/// <summary>
/// Aggregate the proteins based on the data from the mass tags
/// </summary>
private void AggregateProteins()
{
Dictionary <string, Protein> proteinMap = new Dictionary <string, Protein>();
foreach (ConsensusTarget tag in m_massTags)
{
List <Protein> proteins = tag.GetProteins();
foreach (Protein p in proteins)
{
if (!proteinMap.ContainsKey(p.Reference))
{
proteinMap.Add(p.Reference, p);
p.AddConsensusTarget(tag);
}
else
{
Protein existingProtein = proteinMap[p.Reference];
existingProtein.AddConsensusTarget(tag); // Add the mass tag to the protein
tag.AddProtein(existingProtein); // Add the protein to the mass tag...
}
}
}
int globalCount = 0;
foreach (Protein p in proteinMap.Values)
{
p.Id = globalCount++;
Proteins.Add(p);
}
}
/// <summary>
/// Clear all proteins without proteoforms.
/// </summary>
public void ClearEmptyProteins()
{
Proteins.Clear();
foreach (var protein in allProteins)
{
if (protein.Value.Proteoforms.Count > 0)
{
Proteins.Add(protein.Key, protein.Value);
}
}
}
public void AddProtein(Protein protein)
{
Proteins.Add(protein);
if (RepresentativeProtein == null)
{
RepresentativeProtein = protein;
}
else
{
if (protein.Length > RepresentativeProtein.Length)
{
RepresentativeProtein = protein;
}
}
}
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;
}
/// <summary>
/// Add a Protein-Spectrum-Match identification.
/// </summary>
/// <param name="id">Protein-Spectrum-Math to add</param>
public void Add(PrSm id)
{
RemoveUnidentifiedScan(id);
if (!allProteins.ContainsKey(id.ProteinName))
{
return;
////this.allProteins.Add(id.ProteinName, new ProteinId(id.Sequence, id.ProteinName));
}
if (!Proteins.ContainsKey(id.ProteinName))
{
Proteins.Add(id.ProteinName, allProteins[id.ProteinName]);
}
var protein = Proteins[id.ProteinName];
protein.Add(id);
}
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>
/// Add another protein the ConsensusTarget occurs in
/// </summary>
/// <param name="protein"></param>
public void AddProtein(ProteinInformation protein)
{
Proteins.Add(protein);
protein.Consensus.Add(this);
}
/// <summary>
/// Add a protein to the list of proteins evidenced
/// </summary>
/// <param name="protein"></param>
public void AddProtein(ProteinInformation protein)
{
Proteins.Add(protein);
}
/// <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 );
}
}
}