private string Peptides2Html( Protein p ) {
if( p.Peptides.Count == 0 )
return "";
int i;
Tag a = new Tag( "a", "href" );
string str = "";
for( i = 0; i < p.Peptides.Count-1; i++ )
str += a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString()) + ", ";
str += a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString());
return str;
}
/// <summary>
/// Gets the positions of the peptide in the specified protein.
/// </summary>
public string GetPositions( Protein p ) {
if( Sequence == null || Sequence.Length == 0 || p.Sequence == null || p.Sequence.Length == 0 )
return "";
string str = "";
int pos = -1;
while( (pos=p.Sequence.IndexOf(Sequence,pos+1)) != -1 )
str += (pos+1) + ", ";
if( (pos=str.LastIndexOf(',')) != -1 )
return str.Remove( pos );
return str;
}
private void SaveCSVEntry( TextWriter w, Protein p, string grp, string sep ) {
w.Write( p.ID + sep + p.Entry + sep + p.Accession + sep + p.Evidence.ToString() + sep + grp + sep + p.Desc + sep );
foreach( Peptide f in p.Peptides )
w.Write(f.ToString() + ' ');
w.Write( sep + p.Sequence + sep );
String ptm;
foreach( Peptide peptide in p.Peptides ) {
ptm = GetPtmCsv(peptide);
if( ptm.Length != 0 )
w.Write(ptm+" ");
}
w.WriteLine();
}
private SortedList<string,string> LoadProteins() {
SortedList<string,string> SortedAccession = new SortedList<string, string>();
foreach( DBSequenceType prot in m_mzid.ListProteins ) {
if( SortedAccession.ContainsKey(prot.id) ) // Avoids duplicated entries in the same file
continue;
SortedAccession.Add( prot.id, prot.accession );
if( m_SortedProteins.ContainsKey(prot.accession) ) // Avoids duplicated entries between different files
continue;
CVParamType cv = mzidFile1_1.FindCV("MS:1001352", prot.Items);
string entry = cv == null ? "" : cv.value;
cv = mzidFile1_1.FindCV("MS:1001088", prot.Items);
string desc = cv == null ? "" : cv.value;
Protein p = new Protein( m_pid++, entry, prot.accession, desc, prot.Seq );
p.DBRef = prot.id;
Proteins.Add( p );
m_SortedProteins.Add( p.Accession, p );
}
return SortedAccession;
}
private bool IsIndistinguisable( Protein g ) {
List<Peptide> discriminating = new List<Peptide>();
foreach( Protein prot in g.Subset )
foreach( Peptide pep in prot.Peptides )
if( pep.Relation == Peptide.RelationType.Discriminating )
discriminating.Add(pep);
foreach( Protein prot in g.Subset )
foreach( Peptide pep in discriminating )
if( !prot.Peptides.Contains(pep) )
return false;
return true;
/*foreach( Peptide f in g.Subset[0].Peptides ) {
if( f.Relation != Peptide.RelationType.Discriminating )
continue;
foreach( Protein p in g.Subset )
if( !p.HasPeptide(f) )
return false;
}
return true;*/
}
/// <summary>
/// Builds a PDH for the current protein
/// </summary>
private PSIPIanalysisprocessProteinDetectionHypothesisType BuildHypothesis( Protein p, Protein.EvidenceType ev ) {
PSIPIanalysisprocessProteinDetectionHypothesisType h = new PSIPIanalysisprocessProteinDetectionHypothesisType();
h.id = "PDH_" + p.Accession;
h.DBSequence_ref = p.DBRef;
if( p.Evidence == Protein.EvidenceType.NonConclusive || p.Evidence == Protein.EvidenceType.Filtered )
h.passThreshold = false;
else
h.passThreshold = true;
List<PeptideHypothesisType> listPeptides = new List<PeptideHypothesisType>();
foreach( Peptide f in p.Peptides ) {
PeptideHypothesisType peptide = new PeptideHypothesisType();
peptide.PeptideEvidence_Ref = f.Names[p.DBRef];
listPeptides.Add( peptide );
}
if( listPeptides.Count > 0 )
h.PeptideHypothesis = listPeptides.ToArray();
h.cvParam = new FuGECommonOntologycvParamType[1];
h.cvParam[0] = new FuGECommonOntologycvParamType(
"Protein Inference Confidence Category",
"MS:1001600", "PSI-MS" );
h.cvParam[0].value = ParseConfidence( ev );
return h;
}
/// <summary>
/// Gets the positions of the peptide in the specified protein.
/// </summary>
public string GetPositions( Protein p )
{
if( Sequence == null || Sequence.Length == 0 || p.Sequence == null || p.Sequence.Length == 0 )
return "";
string str = "";
int pos = -1;
while( (pos=p.Sequence.IndexOf(Sequence,pos+1)) != -1 )
str += (pos+1) + ", ";
if( (pos=str.LastIndexOf(',')) != -1 )
return str.Remove( pos );
return str;
}
/// <summary>
/// Builds a PDH for the current protein
/// </summary>
private PSIPIanalysisprocessProteinDetectionHypothesisType BuildHypothesis( Protein p, Protein.EvidenceType ev )
{
PSIPIanalysisprocessProteinDetectionHypothesisType h = new PSIPIanalysisprocessProteinDetectionHypothesisType();
h.id = "PDH_" + p.Accession;
h.DBSequence_ref = p.DBRef;
if( p.Evidence == Protein.EvidenceType.NonConclusive || p.Evidence == Protein.EvidenceType.Filtered )
h.passThreshold = false;
else
h.passThreshold = true;
List<PeptideHypothesisType> listPeptides = new List<PeptideHypothesisType>();
foreach( Peptide f in p.Peptides ) {
PeptideHypothesisType peptide = new PeptideHypothesisType();
peptide.PeptideEvidence_Ref = f.Names[p.DBRef];
listPeptides.Add( peptide );
}
if( listPeptides.Count > 0 )
h.PeptideHypothesis = listPeptides.ToArray();
h.cvParam = new FuGECommonOntologycvParamType[1];
h.cvParam[0] = new FuGECommonOntologycvParamType(
"Protein Inference Confidence Category",
"MS:1001600", "PSI-MS" );
h.cvParam[0].value = ParseConfidence( ev );
return h;
}
/// <summary>
/// Loads a mzIdentML file
/// </summary>
protected override void Load( string mzid )
{
m_mzid = new mzidFile1_0();
m_mzid.Load( mzid );
// Proteins
SortedList<string,string> SortedAccession = new SortedList<string, string>();
foreach( PSIPIanalysissearchDBSequenceType element in m_mzid.ListProteins ) {
if( SortedAccession.ContainsKey(element.id) )
continue;
string acc = element.accession;
SortedAccession.Add( element.id, acc );
if( m_SortedProteins.ContainsKey(acc) )
continue;
FuGECommonOntologycvParamType cv;
cv = FuGECommonOntologycvParamType.Find( "MS:1001352", element.cvParam );
string entry = cv == null ? "" : cv.value;
cv = FuGECommonOntologycvParamType.Find( "MS:1001088", element.cvParam );
string desc = cv == null ? "" : cv.value;
string seq = element.seq;//.ToUpper();
Protein p = new Protein(m_pid++, entry, acc, desc, seq);
p.DBRef = element.id;
Proteins.Add( p );
m_SortedProteins.Add( acc, p );
}
// Peptides
SortedList<string,Peptide> SortedPeptides = new SortedList<string, Peptide>();
int id = 1;
foreach( PSIPIpolypeptidePeptideType element in m_mzid.ListPeptides ) {
string seq = element.peptideSequence;//.ToUpper();
Peptide f = new Peptide(id++, seq);
f.Confidence = Peptide.ConfidenceType.PassThreshold; // It will be filtered later if neccessary
SortedPeptides.Add( element.id, f );
f.Runs.Add( m_Run );
if( element.Modification != null )
foreach( PSIPIpolypeptideModificationType mod in element.Modification ) {
PTM ptm = new PTM();
ptm.Pos = mod.locationSpecified ? mod.location : -1;
if( mod.residues != null )
foreach( string residue in mod.residues )
ptm.Residues += residue;
foreach( FuGECommonOntologycvParamType param in mod.cvParam )
if( param.cvRef.Equals("UNIMOD") )
ptm.Name = param.name;
f.AddPTM( ptm );
}
Peptides.Add( f );
}
// Relations
if( m_mzid.Data.DataCollection.AnalysisData.SpectrumIdentificationList.Length != 1 )
throw new ApplicationException( "Multiple spectrum identification lists not supported" );
foreach( PSIPIanalysissearchSpectrumIdentificationResultType idres in
m_mzid.Data.DataCollection.AnalysisData.SpectrumIdentificationList[0].SpectrumIdentificationResult )
foreach( PSIPIanalysissearchSpectrumIdentificationItemType item in idres.SpectrumIdentificationItem ) {
if( !item.passThreshold )
continue;
Peptide f = SortedPeptides[item.Peptide_ref];
if( item.PeptideEvidence == null )
continue;
f.Confidence = Peptide.ConfidenceType.PassThreshold;
foreach( PSIPIanalysisprocessPeptideEvidenceType relation in item.PeptideEvidence ) {
Protein p = m_SortedProteins[SortedAccession[relation.DBSequence_Ref]];
if( f.Proteins.Contains(p) )
continue;
f.Names.Add( relation.DBSequence_Ref, relation.id );
p.Peptides.Add( f );
f.Proteins.Add( p );
}
}
}
private void WriteProteinList( TextWriter w, Tag tr, Protein.EvidenceType evidence )
{
Tag a = new Tag( evidence == Protein.EvidenceType.Filtered ? null : "a", "href" );
Tag td = new Tag( "td" );
Tag tdr = new Tag( "td", "rowspan" );
Tag tdc = new Tag( "td", "colspan" );
foreach( Protein p in Proteins ) {
if( p.Evidence != evidence )
continue;
if( p.Subset.Count == 0 ) {
w.Write( tr.Render(
td.Render(a.Render("#"+p.Accession,p.EntryEx))+
td.Render(p.Evidence.ToString())+
tdc.Render("2",Peptides2Html(p))+
td.Render(p.Desc) ));
continue;
}
w.WriteLine( tr.Render(
tdr.Render(p.Subset.Count.ToString(),p.Entry)+
tdr.Render(p.Subset.Count.ToString(),p.Evidence.ToString())+
td.Render(a.Render("#"+p.Subset[0].Accession,p.Subset[0].EntryEx)+": ")+
td.Render(Peptides2Html(p.Subset[0]))+
td.Render(p.Subset[0].Desc) ));
tr.Hold = true;
for( int i = 1; i < p.Subset.Count; i++ )
w.WriteLine( tr.Render(
td.Render(a.Render("#"+p.Subset[i].Accession,p.Subset[i].EntryEx)+": ")+
td.Render(Peptides2Html(p.Subset[i]))+
td.Render(p.Subset[i].Desc) ));
tr.Hold = false;
}
}
private void WriteProteinDetails( TextWriter w, Protein p )
{
Tag tr = new Tag( "tr", true );
Tag td = new Tag( "td", "colspan" );
Tag th = new Tag( "th", "rowspan" );
Tag a = new Tag( "a", "href" );
int i;
w.WriteLine( "<table>\n<caption><a name=\""+p.Accession+"\"/>Protein "+p.Accession+"</caption>" );
//w.WriteLine( "<col width=\"10%\"/><col width=\"5%\"/><col width=\"10%\"/><col width=\"75%\"/>" );
w.WriteLine( tr.Render(th.Render("Name")+td.Render("3",p.EntryEx)) );
w.WriteLine( tr.Render(th.Render("Description")+td.Render("3",p.Desc)) );
w.WriteLine( tr.Render(th.Render("Sequence")+td.Render("3","<pre>"+p.ParseSeq(10)+"</pre>")) );
w.WriteLine( tr.Render(th.Render("Evidence")+td.Render("3",p.Evidence.ToString())) );
w.Write( tr+th.Render("Peptide list")+"<td colspan=\"3\">" );
if( p.Peptides.Count > 0 ) {
for( i = 0; i < p.Peptides.Count-1; i++ )
w.Write( a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString()) + ", " );
w.Write( a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString()) );
}
w.WriteLine( "</td>"+tr );
if( p.Peptides.Count == 0 ) {
w.WriteLine( "</table><br/>" );
return;
}
int rows = 8;
if( Spectra.Count != 0 )
rows++;
w.Write( tr+th.Render((p.Peptides.Count*rows).ToString(),"Peptides") );
bool first = true;
foreach( Peptide f in p.Peptides ) {
if( first )
first = false;
else
w.Write( tr.ToString() );
w.Write( th.Render(rows.ToString(),f.ToString()) );
w.WriteLine( th.Render("<a name=\""+p.Accession+"__"+f.ID+"\"/>Confidence")+td.Render(f.Confidence.ToString())+tr );
tr.Hold = true;
w.WriteLine( tr.Render(th.Render("Decoy")+td.Render(f.Decoy.ToString())) );
w.Write( tr+th.Render("Runs")+td );
for( i = 0; i < f.Runs.Count-1; i++ )
w.Write( f.Runs[i].ToString() + ", " );
w.WriteLine( f.Runs[i].ToString()+td+tr );
w.WriteLine( tr.Render(th.Render("Relation")+td.Render(f.Relation.ToString())) );
w.Write( tr+th.Render("Proteins")+td );
for( i = 0; i < f.Proteins.Count-1; i++ )
w.Write( a.Render("#"+f.Proteins[i].Accession,f.Proteins[i].EntryEx) + ", " );
w.WriteLine( a.Render("#"+f.Proteins[i].Accession,f.Proteins[i].EntryEx)+td.ToString()+tr );
w.WriteLine( tr.Render(th.Render("Sequence")+td.Render("<pre>"+f.Sequence+"</pre>")) );
w.WriteLine( tr.Render(th.Render("Position")+td.Render(f.GetPositions(p))) );
w.Write( tr+th.Render("PTMs")+td );
if( f.Variants.Count == 1 )
w.Write( Peptide.Variant2Str(f.LastVariant) );
else {
i = 1;
foreach( List<PTM> v in f.Variants )
w.Write( "Variant #"+(i++)+": "+Peptide.Variant2Str(v)+"<br/>" );
}
w.WriteLine( td.ToString()+tr );
if( Spectra.Count != 0 ) {
w.Write( tr+th.Render("PSMs")+td );
if( f.Psm != null ) {
for( i = 0; i < f.Psm.Count-1; i++ )
w.Write( a.Render("#PSM"+f.Psm[i].ID,f.Psm[i].ID.ToString()) + ", " );
w.WriteLine( a.Render("#PSM"+f.Psm[i].ID,f.Psm[i].ID.ToString()) );
}
w.WriteLine( td.ToString()+tr );
}
tr.Hold = false;
}
w.WriteLine( "</table><br/>" );
}
private void WriteProteinDetails( TextWriter w, Protein.EvidenceType evidence )
{
foreach( Protein p in Proteins ) {
if( p.Evidence != evidence )
continue;
if( p.Subset.Count == 0 ) {
WriteProteinDetails( w, p );
continue;
}
foreach( Protein p2 in p.Subset )
WriteProteinDetails( w, p2 );
}
}
private SortedList<string, string> LoadProteins()
{
SortedList<string,string> SortedAccession = new SortedList<string, string>();
foreach( DBSequenceType prot in m_mzid.ListProteins ) {
if( SortedAccession.ContainsKey(prot.id) ) // Avoids duplicated entries in the same file
continue;
SortedAccession.Add( prot.id, prot.accession );
if( m_SortedProteins.ContainsKey(prot.accession) ) // Avoids duplicated entries between different files
continue;
CVParamType cv = mzidFile1_1.FindCV("MS:1001352", prot.Items);
string entry = cv == null ? "" : cv.value;
cv = mzidFile1_1.FindCV("MS:1001088", prot.Items);
string desc = cv == null ? "" : cv.value;
Protein p = new Protein( m_pid++, entry, prot.accession, desc, prot.Seq );
p.DBRef = prot.id;
Proteins.Add( p );
m_SortedProteins.Add( p.Accession, p );
}
return SortedAccession;
}
private void WriteProteinDetails( TextWriter w, Protein p ) {
Tag tr = new Tag( "tr", true );
Tag td = new Tag( "td", "colspan" );
Tag th = new Tag( "th", "rowspan" );
Tag a = new Tag( "a", "href" );
int i;
w.WriteLine( "<table>\n<caption><a name=\""+p.Accession+"\"/>Protein "+p.Accession+"</caption>" );
//w.WriteLine( "<col width=\"10%\"/><col width=\"5%\"/><col width=\"10%\"/><col width=\"75%\"/>" );
w.WriteLine( tr.Render(th.Render("Name")+td.Render("3",p.EntryEx)) );
w.WriteLine( tr.Render(th.Render("Description")+td.Render("3",p.Desc)) );
w.WriteLine( tr.Render(th.Render("Sequence")+td.Render("3","<pre>"+p.ParseSeq(10)+"</pre>")) );
w.WriteLine( tr.Render(th.Render("Evidence")+td.Render("3",p.Evidence.ToString())) );
w.Write( tr+th.Render("Peptide list")+"<td colspan=\"3\">" );
if( p.Peptides.Count > 0 ) {
for( i = 0; i < p.Peptides.Count-1; i++ )
w.Write( a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString()) + ", " );
w.Write( a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString()) );
}
w.WriteLine( "</td>"+tr );
if( p.Peptides.Count == 0 ) {
w.WriteLine( "</table><br/>" );
return;
}
int rows = 8;
if( Spectra.Count != 0 )
rows++;
w.Write( tr+th.Render((p.Peptides.Count*rows).ToString(),"Peptides") );
bool first = true;
foreach( Peptide f in p.Peptides ) {
if( first )
first = false;
else
w.Write( tr.ToString() );
w.Write( th.Render(rows.ToString(),f.ToString()) );
w.WriteLine( th.Render("<a name=\""+p.Accession+"__"+f.ID+"\"/>Confidence")+td.Render(f.Confidence.ToString())+tr );
tr.Hold = true;
w.WriteLine( tr.Render(th.Render("Decoy")+td.Render(f.Decoy.ToString())) );
w.Write( tr+th.Render("Runs")+td );
for( i = 0; i < f.Runs.Count-1; i++ )
w.Write( f.Runs[i].ToString() + ", " );
w.WriteLine( f.Runs[i].ToString()+td+tr );
w.WriteLine( tr.Render(th.Render("Relation")+td.Render(f.Relation.ToString())) );
w.Write( tr+th.Render("Proteins")+td );
for( i = 0; i < f.Proteins.Count-1; i++ )
w.Write( a.Render("#"+f.Proteins[i].Accession,f.Proteins[i].EntryEx) + ", " );
w.WriteLine( a.Render("#"+f.Proteins[i].Accession,f.Proteins[i].EntryEx)+td.ToString()+tr );
w.WriteLine( tr.Render(th.Render("Sequence")+td.Render("<pre>"+f.Sequence+"</pre>")) );
w.WriteLine( tr.Render(th.Render("Position")+td.Render(f.GetPositions(p))) );
w.Write( tr+th.Render("PTMs")+td );
if( f.Variants.Count == 1 )
w.Write( Peptide.Variant2Str(f.LastVariant) );
else {
i = 1;
foreach( List<PTM> v in f.Variants )
w.Write( "Variant #"+(i++)+": "+Peptide.Variant2Str(v)+"<br/>" );
}
w.WriteLine( td.ToString()+tr );
if( Spectra.Count != 0 ) {
w.Write( tr+th.Render("PSMs")+td );
if( f.Psm != null ) {
for( i = 0; i < f.Psm.Count-1; i++ )
w.Write( a.Render("#PSM"+f.Psm[i].ID,f.Psm[i].ID.ToString()) + ", " );
w.WriteLine( a.Render("#PSM"+f.Psm[i].ID,f.Psm[i].ID.ToString()) );
}
w.WriteLine( td.ToString()+tr );
}
tr.Hold = false;
}
w.WriteLine( "</table><br/>" );
}
private void LoadData( string xmlpath ) {
SortedList<int,string> SortedAccession = new SortedList<int, string>();
XmlDocument doc = new XmlDocument();
doc.Load( xmlpath );
// Load proteins
XmlNodeList proteins = doc.GetElementsByTagName( "PROTEIN" );
foreach( XmlElement element in proteins ) {
int id = int.Parse(element.GetAttribute("ID"));
if( SortedAccession.ContainsKey(id) )
continue;
string acc = element.GetElementsByTagName("ACCESSION")[0].InnerText;
SortedAccession.Add( id, acc );
if( m_SortedProteins.ContainsKey(acc) )
continue;
string entry = element.GetElementsByTagName("ENTRY")[0].InnerText;
string desc = element.GetElementsByTagName("DESCRIPTION")[0].InnerText.Replace('+',' ');
string seq = element.GetElementsByTagName("SEQUENCE")[0].InnerText.ToUpper();
Protein p = new Protein(m_pid++, entry, acc, desc, seq);
Proteins.Add( p );
m_SortedProteins.Add( acc, p );
}
// Load peptides
SortedList<int,Peptide> SortedPeptides = new SortedList<int, Peptide>();
XmlNodeList peptides = doc.GetElementsByTagName( "PEPTIDE" );
foreach( XmlElement element in peptides ) {
int id = int.Parse(element.GetAttribute("ID"));
int pid = int.Parse(element.GetAttribute("PROT_ID"));
int mid = 0;
if( PlgsThreshold != Peptide.ConfidenceType.NoThreshold )
mid = int.Parse(element.GetAttribute("QUERY_MASS_ID"));
string seq = element.GetAttribute("SEQUENCE").ToUpper();
Peptide f = new Peptide(id, seq);
XmlNodeList mods = element.GetElementsByTagName( "MATCH_MODIFIER" );
f.Runs.Add( m_Run );
foreach( XmlElement mod in mods ) {
PTM ptm = new PTM();
string[] strs = mod.GetAttribute("NAME").Split(new char[]{'+'});
ptm.Name = strs[0];
string str = mod.GetAttribute("POS");
ptm.Pos = str.Length == 0 ? -1 : int.Parse(str);
if( strs.Length > 1 )
ptm.Residues = strs[1];
f.AddPTM( ptm );
}
Protein p = null;
try {
p = m_SortedProteins[SortedAccession[pid]];
} catch {
Notify( "Peptide '" + id + "' references unknown protein '" + pid + "'" );
}
if( p != null ) {
p.Peptides.Add( f );
f.Proteins.Add( p );
if( !p.Sequence.Contains(f.Sequence) )
throw new ApplicationException( "Inconsistent sequence data" );
}
Peptides.Add( f );
if( PlgsThreshold != Peptide.ConfidenceType.NoThreshold )
SortedPeptides.Add(mid,f);
}
if( PlgsThreshold == Peptide.ConfidenceType.NoThreshold )
return;
// Scores
XmlNodeList scores = doc.GetElementsByTagName( "MASS_MATCH" );
foreach( XmlElement element in scores ) {
int id = int.Parse(element.GetAttribute("ID"));
double score = double.Parse(element.GetAttribute("SCORE"), m_Format);
SortedPeptides[id].Score = score;
}
}
/// <summary>
/// Includes a protein in a group
/// </summary>
private void AddToGroup( Protein p, ref int id, ref List<Protein> proteins )
{
Protein g;
if( p.Group == null ) {
g = new Protein( id++, String.Format("GROUP{0:000}",m_gid++), "", p.Name, "" );
g.Evidence = Protein.EvidenceType.Group;
p.Group = g;
p.ID = id++;
g.Subset.Add( p );
proteins.Add( g );
} else
g = p.Group;
foreach( Peptide f in p.Peptides ) {
if( f.Relation != Peptide.RelationType.Discriminating )
continue;
foreach( Protein t in f.Proteins )
if( t.Evidence == Protein.EvidenceType.Group && t.Group == null ) {
t.ID = id++;
t.Group = g;
g.Subset.Add( t );
g.Desc += " + " + t.Name;
}
}
}
/// <summary>
/// Parses the confidence enum to a string.
/// </summary>
public string ParseConfidence( Protein.EvidenceType e )
{
switch( e ) {
case Protein.EvidenceType.Conclusive:
return "conclusive";
case Protein.EvidenceType.Group:
return "ambiguous group";
case Protein.EvidenceType.Indistinguishable:
return "indistinguishable";
case Protein.EvidenceType.NonConclusive:
return "non conclusive";
}
return e.ToString();
}
/// <summary>
/// Loads a mzIdentML file
/// </summary>
override protected void Load( string mzid ) {
m_mzid = new mzidFile1_0();
m_mzid.Load( mzid );
// Proteins
SortedList<string,string> SortedAccession = new SortedList<string, string>();
foreach( PSIPIanalysissearchDBSequenceType element in m_mzid.ListProteins ) {
if( SortedAccession.ContainsKey(element.id) )
continue;
string acc = element.accession;
SortedAccession.Add( element.id, acc );
if( m_SortedProteins.ContainsKey(acc) )
continue;
FuGECommonOntologycvParamType cv;
cv = FuGECommonOntologycvParamType.Find( "MS:1001352", element.cvParam );
string entry = cv == null ? "" : cv.value;
cv = FuGECommonOntologycvParamType.Find( "MS:1001088", element.cvParam );
string desc = cv == null ? "" : cv.value;
string seq = element.seq;//.ToUpper();
Protein p = new Protein(m_pid++, entry, acc, desc, seq);
p.DBRef = element.id;
Proteins.Add( p );
m_SortedProteins.Add( acc, p );
}
// Peptides
SortedList<string,Peptide> SortedPeptides = new SortedList<string, Peptide>();
int id = 1;
foreach( PSIPIpolypeptidePeptideType element in m_mzid.ListPeptides ) {
string seq = element.peptideSequence;//.ToUpper();
Peptide f = new Peptide(id++, seq);
f.Confidence = Peptide.ConfidenceType.PassThreshold; // It will be filtered later if neccessary
SortedPeptides.Add( element.id, f );
f.Runs.Add( m_Run );
if( element.Modification != null )
foreach( PSIPIpolypeptideModificationType mod in element.Modification ) {
PTM ptm = new PTM();
ptm.Pos = mod.locationSpecified ? mod.location : -1;
if( mod.residues != null )
foreach( string residue in mod.residues )
ptm.Residues += residue;
foreach( FuGECommonOntologycvParamType param in mod.cvParam )
if( param.cvRef.Equals("UNIMOD") )
ptm.Name = param.name;
f.AddPTM( ptm );
}
Peptides.Add( f );
}
// Relations
if( m_mzid.Data.DataCollection.AnalysisData.SpectrumIdentificationList.Length != 1 )
throw new ApplicationException( "Multiple spectrum identification lists not supported" );
foreach( PSIPIanalysissearchSpectrumIdentificationResultType idres in
m_mzid.Data.DataCollection.AnalysisData.SpectrumIdentificationList[0].SpectrumIdentificationResult )
foreach( PSIPIanalysissearchSpectrumIdentificationItemType item in idres.SpectrumIdentificationItem ) {
if( !item.passThreshold )
continue;
Peptide f = SortedPeptides[item.Peptide_ref];
if( item.PeptideEvidence == null )
continue;
f.Confidence = Peptide.ConfidenceType.PassThreshold;
foreach( PSIPIanalysisprocessPeptideEvidenceType relation in item.PeptideEvidence ) {
Protein p = m_SortedProteins[SortedAccession[relation.DBSequence_Ref]];
if( f.Proteins.Contains(p) )
continue;
f.Names.Add( relation.DBSequence_Ref, relation.id );
p.Peptides.Add( f );
f.Proteins.Add( p );
}
}
}
private void LoadData( string xmlpath )
{
SortedList<int,string> SortedAccession = new SortedList<int, string>();
XmlDocument doc = new XmlDocument();
doc.Load( xmlpath );
// Load proteins
XmlNodeList proteins = doc.GetElementsByTagName( "PROTEIN" );
foreach( XmlElement element in proteins ) {
int id = int.Parse(element.GetAttribute("ID"));
if( SortedAccession.ContainsKey(id) )
continue;
string acc = element.GetElementsByTagName("ACCESSION")[0].InnerText;
SortedAccession.Add( id, acc );
if( m_SortedProteins.ContainsKey(acc) )
continue;
string entry = element.GetElementsByTagName("ENTRY")[0].InnerText;
string desc = element.GetElementsByTagName("DESCRIPTION")[0].InnerText.Replace('+',' ');
string seq = element.GetElementsByTagName("SEQUENCE")[0].InnerText.ToUpper();
Protein p = new Protein(m_pid++, entry, acc, desc, seq);
Proteins.Add( p );
m_SortedProteins.Add( acc, p );
}
// Load peptides
SortedList<int,Peptide> SortedPeptides = new SortedList<int, Peptide>();
XmlNodeList peptides = doc.GetElementsByTagName( "PEPTIDE" );
foreach( XmlElement element in peptides ) {
int id = int.Parse(element.GetAttribute("ID"));
int pid = int.Parse(element.GetAttribute("PROT_ID"));
int mid = 0;
if( PlgsThreshold != Peptide.ConfidenceType.NoThreshold )
mid = int.Parse(element.GetAttribute("QUERY_MASS_ID"));
string seq = element.GetAttribute("SEQUENCE").ToUpper();
Peptide f = new Peptide(id, seq);
XmlNodeList mods = element.GetElementsByTagName( "MATCH_MODIFIER" );
f.Runs.Add( m_Run );
foreach( XmlElement mod in mods ) {
PTM ptm = new PTM();
string[] strs = mod.GetAttribute("NAME").Split(new char[]{'+'});
ptm.Name = strs[0];
string str = mod.GetAttribute("POS");
ptm.Pos = str.Length == 0 ? -1 : int.Parse(str);
if( strs.Length > 1 )
ptm.Residues = strs[1];
f.AddPTM( ptm );
}
Protein p = null;
try {
p = m_SortedProteins[SortedAccession[pid]];
} catch {
Notify( "Peptide '" + id + "' references unknown protein '" + pid + "'" );
}
if( p != null ) {
p.Peptides.Add( f );
f.Proteins.Add( p );
if( !p.Sequence.Contains(f.Sequence) )
throw new ApplicationException( "Inconsistent sequence data" );
}
Peptides.Add( f );
if( PlgsThreshold != Peptide.ConfidenceType.NoThreshold )
SortedPeptides.Add(mid,f);
}
if( PlgsThreshold == Peptide.ConfidenceType.NoThreshold )
return;
// Scores
XmlNodeList scores = doc.GetElementsByTagName( "MASS_MATCH" );
foreach( XmlElement element in scores ) {
int id = int.Parse(element.GetAttribute("ID"));
double score = double.Parse(element.GetAttribute("SCORE"), m_Format);
SortedPeptides[id].Score = score;
}
}
private bool IsIndistinguisable( Protein g )
{
List<Peptide> discriminating = new List<Peptide>();
foreach( Protein prot in g.Subset )
foreach( Peptide pep in prot.Peptides )
if( pep.Relation == Peptide.RelationType.Discriminating )
discriminating.Add(pep);
foreach( Protein prot in g.Subset )
foreach( Peptide pep in discriminating )
if( !prot.Peptides.Contains(pep) )
return false;
return true;
/*foreach( Peptide f in g.Subset[0].Peptides ) {
if( f.Relation != Peptide.RelationType.Discriminating )
continue;
foreach( Protein p in g.Subset )
if( !p.HasPeptide(f) )
return false;
}
return true;*/
}
private void SaveCSVEntry( TextWriter w, Protein p, string grp, string sep )
{
w.Write( p.ID + sep + p.Entry + sep + p.Accession + sep + p.Evidence.ToString() + sep + grp + sep + p.Desc + sep );
foreach( Peptide f in p.Peptides )
w.Write(f.ToString() + ' ');
w.Write( sep + p.Sequence + sep );
String ptm;
foreach( Peptide peptide in p.Peptides ) {
ptm = GetPtmCsv(peptide);
if( ptm.Length != 0 )
w.Write(ptm+" ");
}
w.WriteLine();
}
private string Peptides2Html( Protein p )
{
if( p.Peptides.Count == 0 )
return "";
int i;
Tag a = new Tag( "a", "href" );
string str = "";
for( i = 0; i < p.Peptides.Count-1; i++ )
str += a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString()) + ", ";
str += a.Render("#"+p.Accession+"__"+p.Peptides[i].ID,p.Peptides[i].ToString());
return str;
}
private void UpdatePdhCvs( ProteinDetectionHypothesisType pdh, Protein.EvidenceType evidence )
{
List<AbstractParamType> list = new List<AbstractParamType>();
foreach( AbstractParamType item in pdh.Items )
if( !item.name.Contains("PAnalyzer") && !item.name.Contains("leading") )
list.Add( item );
CVParamType ev = new CVParamType(), ld = new CVParamType();
ev.cvRef = "PSI-MS";
ld.cvRef = "PSI-MS";
switch( evidence ) {
case Protein.EvidenceType.Conclusive:
ev.accession = "MS:1002213";
ev.name = "PAnalyzer:conclusive protein";
ld.accession = "MS:1002401";
ld.name = "leading protein";
break;
case Protein.EvidenceType.Indistinguishable:
ev.accession = "MS:1002214";
ev.name = "PAnalyzer:indistinguishable protein";
ld.accession = "MS:1002401";
ld.name = "leading protein";
break;
case Protein.EvidenceType.Group:
ev.accession = "MS:1002216";
ev.name = "PAnalyzer:ambiguous group member";
ld.accession = "MS:1002401";
ld.name = "leading protein";
break;
case Protein.EvidenceType.NonConclusive:
ev.accession = "MS:1002215";
ev.name = "PAnalyzer:non-conclusive protein";
ld.accession = "MS:1002402";
ld.name = "non-leading protein";
break;
default: // filtered
return;
}
list.Add(ld);
list.Add(ev);
pdh.Items = list.ToArray();
}