:\Thibault\Olivier\MQ_vs_Morpheus\Yeast_SwissProt.fasta";//Yeast
//@"G:\Thibault\Olivier\Databases\SProHNoIso_20130430\current\sequences_2013-05-30.fa";
//G:\Thibault\Olivier\MnR\Databases\mini_human_reference_2013-26-03.fasta";//Yeast
string projectFile = @"C:\_IRIC\DATA\PK\NotCalibrated\project_fr1.csv";//Yeast
//@"G:\Thibault\Olivier\MQ_vs_Morpheus\project.csv";//Yeast
//@"G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\JAN22_2013\_Project_FL_Single.csv";
//G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\JUN27_2012\MR 4Rep DS\MassSense\_Test_ProjectFile_MF3.csv";
//G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\MAR18_2013\ProjectFile_TestForProPheus.csv";
DBOptions dbOptions = new DBOptions(fastaFile);
Samples Project = new Samples(projectFile, 0, dbOptions);
dbOptions.precursorMassTolerance = new MassTolerance(80, MassToleranceUnits.ppm);
dbOptions.productMassTolerance = new MassTolerance(80, MassToleranceUnits.ppm);//0.034 is a 60 000 resolution over 2000 range in mz
dbOptions.MaximumPeptideMass = 200000;
dbOptions.OutputFolder = outputDir;
ProteaseDictionary proteases = ProteaseDictionary.Instance;
dbOptions.DigestionEnzyme = proteases["trypsin (no proline rule)"]; //"no enzyme"];
dbOptions.NoEnzymeSearch = false; // true;
dbOptions.DecoyFusion = false;
//dbOptions.protease = proteases["trypsin (no proline rule)"];
dbOptions.ToleratedMissedCleavages = 2;
dbOptions.MinimumPeptideLength = 5;
dbOptions.MaximumPeptideLength = 300;
GraphML_List <Modification> fixMods = new GraphML_List <Modification>();
fixMods.Add(ModificationDictionary.Instance["carbamidomethylation of C"]);
dbOptions.fixedModifications = fixMods;
GraphML_List <Modification> varMods = new GraphML_List <Modification>();
//Oxidation (M);Acetyl (Protein N-term);Phospho (STY)
//Mods for Yeast
if (!restrain)
{
//varMods.Add(ModificationDictionary.Instance["oxidation of M"]);
//varMods.Add(ModificationDictionary.Instance["acetylation of protein N-terminus"]);
//varMods.Add(ModificationDictionary.Instance["phosphorylation of S"]);
//varMods.Add(ModificationDictionary.Instance["phosphorylation of T"]);
//varMods.Add(ModificationDictionary.Instance["phosphorylation of Y"]);//*/
varMods.Add(ModificationDictionary.Instance["Pionylation of peptide N-terminus"]);
dbOptions.maximumVariableModificationIsoforms = 1024;// 2 * (varMods.Count + fixMods.Count);//TODO Evaluate the viability of this parameter
}
else
{
dbOptions.maximumVariableModificationIsoforms = 2;
}
dbOptions.variableModifications = varMods;
dbOptions.addFragmentLoss = false;
dbOptions.addFragmentMods = false;
dbOptions.fragments = new Fragments();
//dbOptions.fragments.Add(new FragmentA());
dbOptions.fragments.Add(new FragmentB());
//dbOptions.fragments.Add(new FragmentC());
//dbOptions.fragments.Add(new FragmentX());
dbOptions.fragments.Add(new FragmentY());
//dbOptions.fragments.Add(new FragmentZ());
//ClusterOptions clusterOptions = new ClusterOptions(Project, outputDir, 5, true, 90, true);//TODO validate its in seconds for all file types
Propheus propheus = new Propheus(dbOptions, Project);
dbOptions.SaveMS1Peaks = false;
dbOptions.SaveMSMSPeaks = true;
dbOptions.LoadSpectraIfFound = true;
dbOptions.NbPSMToKeep = 10;
propheus.Preload(false, false);
propheus.PrepareQueries();
//First pass (used to optimize parameters and score weights)
Result tmp = propheus.SearchLatestVersion(propheus.AllQueries, false);
//Compute number of Moded (Pionylated peptides) at given Ms tolerance
long nbPsmDecoy = 0;
long nbPsmTarget = 0;
double avgShift = 0;
foreach (Query query in tmp.queries)
{
foreach (PeptideSpectrumMatch psm in query.psms)
{
if (string.IsNullOrEmpty(psm.Peptide.VariableModificationsInString))
{
nbPsmTarget++;
avgShift += psm.PrecursorMzError;
}
else
{
nbPsmDecoy++;
}
}
}
avgShift /= (double)nbPsmTarget;
dbOptions.ConSole.WriteLine("Average shift : " + avgShift);
//tmp.WriteInfoToCsv(true);
//tmp.Export(0.02, "FirstPass_02_");
}
public static void Launch(IConSol console)
{
//@"G:\Thibault\Olivier\MnR\Databases\BD_RefGenome_WithReverse_2012-06-20.fasta";
//Trypsin
string outputDir = @"C:\_IRIC\DATA\Yeast\Results\";
string fastaFile = @"C:\_IRIC\DATA\Yeast\Yeast_SwissProt.fasta";//Yeast
//@"G:\Thibault\Olivier\MQ_vs_Morpheus\Yeast_SwissProt.fasta";//Yeast
//@"G:\Thibault\Olivier\Databases\SProHNoIso_20130430\current\sequences_2013-05-30.fa";
//G:\Thibault\Olivier\MnR\Databases\mini_human_reference_2013-26-03.fasta";//Yeast
string projectFile = @"C:\_IRIC\DATA\Yeast\project.csv";//Yeast
//@"G:\Thibault\Olivier\MQ_vs_Morpheus\project.csv";//Yeast
//@"G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\JAN22_2013\_Project_FL_Single.csv";
//G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\JUN27_2012\MR 4Rep DS\MassSense\_Test_ProjectFile_MF3.csv";
//G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\MAR18_2013\ProjectFile_TestForProPheus.csv";
DBOptions dbOptions = new DBOptions(fastaFile, console);
Samples Project = new Samples(projectFile, 0, dbOptions);
dbOptions.precursorMassTolerance = new MassTolerance(8 /*8*//*8withoutisotopes*/, MassToleranceUnits.ppm);
dbOptions.productMassTolerance = new MassTolerance(0.034 /*0.034*//*without isotopes*/, MassToleranceUnits.Da);//0.034 is a 60 000 resolution over 2000 range in mz
//dbOptions.productMassTolerance = new MassTolerance(20, MassToleranceUnits.ppm);
dbOptions.MaximumPeptideMass = 200000;
dbOptions.OutputFolder = outputDir;
ProteaseDictionary proteases = ProteaseDictionary.Instance;
dbOptions.DigestionEnzyme = proteases["trypsin (no proline rule)"]; //"no enzyme"];
dbOptions.NoEnzymeSearch = false; // true;
dbOptions.DecoyFusion = false;
//dbOptions.protease = proteases["trypsin (no proline rule)"];
dbOptions.ToleratedMissedCleavages = 2;
dbOptions.MinimumPeptideLength = 5;
dbOptions.MaximumPeptideLength = 300;
GraphML_List <Modification> fixMods = new GraphML_List <Modification>();
fixMods.Add(ModificationDictionary.Instance["carbamidomethylation of C"]);
dbOptions.fixedModifications = fixMods;
GraphML_List <Modification> varMods = new GraphML_List <Modification>();
//Oxidation (M);Acetyl (Protein N-term);Phospho (STY)
//Mods for Yeast
varMods.Add(ModificationDictionary.Instance["oxidation of M"]);
varMods.Add(ModificationDictionary.Instance["acetylation of protein N-terminus"]);
varMods.Add(ModificationDictionary.Instance["phosphorylation of S"]);
varMods.Add(ModificationDictionary.Instance["phosphorylation of T"]);
varMods.Add(ModificationDictionary.Instance["phosphorylation of Y"]); //*/
dbOptions.maximumVariableModificationIsoforms = 2 * (varMods.Count + fixMods.Count); //TODO Evaluate the viability of this parameter
dbOptions.variableModifications = varMods;
dbOptions.NbPSMToKeep = 16;
dbOptions.addFragmentLoss = false;
dbOptions.addFragmentMods = false;
dbOptions.fragments = new Fragments();
dbOptions.fragments.Add(new FragmentA());
dbOptions.fragments.Add(new FragmentB());
dbOptions.fragments.Add(new FragmentC());
dbOptions.fragments.Add(new FragmentX());
dbOptions.fragments.Add(new FragmentY());
dbOptions.fragments.Add(new FragmentZ());
dbOptions.dProduct = 0.0;
dbOptions.dPrecursor = 0.1; // 0.12;
dbOptions.dMatchingProductFraction = 0.8; // 0.45;
dbOptions.dMatchingProduct = 0.0; // 0.5;
dbOptions.dIntensityFraction = 0.1; // 45;// 0.0;//0.13;
dbOptions.dIntensity = 0;
dbOptions.dProtein = 0;
dbOptions.dPeptideScore = 0.0; // 0.3;
dbOptions.dFragmentScore = 0.0; // 0.5;
//ClusterOptions clusterOptions = new ClusterOptions(Project, outputDir, 5, true, 90, true);//TODO validate its in seconds for all file types
dbOptions.SaveMS1Peaks = true;
dbOptions.SaveMSMSPeaks = true;
dbOptions.LoadSpectraIfFound = true;
Propheus propheus = new Propheus(dbOptions, Project);
propheus.Preload(false, false);
propheus.PrepareQueries();
//To beat : 4653 (MaxQuant) Psm at 2%FDR
//First pass (used to optimize parameters and score weights)
Result tmp = propheus.SearchLatestVersion(propheus.AllQueries, true, false);//, 1.0, false, false, null);
tmp.WriteInfoToCsv(true);
tmp.Export(0.02, "FirstPass_02_");
//Second search
propheus.Preload(true);
propheus.PrepareQueries();
Result finalRez = propheus.SearchLatestVersion(propheus.AllQueries, false);//, 1.0, false, false, null);
//tmp.Export(0.05, "05_");
tmp.Export(0.02, "02_");
//tmp.Export(0.05, "05_AllFragments");
// tmp.Export(0.01, "01_");
//tmp.Export(double.MaxValue, "All_");
//tmp.WriteInfoToConsole();
/*
* Optimizer op = new Optimizer(propheus);
* op.LaunchBestPSMOptimization(tmp);//.proteins, propheus.AllQueries);
* //*/
//Optimizer op = new Optimizer(propheus);
//MSSearcher.Export(dbOptions.outputFolder + "5PercentOptimized_precursors.csv", Optimizer.PrecursorOptimizer(tmp.precursors, 0.05));
//op.LaunchBestPSMOptimization(tmp);//.proteins, propheus.AllQueries);
//op.LaunchPrecursorScoreOptimization(tmp);//.proteins, propheus.AllQueries);
//op.Launch(tmp.proteins, propheus.AllQueries);
/*
* propheus.Align(tmp);
*
* Result tmp2 = propheus.Search(1.0, false, null, propheus.CreateQueries(propheus.AllSpectras));
* tmp2.Export(0.05, "Aligned_05_");
* tmp2.Export(double.MaxValue, "Aligned_All_");
* MSSearcher.Export(dbOptions.outputFolder + "Aligned_5PercentOptimized_precursors.csv", Optimizer.PrecursorOptimizer(tmp2.precursors, 0.05));
* tmp.WriteInfoToConsole();//*/
}
public static bool Run()
{
//TODO test GPU instead
DBOptions dbOptions = MhcSample.CreateOptions("");
Dictionary <string, int> sequences = new Dictionary <string, int>();
List <Protein> proteins = Propheus.ReadProteomeFromFasta(Path.Combine(Path.GetDirectoryName(Environment.GetCommandLineArgs()[0]), "UnitTest", "proteins.fasta"), false, dbOptions);//ETLPAMCNVYYVNCMAPLTE
string sequence = proteins[0].BaseSequence;
double[] proteinMasses = new double[sequence.Length];
List <double> precursors = new List <double>();
for (int i = 0; i < sequence.Length; i++)
{
for (int j = i + dbOptions.MinimumPeptideLength - 1; j < sequence.Length; j++)
{
int size = j - i + 1;
if (size <= dbOptions.MaximumPeptideLength)
{
string subStr = sequence.Substring(i, j - i + 1);
if (!sequences.ContainsKey(subStr))
{
sequences.Add(subStr, 1);
}
else
{
sequences[subStr]++;
}
double mass = Constants.WATER_MONOISOTOPIC_MASS;
for (int k = 0; k < subStr.Length; k++)
{
mass += AminoAcidMasses.GetMonoisotopicMass(subStr[k]);
}
precursors.Add(mass);
}
}
proteinMasses[i] = AminoAcidMasses.GetMonoisotopicMass(sequence[i]);
}
precursors.Sort();
Queries queries = new Queries(dbOptions, precursors.ToArray());
Digestion ps = new Digestion(dbOptions);
List <Protein> lProt = new List <Protein>();
lProt.Add(proteins[0]);
//for each protein, build matrix of mass
//Trinity_Gpu.ProteinDigest pg = new Trinity_Gpu.ProteinDigest(precursors.ToArray(), sequence.Length);
//Test twice to test that precursor list stays in gpu memory
for (int iter = 0; iter < 2; iter++)
{
Dictionary <string, int> sequencesTmp = new Dictionary <string, int>(sequences);
foreach (Tuple <Peptide, int> item in ps.DigestProteomeOnTheFlyNoEnzyme(lProt, queries))
{
sequencesTmp[item.Item1.BaseSequence] -= 1;//TODO add modifications
}
/*
* foreach (Trinity_Gpu.ProteinPrecursorMatch match in pg.Execute(proteinMasses, 0.00005, 10000000))//TODO compute correct tolerance window
* {
* int size = match.proteinEndPos - match.proteinStartPos;
* string str = sequence.Substring(match.proteinStartPos, size);
* if (size >= dbOptions.MinimumPeptideLength)
* {
* sequencesTmp[str] -= 1;//TODO add modifications
* }
* }//*/
foreach (int val in sequencesTmp.Values)
{
if (val != 0)
{
return(false);//*/
}
}
}
//pg.Dispose();
return(true);
}
public static void Launch()//Trinity.UnitTest.SettePeptideSample.Launch()
{
string outputDir = @"C:\_IRIC\DATA\Test\testRTMHC\";
string fastaFile = @"C:\_IRIC\DATA\MHC Sette\MHC_Sette_Peptides_20091001.fasta";
string projectFile = //@"G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\AUG06_2013\RT_MHC\Project_TEST_600mM.csv";
@"G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\MAR26_2013\Project_NonFAIMS.csv";
// @"G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\SEP10_2013\Project_TEST_75_100_300mM.csv";
DBOptions dbOptions = new DBOptions(fastaFile);
try
{
Samples Project = new Samples(projectFile, 0, dbOptions);
dbOptions.precursorMassTolerance = new MassTolerance(5, MassToleranceUnits.ppm);
dbOptions.productMassTolerance = new MassTolerance(0.068, MassToleranceUnits.Da);//0.034 is a 60 000 resolution over 2000 range in mz
dbOptions.MaximumPeptideMass = 200000;
dbOptions.OutputFolder = outputDir;
ProteaseDictionary proteases = ProteaseDictionary.Instance;
dbOptions.DigestionEnzyme = proteases["no enzyme"];
dbOptions.NoEnzymeSearch = false;
dbOptions.ToleratedMissedCleavages = 20;// 2;
GraphML_List <Modification> fixMods = new GraphML_List <Modification>();
//fixMods.Add(ModificationDictionary.Instance["carbamidomethylation of C"]);
dbOptions.fixedModifications = fixMods;
GraphML_List <Modification> varMods = new GraphML_List <Modification>();
varMods.Add(ModificationDictionary.Instance["oxidation of M"]); //+ Deamidation M Q
varMods.Add(ModificationDictionary.Instance["phosphorylation of S"]);
varMods.Add(ModificationDictionary.Instance["phosphorylation of T"]);
varMods.Add(ModificationDictionary.Instance["phosphorylation of Y"]); //*/
varMods.Add(ModificationDictionary.Instance["deamidation of N"]);
varMods.Add(ModificationDictionary.Instance["deamidation of Q"]);
varMods.Add(ModificationDictionary.Instance["cysteinylation of C"]);
dbOptions.variableModifications = varMods;
dbOptions.maximumVariableModificationIsoforms = 1024;// 2 * (varMods.Count + fixMods.Count);//TODO Evaluate the viability of this parameter
dbOptions.PSMFalseDiscoveryRate = 0.05;
dbOptions.addFragmentMods = false;
dbOptions.addFragmentLoss = false;
dbOptions.fragments = new Fragments();
dbOptions.fragments.Add(new FragmentB());
dbOptions.fragments.Add(new FragmentY());
dbOptions.MinimumPrecursorIntensityRatioInIsolationWindow = 0.05;
Propheus propheus = new Propheus(dbOptions, Project);
dbOptions.SaveMS1Peaks = true;
dbOptions.SaveMSMSPeaks = true;
dbOptions.LoadSpectraIfFound = true;
propheus.Preload(true);
propheus.PrepareQueries();
}
catch (Exception ex)
{
dbOptions.ConSole.WriteLine("Error in SettePeptideSample : " + ex.Message);
dbOptions.ConSole.WriteLine(ex.StackTrace);
}
}
public void Solve(string[] spikedRaws, string[] mixedRaws, string fastaFile, string folderToOutputTo, IConSol conSol)
{
dbOptions = CreateOptions(fastaFile, folderToOutputTo, conSol);
SpikedSamples = new Samples(dbOptions);
for (int i = 0; i < spikedRaws.Length; i++)
{
SpikedSamples.Add(new Sample(i + 1, 1, 1, spikedRaws[i], spikedRaws[i], 0, ""));
}
//Precompute Spiked peptide identifications
SpikedResult = Propheus.Start(dbOptions, SpikedSamples, false, false, true, false);
SpikedResult.ExportPSMs(1, dbOptions.OutputFolder + "Identifications" + System.IO.Path.DirectorySeparatorChar + "SpikedSamplesPSMs.csv");
MixedSamples = new Samples(dbOptions);
for (int i = 0; i < mixedRaws.Length; i++)
{
MixedSamples.Add(new Sample(i + 1, 1, 1, mixedRaws[i], mixedRaws[i], 0, ""));
}
//Precompute Mixed peptide identifications
mixedResult = Propheus.Start(dbOptions, MixedSamples, false, false, true, false);
mixedResult.ExportPSMs(1, dbOptions.OutputFolder + "Identifications" + System.IO.Path.DirectorySeparatorChar + "MixedSamplesPSMs.csv");
conSol.WriteLine("Computing gradient descents...");
//Compute all usable spiked peptides
characterizedPeptides = CharacterizedPrecursor.GetSpikedPrecursors(SpikedSamples, SpikedResult, dbOptions, nbMinFragments, nbMaxFragments, precision);
ExportSpikedSampleResult(characterizedPeptides, dbOptions);
vsCSVWriter writerCumul = new vsCSVWriter(OutputFolder + "Results.csv");
string titleCombined = "Mixed Sample,Precursor";
string curveStr = "Polynomial Curve,";
string spikedIntensityStr = "Area under the curve,";
foreach (double precursor in characterizedPeptides.Keys)
{
foreach (CharacterizedPrecursor charPrec in characterizedPeptides[precursor].Values)
{
titleCombined += "," + charPrec.Peptide.Sequence + " Charge " + charPrec.Charge;
if (charPrec.eCurve.Coefficients != null && charPrec.eCurve.Coefficients.Length == 3)
{
curveStr += "," + charPrec.eCurve.Coefficients[0] + "x^2 + " + charPrec.eCurve.Coefficients[1] + "x" + charPrec.eCurve.Coefficients[2];
}
else
{
curveStr += ",NA";
}
spikedIntensityStr += "," + charPrec.eCurve.Area;
}
}
writerCumul.AddLine(titleCombined);
writerCumul.AddLine(curveStr);
writerCumul.AddLine(spikedIntensityStr);
//mixedPrecursors = new Dictionary<Sample, Dictionary<double, MixedPrecursor>>();
mixedPrecursors = new Dictionary <Sample, List <MixedPrecursor> >();
foreach (Sample mixedSample in MixedSamples)
{
mixedPrecursors.Add(mixedSample, MixedPrecursor.GetMixedPrecursors(mixedSample, mixedResult, dbOptions, characterizedPeptides));
}
//Get the list of precursors to characterize
foreach (Sample mixedSample in MixedSamples)
{
foreach (double keyMz in characterizedPeptides.Keys)
{
List <Dictionary <CharacterizedPrecursor, ElutionCurve> > listOfRatios = new List <Dictionary <CharacterizedPrecursor, ElutionCurve> >();
foreach (MixedPrecursor mPrec in mixedPrecursors[mixedSample])
{
if (mPrec.MZ == keyMz)
{
// Compute Max Flow for this precursor
Dictionary <CharacterizedPrecursor, ElutionCurve> ratios = GetRatios(characterizedPeptides, mPrec);
listOfRatios.Add(ratios);
ExportMixedSampleResult(ratios, mixedSample, mPrec, keyMz, dbOptions);
}
}
bool isEmpty = true;
string resultStr = vsCSV.GetFileName(mixedSample.sSDF) + "," + keyMz;
foreach (double precursor in characterizedPeptides.Keys)
{
foreach (CharacterizedPrecursor charPrec in characterizedPeptides[precursor].Values)
{
double cumulArea = 0.0;
foreach (Dictionary <CharacterizedPrecursor, ElutionCurve> ratios in listOfRatios)
{
if (ratios.ContainsKey(charPrec))
{
cumulArea += ratios[charPrec].Area;
}
}
resultStr += "," + cumulArea;
if (cumulArea > 0)
{
isEmpty = false;
}
}
}
if (!isEmpty)
{
writerCumul.AddLine(resultStr);
}
}
}
writerCumul.WriteToFile();
//List Modifications
Dictionary <Modification, double> dicOfIntensityPerMod = new Dictionary <Modification, double>();
foreach (Sample sample in mixedPrecursors.Keys)
{
foreach (MixedPrecursor mP in mixedPrecursors[sample])
{
foreach (CharacterizedPrecursor cP in mP.PeptideRatios.Keys)
{
if (cP.Peptide.VariableModifications != null)
{
foreach (Modification mod in cP.Peptide.VariableModifications.Values)
{
if (!dicOfIntensityPerMod.ContainsKey(mod))
{
dicOfIntensityPerMod.Add(mod, 0.0);
}
}
}
}
}
}
//Compute site occupancy for identical sequences (real positionnal isomers)
vsCSVWriter writerSitesOccupancy = new vsCSVWriter(OutputFolder + "Results_SiteOccupancy.csv");
List <Protein> AllProteins = Propheus.ReadProteomeFromFasta(fastaFile, false, dbOptions);
foreach (Protein protein in AllProteins)
{
string newTitleProtein = protein.Description.Replace(',', ' ') + "," + protein.Sequence;
for (int i = 0; i < protein.Sequence.Length; i++)
{
newTitleProtein += "," + protein[i].ToString();
}
writerSitesOccupancy.AddLine(newTitleProtein);
foreach (Sample mixedSample in mixedPrecursors.Keys)
{
string coverage = "Coverage," + mixedSample.Name;
for (int i = 0; i < protein.Sequence.Length; i++)
{
double cumulSite = 0.0;
newTitleProtein += "," + protein[i].ToString();
foreach (MixedPrecursor mP in mixedPrecursors[mixedSample])
{
foreach (CharacterizedPrecursor cP in mP.PeptideRatios.Keys)
{
if (i + 1 >= cP.Peptide.StartResidueNumber && i + 1 <= cP.Peptide.EndResidueNumber)
{
cumulSite += mP.PeptideRatios[cP].Area;
}
}
}
coverage += "," + cumulSite;
}
writerSitesOccupancy.AddLine(coverage);
}
foreach (Modification mod in dicOfIntensityPerMod.Keys)
{
Dictionary <Sample, string> dicOfLines = new Dictionary <Sample, string>();
for (int i = 0; i < protein.Sequence.Length; i++)
{
foreach (Sample mixedSample in mixedPrecursors.Keys)
{
double cumulModArea = 0.0;
foreach (MixedPrecursor mP in mixedPrecursors[mixedSample])
{
foreach (CharacterizedPrecursor cP in mP.PeptideRatios.Keys)
{
if (i + 1 >= cP.Peptide.StartResidueNumber && i + 1 <= cP.Peptide.EndResidueNumber &&
cP.Peptide.VariableModifications != null)
{
foreach (int pos in cP.Peptide.VariableModifications.Keys)
{
if (cP.Peptide.StartResidueNumber + pos - 2 == i + 1 && cP.Peptide.VariableModifications[pos] == mod)
{
cumulModArea += mP.PeptideRatios[cP].Area;
}
}
}
}
}
if (!dicOfLines.ContainsKey(mixedSample))
{
dicOfLines.Add(mixedSample, mod.Description + "," + mixedSample.Name + "," + cumulModArea);
}
else
{
dicOfLines[mixedSample] += "," + cumulModArea;
}
}
}
foreach (string line in dicOfLines.Values)
{
writerSitesOccupancy.AddLine(line);
}
}
}
writerSitesOccupancy.WriteToFile();
}
public static void Launch()//Trinity.UnitTest.SettePeptideSample.Launch()
{
string outputDir = @"C:\_IRIC\DATA\Test\testMHCSette\";
string fastaFile = @"C:\_IRIC\DATA\MHC Sette\MHC_Sette_Peptides_20091001.fasta";
string projectFile = @"G:\Thibault\-=Proteomics_Raw_Data=-\ELITE\JUL03_2013\ProjectFile_SETTEpep_OneRAW.csv";
DBOptions dbOptions = new DBOptions(fastaFile);
try
{
Samples Project = new Samples(projectFile, 0, dbOptions);
dbOptions.precursorMassTolerance = new MassTolerance(5, MassToleranceUnits.ppm);
dbOptions.productMassTolerance = new MassTolerance(0.068, MassToleranceUnits.Da);//0.034 is a 60 000 resolution over 2000 range in mz
dbOptions.MaximumPeptideMass = 200000;
dbOptions.OutputFolder = outputDir;
ProteaseDictionary proteases = ProteaseDictionary.Instance;
dbOptions.DigestionEnzyme = proteases["no enzyme"];
dbOptions.NoEnzymeSearch = false;
dbOptions.ToleratedMissedCleavages = 20;// 2;
GraphML_List <Modification> fixMods = new GraphML_List <Modification>();
//fixMods.Add(ModificationDictionary.Instance["carbamidomethylation of C"]);
dbOptions.fixedModifications = fixMods;
GraphML_List <Modification> varMods = new GraphML_List <Modification>();
varMods.Add(ModificationDictionary.Instance["oxidation of M"]); //+ Deamidation M Q
varMods.Add(ModificationDictionary.Instance["phosphorylation of S"]);
varMods.Add(ModificationDictionary.Instance["phosphorylation of T"]);
varMods.Add(ModificationDictionary.Instance["phosphorylation of Y"]); //*/
varMods.Add(ModificationDictionary.Instance["deamidation of N"]);
varMods.Add(ModificationDictionary.Instance["deamidation of Q"]);
varMods.Add(ModificationDictionary.Instance["cysteinylation of C"]);
dbOptions.variableModifications = varMods;
dbOptions.maximumVariableModificationIsoforms = 1024;// 2 * (varMods.Count + fixMods.Count);//TODO Evaluate the viability of this parameter
dbOptions.PSMFalseDiscoveryRate = 0.05;
dbOptions.addFragmentMods = false;
dbOptions.addFragmentLoss = false;
dbOptions.fragments = new Fragments();
// dbOptions.fragments.Add(new FragmentA());
dbOptions.fragments.Add(new FragmentB());
// dbOptions.fragments.Add(new FragmentC());
// dbOptions.fragments.Add(new FragmentX());
dbOptions.fragments.Add(new FragmentY());
// dbOptions.fragments.Add(new FragmentZ());
//ClusterOptions clusterOptions = new ClusterOptions(Project, outputDir, 5, true, 90, true);//TODO validate its in seconds for all file types
Propheus propheus = new Propheus(dbOptions, Project);
dbOptions.SaveMS1Peaks = false;
dbOptions.SaveMSMSPeaks = false;
dbOptions.LoadSpectraIfFound = true;
propheus.Preload(true);
propheus.PrepareQueries();
//First pass (used to optimize parameters and score weights)
Result tmp = propheus.SearchVersionAugust2013(propheus.AllQueries, true);//, 1.0, false, false, null);
tmp.WriteInfoToCsv(true);
tmp.Export(0.02, "FirstPass_02_");
dbOptions.SaveMS1Peaks = true;
dbOptions.SaveMSMSPeaks = true;
//Second search
propheus.Preload(true);
propheus.PrepareQueries();
Result finalRez = propheus.SearchLatestVersion(propheus.AllQueries, false);//, 1.0, false, false, null);
//tmp.Export(0.05, "05_");
tmp.Export(0.02, "02_");
//tmp.Export(1.0, "All_");
//UnitTest.Tests.MatchAllFragments(tmp);
//tmp.WriteInfoToConsole();
//tmp.Save();
/*
* Optimizer op = new Optimizer(propheus);
* op.LaunchBestPSMOptimization(tmp);
* op.LaunchPrecursorScoreOptimization(tmp);
* /*
* propheus.Align(tmp);
*
* Result tmp2 = propheus.Search(1.0, false, null, propheus.CreateQueries(propheus.AllSpectras));
* tmp2.Export(0.05, "Aligned_05_");
* tmp2.Export(0.02, "Aligned_02_");
* tmp2.Export(double.MaxValue, "Aligned_All_");
* MSSearcher.Export(dbOptions.outputFolder + "Aligned_5PercentOptimized_precursors.csv", Optimizer.PrecursorOptimizer(tmp2.precursors, 0.05));
* tmp2.WriteInfoToConsole();//*/
}
catch (Exception ex)
{
dbOptions.ConSole.WriteLine("Error in SettePeptideSample : " + ex.Message);
dbOptions.ConSole.WriteLine(ex.StackTrace);
}
}
public void Solve(string[] mixedRaws, string fastaFile, string folderToOutputTo, IConSol conSol)
{
dbOptions = CreateOptions(fastaFile, folderToOutputTo, conSol);
MixedSamples = new Samples(dbOptions);
for (int i = 0; i < mixedRaws.Length; i++)
{
MixedSamples.Add(new Sample(i + 1, 1, 1, mixedRaws[i], mixedRaws[i], 0, ""));
}
//Precompute Mixed peptide identifications
mixedResult = Propheus.Start(dbOptions, MixedSamples, false, false, true, false);
conSol.WriteLine("Computing gradient descents...");
//Compute all usable spiked peptides
characterizedPeptides = CharacterizedPrecursor.GetSpikedPrecursors(MixedSamples, mixedResult, dbOptions, nbMinFragments, nbMaxFragments, precision);
ExportSpikedSampleResult(characterizedPeptides, dbOptions);
vsCSVWriter writerCumul = new vsCSVWriter(OutputFolder + "Results.csv");
string titleCombined = "Mixed Sample,Precursor";
string curveStr = "Polynomial Curve,";
string spikedIntensityStr = "Area under the curve,";
foreach (double precursor in characterizedPeptides.Keys)
{
foreach (CharacterizedPrecursor charPrec in characterizedPeptides[precursor].Values)
{
titleCombined += "," + charPrec.Peptide.Sequence + " Charge " + charPrec.Charge;
if (charPrec.eCurve.Coefficients != null && charPrec.eCurve.Coefficients.Length == 3)
{
curveStr += "," + charPrec.eCurve.Coefficients[0] + "x^2 + " + charPrec.eCurve.Coefficients[1] + "x" + charPrec.eCurve.Coefficients[2];
}
else
{
curveStr += ",NA";
}
spikedIntensityStr += "," + charPrec.eCurve.Area;
}
}
writerCumul.AddLine(titleCombined);
writerCumul.AddLine(curveStr);
writerCumul.AddLine(spikedIntensityStr);
//mixedPrecursors = new Dictionary<Sample, Dictionary<double, MixedPrecursor>>();
mixedPrecursors = new Dictionary <Sample, List <MixedPrecursor> >();
foreach (Sample mixedSample in MixedSamples)
{
mixedPrecursors.Add(mixedSample, MixedPrecursor.GetMixedPrecursors(mixedSample, mixedResult, dbOptions, characterizedPeptides));
}
//Get the list of precursors to characterize
foreach (Sample mixedSample in MixedSamples)
{
foreach (double keyMz in characterizedPeptides.Keys)
{
List <Dictionary <Peptide, ElutionCurve> > listOfRatios = new List <Dictionary <Peptide, ElutionCurve> >();
foreach (MixedPrecursor mPrec in mixedPrecursors[mixedSample])
{
if (mPrec.MZ == keyMz)
{
// Compute Max Flow for this precursor
Dictionary <Peptide, ElutionCurve> ratios = GetRatiosNoSpikes(mPrec, precision);
listOfRatios.Add(ratios);
ExportMixedSampleResult(ratios, mixedSample, mPrec, keyMz, dbOptions);
}
}
/*
* string resultStr = vsCSV.GetFileName(mixedSample.sSDF) + "," + keyMz;
* foreach (double precursor in characterizedPeptides.Keys)
* {
* foreach (Peptide charPrec in characterizedPeptides[precursor].Values)
* {
* double cumulArea = 0.0;
* foreach (Dictionary<Peptide, ElutionCurve> ratios in listOfRatios)
* if (ratios.ContainsKey(charPrec))
* cumulArea += ratios[charPrec].Area;
* resultStr += "," + cumulArea;
* }
* }
* writerCumul.AddLine(resultStr);//*/
}
}
writerCumul.WriteToFile();
}
