public static IIdentifiedResult DoBuildGroupByPeptide(List <IIdentifiedSpectrum> spectra, Func <IIdentifiedPeptide, string> func)
{
IdentifiedResult result = new IdentifiedResult();
var singlePeptides = (from s in spectra
where s.Peptides.Count == 1
select s).GroupBy(m => func(m.Peptide));
var multiplePeptides = (from s in spectra
where s.Peptides.Count > 1
select s).ToList();
Dictionary <string, List <IIdentifiedPeptide> > dic = new Dictionary <string, List <IIdentifiedPeptide> >();
foreach (var g in singlePeptides)
{
dic[g.Key] = new List <IIdentifiedPeptide>(from s in g
select s.Peptide);
}
foreach (var o in multiplePeptides)
{
var pc = (from p in o.Peptides
let c = dic.ContainsKey(func(p)) ? dic[func(p)].Count : 0
orderby c descending
select p).First();
if (!dic.ContainsKey(func(pc)))
{
dic[func(pc)] = new List <IIdentifiedPeptide>();
}
dic[func(pc)].Add(pc);
}
var keys = new List <string>(dic.Keys);
keys.Sort();
foreach (var key in keys)
{
IdentifiedProtein protein = new IdentifiedProtein(key);
protein.Peptides.AddRange(dic[key]);
protein.UniquePeptideCount = 1;
protein.Description = dic[key][0].Proteins.Merge('/');
IdentifiedProteinGroup group = new IdentifiedProteinGroup();
group.Add(protein);
result.Add(group);
}
result.BuildGroupIndex();
//result.Sort();
return(result);
}
public object Clone()
{
IdentifiedProteinGroup result = new IdentifiedProteinGroup();
this.ForEach(p => result.Add((IIdentifiedProtein)p.Clone()));
result.Index = this.Index;
result.Enabled = this.Enabled;
return(result);
}
public override IIdentifiedResult ReadFromFile(string fileName)
{
if (!File.Exists(fileName))
{
throw new FileNotFoundException("Protein file not exist : " + fileName);
}
string peptideFilename = GetPeptideFileName(fileName);
if (!File.Exists(peptideFilename))
{
throw new FileNotFoundException("Peptide file not exist : " + peptideFilename);
}
string linkFileName = GetLinkFileName(fileName);
if (!File.Exists(linkFileName))
{
throw new FileNotFoundException("Peptide2group file not exist : " + linkFileName);
}
var pepFileReader = new PeptideTextReader(GetEngineName());
List <IIdentifiedSpectrum> spectra = pepFileReader.ReadFromFile(peptideFilename);
this.PeptideFormat = pepFileReader.PeptideFormat;
var proFileReader = new ProteinTextReader(GetEngineName());
List <IIdentifiedProtein> proteins = proFileReader.ReadFromFile(fileName);
this.ProteinFormat = proFileReader.ProteinFormat;
var peptideMap = spectra.ToDictionary(m => m.Id);
var proteinMap = proteins.GroupBy(m => m.GroupIndex);
IIdentifiedResult result = Allocate();
foreach (var pros in proteinMap)
{
var group = new IdentifiedProteinGroup();
pros.ToList().ForEach(m => group.Add(m));
result.Add(group);
}
new Peptide2GroupTextReader().LinkPeptideToGroup(linkFileName, peptideMap, result.ToDictionary(m => m.Index));
string fastaFile = fileName + ".fasta";
if (File.Exists(fastaFile))
{
IdentifiedResultUtils.FillSequenceFromFasta(fastaFile, result, null);
}
return(result);
}
public void Run()
{
IdentifiedProteinGroupContaminationDescriptionFilter filter = new IdentifiedProteinGroupContaminationDescriptionFilter("KERATIN");
IdentifiedProteinGroup group = new IdentifiedProteinGroup();
group.Add(new IdentifiedProtein("P1")
{
Description = "P1 Keratin"
});
Assert.IsTrue(filter.Accept(group));
}
public void TestFilter2()
{
var spectrum = new IdentifiedSpectrum();
spectrum.Query.FileScan.LongFileName = "ABDCDD.12.123.2.dat";
var pro1 = new IdentifiedProtein("P1");
pro1.Peptides.Add(new IdentifiedPeptide(spectrum)
{
Sequence = "AAAAAAA"
});
var pro2 = new IdentifiedProtein("P2");
pro2.Peptides.Add(new IdentifiedPeptide(spectrum)
{
Sequence = "BBBBBBB"
});
var g1 = new IdentifiedProteinGroup();
g1.Add(pro1);
g1.Add(pro2);
IdentifiedResult ir = new IdentifiedResult();
ir.Add(g1);
Assert.AreEqual(1, ir.Count);
Assert.AreEqual(2, ir[0].Count);
Assert.AreEqual(1, ir.GetSpectra().Count);
ir.Filter(m =>
{
return(m.Sequence.Contains('A'));
});
Assert.AreEqual(1, ir.Count);
Assert.AreEqual(1, ir[0].Count);
Assert.AreEqual(1, ir.GetSpectra().Count);
Assert.AreSame(pro1, ir[0][0]);
ir.Filter(m =>
{
return(m.Sequence.Contains('C'));
});
Assert.AreEqual(0, ir.Count);
}
public void TestFilter()
{
var pro1 = new IdentifiedProtein("P1");
pro1.Peptides.Add(new IdentifiedPeptide(new IdentifiedSpectrum()
{
Charge = 1
}));
pro1.Peptides.Add(new IdentifiedPeptide(new IdentifiedSpectrum()
{
Charge = 2
}));
var pro2 = new IdentifiedProtein("P2");
pro2.Peptides.Add(new IdentifiedPeptide(new IdentifiedSpectrum()
{
Charge = 3
}));
var g1 = new IdentifiedProteinGroup();
g1.Add(pro1);
var g2 = new IdentifiedProteinGroup();
g2.Add(pro2);
IdentifiedResult ir = new IdentifiedResult();
ir.Add(g1);
ir.Add(g2);
Assert.AreEqual(2, ir.Count);
Assert.AreEqual(3, ir.GetSpectra().Count);
ir.Filter(m => { return(m.Spectrum.Query.Charge > 1); });
Assert.AreEqual(2, ir.Count);
Assert.AreEqual(2, ir.GetSpectra().Count);
ir.GetSpectra().All(m => { return(m.Charge > 1); });
ir.Filter(m => { return(m.Spectrum.Query.Charge > 2); });
Assert.AreEqual(1, ir.Count);
Assert.AreEqual(1, ir.GetSpectra().Count);
ir.GetSpectra().All(m => { return(m.Charge > 2); });
Assert.AreEqual("P2", ir[0][0].Name);
}
public int CompareTo(IdentifiedProteinGroup other)
{
return(CompareTo(other as IIdentifiedProteinGroup));
}
public List <IIdentifiedProteinGroup> Build(List <IIdentifiedProtein> proteins)
{
var result = new List <IIdentifiedProteinGroup>();
Progress.SetMessage("Initializing protein group/spectra map ...");
var groupMap = new Dictionary <IIdentifiedProteinGroup, HashSet <IIdentifiedSpectrum> >();
foreach (IIdentifiedProtein protein in proteins)
{
IIdentifiedProteinGroup group = new IdentifiedProteinGroup();
group.Add(protein);
var spectraSet = new HashSet <IIdentifiedSpectrum>(protein.GetSpectra());
groupMap[group] = spectraSet;
result.Add(group);
}
Progress.SetMessage("Sorting protein groups ...");
result.Sort((m1, m2) =>
{
int ret = -m1[0].PeptideCount.CompareTo(m2[0].PeptideCount);
if (ret == 0)
{
ret = -m1[0].UniquePeptideCount.CompareTo(m2[0].UniquePeptideCount);
}
return(ret);
});
Progress.SetMessage("Merging proteins with same peptide-spectrum matches ...");
Progress.SetRange(0, result.Count);
//首先合并所有内容相同的group
for (int i = result.Count - 1; i > 0; i--)
{
Progress.SetPosition(result.Count - i);
HashSet <IIdentifiedSpectrum> iSpectra = groupMap[result[i]];
for (int j = i - 1; j >= 0; j--)
{
if (result[j][0].PeptideCount == result[i][0].PeptideCount &&
result[j][0].UniquePeptideCount == result[i][0].UniquePeptideCount)
{
HashSet <IIdentifiedSpectrum> jSpectra = groupMap[result[j]];
if (jSpectra.SetEquals(iSpectra))
{
//如果内容一致，则合并两个group
foreach (IIdentifiedProtein protein in result[i])
{
result[j].Add(protein);
}
//删除group i
result.RemoveAt(i);
break;
}
}
else
{
break;
}
}
}
Progress.SetMessage("Initializing peptide group count ...");
InitializePeptideGroupCount(result);
Progress.SetMessage("Extracting distinct protein groups ...");
var temp = result;
result = new List <IIdentifiedProteinGroup>();
for (int i = temp.Count - 1; i > 0; i--)
{
if (temp[i].GetPeptides().All(m => m.GroupCount == 1))
{
result.Add(temp[i]);
temp.RemoveAt(i);
}
}
Progress.SetMessage("There are {0} distinct and {1} undistinct protein groups. ", result.Count, temp.Count);
Progress.SetMessage("Removing redundant protein groups from undistinct protein groups...");
var oldcount = temp.Count;
Progress.SetRange(0, oldcount);
//删除被包含的group
for (int i = temp.Count - 1; i > 0; i--)
{
Progress.SetPosition(oldcount - i);
HashSet <IIdentifiedSpectrum> iSpectra = groupMap[temp[i]];
for (int j = i - 1; j >= 0; j--)
{
HashSet <IIdentifiedSpectrum> jSpectra = groupMap[temp[j]];
if (jSpectra.Count == iSpectra.Count)
{
continue;
}
if (iSpectra.All(l => jSpectra.Contains(l)))
{
//删除group i
temp.RemoveAt(i);
break;
}
}
}
RemoveUndistinctProteinGroups(temp);
result.AddRange(temp);
Progress.SetMessage("Sorting proteins in group ...");
result.ForEach(m => m.SortByProteinName());
Progress.SetMessage("Building protein groups done.");
return(result);
}
private IIdentifiedProteinGroup ReadNextProteinGroup(StreamReader filein, Dictionary <string, IIdentifiedSpectrum> peptideMap, ref string lastLine)
{
Progress.SetPosition(filein.BaseStream.Position);
while (!IdentifiedResultUtils.IsProteinLine(lastLine) && (lastLine = filein.ReadLine()) != null)
{
}
if (lastLine == null)
{
return(null);
}
IIdentifiedProteinGroup result = new IdentifiedProteinGroup();
while (IdentifiedResultUtils.IsProteinLine(lastLine))
{
IIdentifiedProtein protein = ProteinFormat.ParseString(lastLine);
result.Add(protein);
protein.GroupIndex = IdentifiedResultUtils.GetGroupIndex(lastLine);
lastLine = filein.ReadLine();
}
List <IIdentifiedSpectrum> peptides = new List <IIdentifiedSpectrum>();
while (!IdentifiedResultUtils.IsProteinLine(lastLine))
{
if (Progress.IsCancellationPending())
{
throw new UserTerminatedException();
}
IIdentifiedSpectrum mphit = PeptideFormat.ParseString(lastLine);
string id = string.Format("{0}-{1}-{2}-{3}", mphit.Query.FileScan.LongFileName, mphit.Rank, mphit.Engine, mphit.Tag);
if (!peptideMap.ContainsKey(id))
{
peptideMap[id] = mphit;
}
else
{
mphit = peptideMap[id];
}
peptides.Add(mphit);
lastLine = filein.ReadLine();
if (lastLine == null || lastLine.Trim().Length == 0)
{
break;
}
}
foreach (IIdentifiedSpectrum hit in peptides)
{
result.AddIdentifiedSpectrum(hit);
}
return(result);
}
