// find matches using the background proteome
public void UseBackgroundProteome()
{
if (_parent.Document.Settings.PeptideSettings.BackgroundProteome.Equals(BackgroundProteome.NONE))
{
MessageDlg.Show(this, Resources.AssociateProteinsDlg_UseBackgroundProteome_No_background_proteome_defined);
return;
}
_isFasta = false;
_fileName = _parent.Document.Settings.PeptideSettings.BackgroundProteome.DatabasePath;
checkBoxListMatches.Items.Clear();
BackgroundProteome proteome = _parent.Document.Settings.PeptideSettings.BackgroundProteome;
var proteinAssociations = new List <KeyValuePair <FastaSequence, List <PeptideDocNode> > >();
var peptidesForMatching = ListPeptidesForMatching();
IDictionary <String, List <Protein> > proteinsWithSequences = null;
using (var longWaitDlg = new LongWaitDlg())
{
longWaitDlg.PerformWork(this, 1000, longWaitBroker =>
{
using (var proteomeDb = proteome.OpenProteomeDb(longWaitBroker.CancellationToken))
{
proteinsWithSequences = proteomeDb.GetDigestion()
.GetProteinsWithSequences(peptidesForMatching.Select(pep => pep.Peptide.Target.Sequence));
}
});
}
if (proteinsWithSequences == null)
{
return;
}
HashSet <String> processedProteinSequence = new HashSet <string>();
foreach (var entry in proteinsWithSequences)
{
foreach (var protein in entry.Value)
{
if (!processedProteinSequence.Add(protein.Sequence))
{
continue;
}
var matches = peptidesForMatching.Where(pep => protein.Sequence.Contains(pep.Peptide.Target.Sequence)).ToList();
if (matches.Count == 0)
{
continue;
}
FastaSequence fastaSequence = proteome.MakeFastaSequence(protein);
if (fastaSequence != null)
{
proteinAssociations.Add(new KeyValuePair <FastaSequence, List <PeptideDocNode> >(fastaSequence, matches));
}
}
}
SetCheckBoxListItems(proteinAssociations,
Resources.AssociateProteinsDlg_UseBackgroundProteome_No_matches_were_found_using_the_background_proteome_);
}
/// <summary>
/// Tries to match each library peptide to document settings.
/// </summary>
public void MatchAllPeptides(ILongWaitBroker broker)
{
_chargeSettingsMap = new AdductMap <SrmSettings>();
// Build a dictionary mapping sequence to proteins because getting this information is slow.
var dictSequenceProteins = new Dictionary <string, IList <ProteinInfo> >();
var dictNewNodePeps = new Dictionary <PeptideSequenceModKey, PeptideMatch>();
PeptideMatches = null;
MatchedPeptideCount = 0;
int peptides = 0;
int totalPeptides = _libraryPepInfos.Count;
ProteomeDb proteomeDb = null;
IStatelessSession session = null;
try
{
foreach (ViewLibraryPepInfo pepInfo in _libraryPepInfos)
{
if (broker.IsCanceled)
{
return;
}
var charge = pepInfo.Key.Adduct;
// Find the matching peptide.
var nodePepMatched = AssociateMatchingPeptide(pepInfo, charge).PeptideNode;
if (nodePepMatched != null)
{
MatchedPeptideCount++;
PeptideMatch peptideMatchInDict;
// If peptide is already in the dictionary of peptides to add, merge the children.
if (!dictNewNodePeps.TryGetValue(nodePepMatched.SequenceKey, out peptideMatchInDict))
{
IList <ProteinInfo> matchedProteins = null;
var target = nodePepMatched.Peptide.Target;
// This is only set if the user has checked the associate peptide box.
if (target.IsProteomic && _backgroundProteome != null)
{
var sequence = target.Sequence;
// We want to query the background proteome as little as possible,
// so sequences are mapped to protein lists in a dictionary.
if (!dictSequenceProteins.TryGetValue(sequence, out matchedProteins))
{
if (proteomeDb == null)
{
proteomeDb = _backgroundProteome.OpenProteomeDb(broker.CancellationToken);
session = proteomeDb.OpenStatelessSession(false);
}
var digestion = proteomeDb.GetDigestion();
if (sequence.Length >= MIN_PEPTIDE_LENGTH)
{
matchedProteins = digestion.GetProteinsWithSequence(session, sequence)
.Select(protein => new ProteinInfo(protein)).ToList();
}
if (matchedProteins == null || matchedProteins.Count > MAX_PROTEIN_MATCHES)
{
// If the peptide was too short, or matched too many proteins, then
// treat it as if it was not found in any proteins.
matchedProteins = new List <ProteinInfo>();
}
dictSequenceProteins.Add(sequence, matchedProteins);
}
}
dictNewNodePeps.Add(nodePepMatched.SequenceKey,
new PeptideMatch(nodePepMatched, matchedProteins,
MatchesFilter(target, charge)));
}
else
{
PeptideDocNode nodePepInDictionary = peptideMatchInDict.NodePep;
if (!nodePepInDictionary.HasChildCharge(charge))
{
List <DocNode> newChildren = nodePepInDictionary.Children.ToList();
newChildren.AddRange(nodePepMatched.Children);
newChildren.Sort(Peptide.CompareGroups);
var key = nodePepMatched.SequenceKey;
dictNewNodePeps.Remove(key);
dictNewNodePeps.Add(key,
new PeptideMatch((PeptideDocNode)nodePepInDictionary.ChangeChildren(newChildren),
peptideMatchInDict.Proteins, peptideMatchInDict.MatchesFilterSettings));
}
}
}
peptides++;
int progressValue = (int)((peptides + 0.0) / totalPeptides * PERCENT_PEPTIDE_MATCH);
broker.ProgressValue = progressValue;
}
PeptideMatches = dictNewNodePeps;
}
finally
{
using (proteomeDb)
using (session)
{
}
}
}
/// <summary>
/// Tries to match each library peptide to document settings.
/// </summary>
public void MatchAllPeptides(ILongWaitBroker broker)
{
_chargeSettingsMap = new SrmSettings[128];
// Build a dictionary mapping sequence to proteins because getting this information is slow.
var dictSequenceProteins = new Dictionary <string, IList <ProteinInfo> >();
var dictNewNodePeps = new Dictionary <PeptideSequenceModKey, PeptideMatch>();
PeptideMatches = null;
MatchedPeptideCount = 0;
int peptides = 0;
int totalPeptides = _libraryPepInfos.Length;
foreach (ViewLibraryPepInfo pepInfo in _libraryPepInfos)
{
if (broker.IsCanceled)
{
return;
}
int charge = pepInfo.Key.Charge;
// Find the matching peptide.
var nodePepMatched = AssociateMatchingPeptide(pepInfo, charge).PeptideNode;
if (nodePepMatched != null)
{
MatchedPeptideCount++;
PeptideMatch peptideMatchInDict;
// If peptide is already in the dictionary of peptides to add, merge the children.
if (!dictNewNodePeps.TryGetValue(nodePepMatched.SequenceKey, out peptideMatchInDict))
{
IList <ProteinInfo> matchedProteins = null;
var sequence = nodePepMatched.Peptide.Sequence;
// This is only set if the user has checked the associate peptide box.
if (_backgroundProteome != null)
{
// We want to query the background proteome as little as possible,
// so sequences are mapped to protein lists in a dictionary.
if (!dictSequenceProteins.TryGetValue(sequence, out matchedProteins))
{
using (var proteomeDb = _backgroundProteome.OpenProteomeDb())
{
var digestion = _backgroundProteome.GetDigestion(proteomeDb, Settings.PeptideSettings);
if (digestion != null)
{
matchedProteins = digestion.GetProteinsWithSequence(sequence).Select(protein => new ProteinInfo(protein)).ToList();
dictSequenceProteins.Add(sequence, matchedProteins);
}
}
}
}
dictNewNodePeps.Add(nodePepMatched.SequenceKey,
new PeptideMatch(nodePepMatched, matchedProteins,
MatchesFilter(sequence, charge)));
}
else
{
PeptideDocNode nodePepInDictionary = peptideMatchInDict.NodePep;
if (!nodePepInDictionary.HasChildCharge(charge))
{
List <DocNode> newChildren = nodePepInDictionary.Children.ToList();
newChildren.AddRange(nodePepMatched.Children);
newChildren.Sort(Peptide.CompareGroups);
var key = nodePepMatched.SequenceKey;
dictNewNodePeps.Remove(key);
dictNewNodePeps.Add(key,
new PeptideMatch((PeptideDocNode)nodePepInDictionary.ChangeChildren(newChildren),
peptideMatchInDict.Proteins, peptideMatchInDict.MatchesFilterSettings));
}
}
}
peptides++;
int progressValue = (int)((peptides + 0.0) / totalPeptides * PERCENT_PEPTIDE_MATCH);
broker.ProgressValue = progressValue;
}
PeptideMatches = dictNewNodePeps;
}
