public void TestBuild()
{
List <IIdentifiedSpectrum> spectra = new SequestPeptideTextFormat().ReadFromFile(TestContext.CurrentContext.TestDirectory + "/../../../data/TestBuilder.peptides");
Assert.AreEqual(4, spectra.Count);
IAccessNumberParser parser = AccessNumberParserFactory.FindOrCreateParser(@"(IPI\d+)", "IPI");
List <IIdentifiedProtein> proteins = new IdentifiedProteinBuilder().Build(spectra);
Assert.AreEqual(4, proteins.Count);
List <IIdentifiedProteinGroup> groups = new IdentifiedProteinGroupBuilder().Build(proteins);
Assert.AreEqual(2, groups.Count);
Assert.AreEqual(1, groups[0].Count);
Assert.AreEqual("IPI:IPI00784154.1|SW", groups[0][0].Name);
Assert.AreEqual(2, groups[1].Count);
Assert.AreEqual("REVERSED_00000001", groups[1][0].Name);
Assert.AreEqual("REVERSED_00000002", groups[1][1].Name);
IIdentifiedResult result = new IdentifiedResultBuilder(parser, "").Build(groups);
}
public void TestBuild()
{
var pep1 = new IdentifiedPeptide(new IdentifiedSpectrum(new SequestFilename("A", 1, 1, 1, ".dta")))
{
Sequence = "A"
};
var pep2 = new IdentifiedPeptide(new IdentifiedSpectrum(new SequestFilename("B", 1, 1, 1, ".dta")))
{
Sequence = "B"
};
var pep3 = new IdentifiedPeptide(new IdentifiedSpectrum(new SequestFilename("C", 1, 1, 1, ".dta")))
{
Sequence = "C"
};
var pep4 = new IdentifiedPeptide(new IdentifiedSpectrum(new SequestFilename("D", 1, 1, 1, ".dta")))
{
Sequence = "D"
};
var pep5 = new IdentifiedPeptide(new IdentifiedSpectrum(new SequestFilename("E", 1, 1, 1, ".dta")))
{
Sequence = "E"
};
var pep6 = new IdentifiedPeptide(new IdentifiedSpectrum(new SequestFilename("F", 1, 1, 1, ".dta")))
{
Sequence = "F"
};
var protein1 = new IdentifiedProtein()
{
Peptides = new IIdentifiedPeptide[] { pep1, pep3, pep5, pep6 }.ToList()
};
var protein2 = new IdentifiedProtein()
{
Peptides = new IIdentifiedPeptide[] { pep2, pep3, pep4 }.ToList()
};
//should be removed from final result since all peptides has been included in protein1 and protein2, even one protein contains both peptides
var protein3 = new IdentifiedProtein()
{
Peptides = new IIdentifiedPeptide[] { pep1, pep2 }.ToList()
};
//should be removed from final result since all peptides has been included in protein1
var protein4 = new IdentifiedProtein()
{
Peptides = new IIdentifiedPeptide[] { pep1, pep5 }.ToList()
};
var actual = new IdentifiedProteinGroupBuilder().Build(new IIdentifiedProtein[] { protein1, protein2, protein3 }.ToList());
Assert.AreEqual(2, actual.Count);
Assert.AreSame(protein1, actual[0][0]);
Assert.AreSame(protein2, actual[1][0]);
}
private void RunCurrentParameter(string parameterFile, List <string> result, BuildSummaryOptions conf)
{
IStringParser <string> acParser = conf.Database.GetAccessNumberParser();
IIdentifiedProteinBuilder proteinBuilder = new IdentifiedProteinBuilder();
IIdentifiedProteinGroupBuilder groupBuilder = new IdentifiedProteinGroupBuilder()
{
Progress = this.Progress
};
IdentifiedSpectrumBuilderResult isbr;
List <IIdentifiedSpectrum> finalPeptides;
if (string.IsNullOrEmpty(options.PeptideFile))
{ //parse from configuration
//build spectrum list
IIdentifiedSpectrumBuilder spectrumBuilder = conf.GetSpectrumBuilder();
if (spectrumBuilder is IProgress)
{
(spectrumBuilder as IProgress).Progress = this.Progress;
}
isbr = spectrumBuilder.Build(parameterFile);
finalPeptides = isbr.Spectra;
}
else
{
Progress.SetMessage("Reading peptides from {0} ...", options.PeptideFile);
finalPeptides = new MascotPeptideTextFormat().ReadFromFile(options.PeptideFile);
conf.SavePeptidesFile = false;
isbr = null;
}
CalculateIsoelectricPoint(finalPeptides);
//如果需要通过蛋白质注释去除contamination，首先需要在肽段水平删除
if (conf.Database.HasContaminationDescriptionFilter() && (conf.FalseDiscoveryRate.FdrLevel != FalseDiscoveryRateLevel.Protein))
{
Progress.SetMessage("Removing contamination by description ...");
var notConGroupFilter = conf.Database.GetNotContaminationDescriptionFilter(Progress);
var tempResultBuilder = new IdentifiedResultBuilder(null, null);
while (true)
{
List <IIdentifiedProtein> proteins = proteinBuilder.Build(finalPeptides);
List <IIdentifiedProteinGroup> groups = groupBuilder.Build(proteins);
IIdentifiedResult tmpResult = tempResultBuilder.Build(groups);
HashSet <IIdentifiedSpectrum> notConSpectra = new HashSet <IIdentifiedSpectrum>();
foreach (var group in tmpResult)
{
if (notConGroupFilter.Accept(group))
{
notConSpectra.UnionWith(group[0].GetSpectra());
}
}
if (notConSpectra.Count == finalPeptides.Count)
{
break;
}
finalPeptides = notConSpectra.ToList();
}
}
if (conf.FalseDiscoveryRate.FilterOneHitWonder && conf.FalseDiscoveryRate.MinOneHitWonderPeptideCount > 1)
{
Progress.SetMessage("Filtering single wonders ...");
var proteinFilter = new IdentifiedProteinSingleWonderPeptideCountFilter(conf.FalseDiscoveryRate.MinOneHitWonderPeptideCount);
List <IIdentifiedProtein> proteins = proteinBuilder.Build(finalPeptides);
int oldProteinCount = proteins.Count;
proteins.RemoveAll(l => !proteinFilter.Accept(l));
if (oldProteinCount != proteins.Count)
{
HashSet <IIdentifiedSpectrum> newspectra = new HashSet <IIdentifiedSpectrum>();
foreach (var protein in proteins)
{
newspectra.UnionWith(protein.GetSpectra());
}
finalPeptides = newspectra.ToList();
}
}
//if (conf.SavePeptidesFile && !(conf.FalseDiscoveryRate.FilterOneHitWonder && conf.FalseDiscoveryRate.MinOneHitWonderPeptideCount > 1))
if (conf.SavePeptidesFile)
{
if (conf.Database.RemovePeptideFromDecoyDB)
{
DecoyPeptideBuilder.AssignDecoy(finalPeptides, conf.GetDecoySpectrumFilter());
for (int i = finalPeptides.Count - 1; i >= 0; i--)
{
if (finalPeptides[i].FromDecoy)
{
finalPeptides.RemoveAt(i);
}
}
}
finalPeptides.Sort();
//保存肽段文件
IFileFormat <List <IIdentifiedSpectrum> > peptideFormat = conf.GetIdentifiedSpectrumFormat();
string peptideFile = FileUtils.ChangeExtension(parameterFile, ".peptides");
Progress.SetMessage("Writing peptides file...");
peptideFormat.WriteToFile(peptideFile, finalPeptides);
result.Add(peptideFile);
if (!conf.FalseDiscoveryRate.FilterByFdr && conf.Database.DecoyPatternDefined)
{
WriteFdrFile(parameterFile, conf, finalPeptides);
}
Progress.SetMessage("Calculating precursor offset...");
result.AddRange(new PrecursorOffsetCalculator(finalPeptides).Process(peptideFile));
}
Progress.SetMessage("Building protein...");
//构建蛋白质列表
List <IIdentifiedProtein> finalProteins = proteinBuilder.Build(finalPeptides);
Progress.SetMessage("Building protein group...");
//构建蛋白质群列表
List <IIdentifiedProteinGroup> finalGroups = groupBuilder.Build(finalProteins);
if (conf.Database.HasContaminationDescriptionFilter())
{
var notConGroupFilter = conf.Database.GetNotContaminationDescriptionFilter(Progress);
for (int i = finalGroups.Count - 1; i >= 0; i--)
{
if (!notConGroupFilter.Accept(finalGroups[i]))
{
finalGroups.RemoveAt(i);
}
}
}
//构建最终鉴定结果
var resultBuilder = conf.GetIdentifiedResultBuilder();
resultBuilder.Progress = Progress;
IIdentifiedResult finalResult = resultBuilder.Build(finalGroups);
finalResult.BuildGroupIndex();
if (conf.FalseDiscoveryRate.FilterByFdr)
{
var decoyGroupFilter = conf.GetDecoyGroupFilter();
foreach (var group in finalResult)
{
group.FromDecoy = decoyGroupFilter.Accept(group);
foreach (var protein in group)
{
protein.FromDecoy = group.FromDecoy;
}
}
finalResult.ProteinFDR = conf.FalseDiscoveryRate.GetFalseDiscoveryRateCalculator().Calculate(finalResult.Count(l => l[0].FromDecoy), finalResult.Count(l => !l[0].FromDecoy));
}
CalculateIsoelectricPoint(finalResult.GetProteins());
if (isbr != null)
{
finalResult.PeptideFDR = isbr.PeptideFDR;
}
//保存非冗余蛋白质列表文件
var resultFormat = conf.GetIdetifiedResultFormat(finalResult, this.Progress);
string noredundantFile = FileUtils.ChangeExtension(parameterFile, ".noredundant");
Progress.SetMessage("Writing noredundant file...");
resultFormat.WriteToFile(noredundantFile, finalResult);
result.Add(noredundantFile);
Progress.SetMessage("Finished!");
}