public ViewLibraryPepMatching(SrmDocument document,
Library library,
LibrarySpec spec,
LibKeyModificationMatcher matcher,
IReadOnlyList <ViewLibraryPepInfo> peptides)
{
_document = document;
_selectedLibrary = library;
_selectedSpec = spec;
_matcher = matcher;
_libraryPepInfos = peptides;
_chargeSettingsMap = new AdductMap <SrmSettings>();
}
public void UseCurrentData()
{
var arrayData = GetRegressionDatas();
if (arrayData == null)
{
return;
}
bool hasRegressionLines = false;
var regressionLines = new AdductMap <RegressionLine>();
foreach (var i in arrayData.Keys)
{
if (arrayData[i] == null)
{
continue;
}
regressionLines[i] = arrayData[i].RegressionLine;
if (regressionLines[i] != null)
{
hasRegressionLines = true;
}
}
if (!hasRegressionLines)
{
MessageDlg.Show(this, Resources.EditCEDlg_UseCurrentData_Insufficient_data_found_to_calculate_a_new_regression);
return;
}
gridRegression.Rows.Clear();
foreach (var i in regressionLines.Keys)
{
var regressionLine = regressionLines[i];
if (regressionLine == null)
{
continue;
}
gridRegression.Rows.Add(new object[]
{
i.AdductCharge.ToString(LocalizationHelper.CurrentCulture),
string.Format("{0:F04}", regressionLine.Slope), // Not L10N
string.Format("{0:F04}", regressionLine.Intercept) // Not L10N
});
}
}
private AdductMap <CERegressionData> GetRegressionDatas()
{
var document = Program.ActiveDocumentUI;
if (!document.Settings.HasResults)
{
return(null);
}
if (!document.Settings.MeasuredResults.IsLoaded)
{
MessageBox.Show(this, Resources.EditCEDlg_GetRegressionDatas_Measured_results_must_be_completely_loaded_before_they_can_be_used_to_create_a_collision_energy_regression,
Program.Name);
return(null);
}
var regressionCurrent = _regression ??
document.Settings.TransitionSettings.Prediction.CollisionEnergy;
var arrayData = new AdductMap <CERegressionData>();
var chromatograms = document.Settings.MeasuredResults.Chromatograms;
for (int i = 0; i < chromatograms.Count; i++)
{
var chromSet = chromatograms[i];
var regression = chromSet.OptimizationFunction as CollisionEnergyRegression;
if (regression == null)
{
continue;
}
foreach (var nodeGroup in document.MoleculeTransitionGroups)
{
var charge = nodeGroup.TransitionGroup.PrecursorAdduct;
if (arrayData[charge] == null)
{
var chargeRegression = (regressionCurrent != null ?
regressionCurrent.GetRegressionLine(charge) : null);
arrayData[charge] = new CERegressionData(chargeRegression != null ?
chargeRegression.RegressionLine : null);
}
arrayData[charge].Add(regression, nodeGroup, i);
}
}
return(arrayData);
}
public PeptideDocNode MatchSinglePeptide(ViewLibraryPepInfo pepInfo)
{
_chargeSettingsMap = new AdductMap <SrmSettings>();
var nodePep = AssociateMatchingPeptide(pepInfo, pepInfo.Key.Adduct).PeptideNode;
if (nodePep == null)
{
return(null);
}
IList <ProteinInfo> matchedProteins = null;
var target = nodePep.Peptide.Target;
if (pepInfo.Target.IsProteomic)
{
// This is only set if the user has checked the associate peptide box.
if (_backgroundProteome != null)
{
using (var proteomeDb = _backgroundProteome.OpenProteomeDb())
{
var digestion = _backgroundProteome.GetDigestion(proteomeDb, Settings.PeptideSettings);
if (digestion != null)
{
matchedProteins = digestion.GetProteinsWithSequence(target.Sequence)
.Select(protein => new ProteinInfo(protein)).ToArray();
}
}
}
}
PeptideMatches = new Dictionary <PeptideSequenceModKey, PeptideMatch>
{
{ nodePep.SequenceKey, new PeptideMatch(nodePep, matchedProteins,
pepInfo.LibInfo,
MatchesFilter(target, pepInfo.Key.Adduct)) }
};
return(nodePep);
}
private void TestAdductOperators()
{
// Test some underlying formula handling for fanciful user-supplied values
Assert.IsTrue(Molecule.AreEquivalentFormulas("C10H30Si5O5H-CH4", "C9H27O5Si5"));
Assert.AreEqual("C7H27O5Si4", BioMassCalc.MONOISOTOPIC.FindFormulaIntersection(new[] { "C8H30Si5O5H-CH4", "C9H27O5Si4", "C9H27O5Si5Na" }));
Assert.AreEqual("C7H27O5Si4", BioMassCalc.MONOISOTOPIC.FindFormulaIntersectionUnlabeled(new[] { "C7C'H30Si5O5H-CH4", "C9H27O5Si4", "C9H25H'2O5Si5Na" }));
// There is a difference between a proteomic adduct and non proteomic, primarily in how they display
Assert.AreEqual(Adduct.FromStringAssumeChargeOnly("M+H"), Adduct.M_PLUS_H);
Assert.AreEqual(Adduct.FromStringAssumeProtonatedNonProteomic("1"), Adduct.M_PLUS_H);
Assert.AreEqual(Adduct.FromStringAssumeChargeOnly("1"), Adduct.M_PLUS);
Assert.AreEqual(Adduct.FromStringAssumeProtonatedNonProteomic("M+H"), Adduct.M_PLUS_H);
Assert.AreEqual(Adduct.FromStringAssumeProtonated("1"), Adduct.SINGLY_PROTONATED);
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M+H"), Adduct.SINGLY_PROTONATED);
Assert.AreEqual(Adduct.FromStringAssumeChargeOnly("M+H").AsFormula(), Adduct.SINGLY_PROTONATED.AsFormula()); // But the underlying chemistry is the same
Assert.AreEqual(Adduct.FromStringAssumeProtonated("[M+S]+"), Adduct.FromStringAssumeProtonated("M+S").ChangeCharge(1));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M(-1.234)+2Na"), Adduct.FromStringAssumeProtonated("M(-1.234)+3Na").ChangeCharge(2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M1.234+2Na"), Adduct.FromStringAssumeProtonated("M1.234+3Na").ChangeCharge(2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M2Cl37-2Na"), Adduct.FromStringAssumeProtonated("M2Cl37+3Na").ChangeCharge(-2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M1.234-2Na"), Adduct.FromStringAssumeProtonated("M1.234+3Na").ChangeCharge(-2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M(-1.234)-2Na"), Adduct.FromStringAssumeProtonated("M(-1.234)+3Na").ChangeCharge(-2));
Assert.IsFalse(Adduct.M_PLUS_H.IsProteomic);
Assert.IsTrue(Adduct.M_PLUS_H.IsProtonated);
Assert.IsTrue(Adduct.SINGLY_PROTONATED.IsProteomic);
Assert.IsTrue(Adduct.SINGLY_PROTONATED.IsProtonated);
Assert.IsFalse(Adduct.SINGLY_PROTONATED.IsEmpty);
Assert.IsFalse(Adduct.EMPTY.IsProteomic);
Assert.IsTrue(Adduct.EMPTY.IsEmpty);
// Exercise the ability to work with masses and isotope labels
Assert.IsTrue(ReferenceEquals(Adduct.SINGLY_PROTONATED, Adduct.SINGLY_PROTONATED.Unlabeled));
var nolabel = Adduct.FromStringAssumeProtonated("M-2Na");
var label = Adduct.FromStringAssumeProtonated("M2Cl37-2Na");
Assert.AreEqual(nolabel, label.Unlabeled);
Assert.IsTrue(ReferenceEquals(nolabel, nolabel.Unlabeled));
Assert.IsFalse(nolabel.MassFromMz(300.0, MassType.Monoisotopic).IsHeavy());
Assert.IsFalse(label.MassFromMz(300.0, MassType.Monoisotopic).IsHeavy());
Assert.IsTrue(label.MassFromMz(300.0, MassType.MonoisotopicHeavy).IsHeavy());
Assert.IsTrue(label.MassFromMz(300.0, MassType.Monoisotopic).IsMonoIsotopic());
Assert.IsFalse(nolabel.MassFromMz(300.0, MassType.Average).IsHeavy());
Assert.IsFalse(label.MassFromMz(300.0, MassType.Average).IsHeavy());
Assert.IsTrue(label.MassFromMz(300.0, MassType.AverageHeavy).IsHeavy());
Assert.IsTrue(label.MassFromMz(300.0, MassType.Average).IsAverage());
var massHeavy = label.ApplyToMass(new TypedMass(300, MassType.MonoisotopicHeavy)); // Will not have isotope effect added in mz calc, as it's already heavy
var massLight = label.ApplyToMass(new TypedMass(300, MassType.Monoisotopic)); // Will have isotope effect added in mz calc
Assert.AreNotEqual(massHeavy, massLight);
Assert.AreNotEqual(label.MzFromNeutralMass(massHeavy), label.MzFromNeutralMass(massLight));
Assert.IsTrue(Adduct.PossibleAdductDescriptionStart("["));
Assert.IsTrue(Adduct.PossibleAdductDescriptionStart("M"));
Assert.IsTrue(Adduct.PossibleAdductDescriptionStart("[2M+CH3COO]"));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M+2CH3COO"), Adduct.FromStringAssumeProtonated("M+CH3COO").ChangeCharge(-2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M-2CH3COO"), Adduct.FromStringAssumeProtonated("M+CH3COO").ChangeCharge(2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M-2Na"), Adduct.FromStringAssumeProtonated("M+Na").ChangeCharge(-2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M+Na"), Adduct.FromStringAssumeProtonated("M+Na").ChangeCharge(1));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M+2Na"), Adduct.FromStringAssumeProtonated("M+Na").ChangeCharge(2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M+2Na"), Adduct.FromStringAssumeProtonated("M+3Na").ChangeCharge(2));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M2Cl37+2Na"), Adduct.FromStringAssumeProtonated("M2Cl37+3Na").ChangeCharge(2));
AssertEx.ThrowsException <InvalidOperationException>(() => Adduct.FromStringAssumeProtonated("M+2Na-H").ChangeCharge(2)); // Too complex to adjust formula
AssertEx.ThrowsException <InvalidOperationException>(() => Adduct.FromStringAssumeProtonatedNonProteomic("[M2")); // Seen in the wild, wasn't handled well
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M++++").AdductCharge, Adduct.FromChargeNoMass(4).AdductCharge);
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M+4"), Adduct.FromChargeNoMass(4));
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M--").AdductCharge, Adduct.FromChargeNoMass(-2).AdductCharge);
Assert.AreEqual(Adduct.FromStringAssumeProtonated("M-2"), Adduct.FromChargeNoMass(-2));
Assert.IsTrue(ReferenceEquals(Adduct.FromStringAssumeChargeOnly("M-"), Adduct.FromChargeNoMass(-1))); // Both should return Adduct.M_MINUS
Assert.IsTrue(ReferenceEquals(Adduct.FromStringAssumeChargeOnly("M+"), Adduct.FromChargeNoMass(1))); // Both should return Adduct.M_PLUS
Assert.IsTrue(ReferenceEquals(Adduct.FromStringAssumeChargeOnly("[M+]"), Adduct.FromChargeNoMass(1))); // Both should return Adduct.M_PLUS
Assert.IsTrue(ReferenceEquals(Adduct.FromStringAssumeChargeOnly("M-H"), Adduct.M_MINUS_H));
Assert.IsTrue(ReferenceEquals(Adduct.FromStringAssumeChargeOnly("M+H"), Adduct.M_PLUS_H));
var a = Adduct.FromChargeProtonated(-1);
var aa = Adduct.FromStringAssumeProtonated("M+CH3COO");
var b = Adduct.FromChargeProtonated(2);
var bb = Adduct.FromChargeProtonated(2);
var bbb = Adduct.FromChargeProtonated(2);
var bbbb = Adduct.FromChargeProtonated(2);
var c = Adduct.FromChargeProtonated(3);
var cc = Adduct.FromStringAssumeChargeOnly("M+3H");
var ccc = Adduct.FromStringAssumeChargeOnly("[M+3H]");
Assert.AreEqual(a.AdductCharge, aa.AdductCharge);
Assert.IsTrue(b == bb);
Assert.IsTrue(b == bbb);
Assert.IsTrue(ReferenceEquals(bbb, bbbb));
Assert.IsTrue(c.AdductCharge == cc.AdductCharge);
Assert.IsFalse(c == cc);
Assert.IsTrue(c != cc);
Assert.IsTrue(cc == ccc);
Assert.IsTrue(a < aa);
Assert.IsTrue(a < b);
Assert.IsTrue(b > a);
Assert.IsTrue(b != a);
var sorted = new List <Adduct> {
a, aa, b, bb, bbb, bbbb, c, cc, ccc
};
var unsorted = new List <Adduct> {
bb, aa, ccc, b, c, bbb, a, bbbb, cc
};
Assert.IsFalse(sorted.SequenceEqual(unsorted));
unsorted.Sort();
Assert.IsTrue(sorted.SequenceEqual(unsorted));
var ints = new AdductMap <int>();
Assert.AreEqual(0, ints[a]);
ints[a] = 7;
Assert.AreEqual(7, ints[a]);
var adducts = new AdductMap <Adduct>();
Assert.AreEqual(null, adducts[a]);
adducts[a] = b;
Assert.AreEqual(b, adducts[a]);
adducts[a] = c;
Assert.AreEqual(c, adducts[a]);
var d = Adduct.FromStringAssumeProtonated("[2M+3H]");
var dd = Adduct.FromStringAssumeProtonated("[M+3H]");
var ddd = Adduct.FromStringAssumeProtonated("[M-Na]");
Assert.IsTrue(d.ChangeMassMultiplier(1).SameEffect(dd));
Assert.IsTrue(dd.ChangeMassMultiplier(2).SameEffect(d));
Assert.AreEqual(dd.ChangeIonFormula("-Na"), ddd);
Assert.AreEqual(d.ChangeMassMultiplier(1).ChangeIonFormula("-Na"), ddd);
CheckLabel(BioMassCalc.Cl37);
CheckLabel(BioMassCalc.Br81);
CheckLabel(BioMassCalc.S33);
CheckLabel(BioMassCalc.S34);
CheckLabel(BioMassCalc.P32);
CheckLabel(BioMassCalc.C14);
CheckLabel(BioMassCalc.O17);
CheckLabel(BioMassCalc.O18);
var tips = Adduct.Tips;
foreach (var nickname in Adduct.DICT_ADDUCT_NICKNAMES)
{
Assert.IsTrue(tips.Contains(nickname.Key));
}
foreach (var nickname in Adduct.DICT_ADDUCT_ISOTOPE_NICKNAMES)
{
Assert.IsTrue(tips.Contains(nickname.Key));
}
}
/// <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)
{
}
}
}
