/// <summary>
/// Returns the string that MSstats code uses to identify a row of data in the MSstats Input report.
/// </summary>
private static string GetFeatureKey(SrmDocument document, IdentityPath identityPath)
{
PeptideGroupDocNode peptideGroup = (PeptideGroupDocNode)document.FindNode(identityPath.GetIdentity(0));
PeptideDocNode peptide = (PeptideDocNode)peptideGroup.FindNode(identityPath.GetIdentity(1));
TransitionGroupDocNode transitionGroup =
(TransitionGroupDocNode)peptide.FindNode(identityPath.GetIdentity(2));
TransitionDocNode transition = (TransitionDocNode)transitionGroup.FindNode(identityPath.GetIdentity(3));
return(peptide.ModifiedSequenceDisplay + '_' + transitionGroup.PrecursorCharge + '_' + GetFragmentIon(transition) +
'_' + transition.Transition.Charge);
}
public void IdentityPathTest()
{
SimpleIdentity id = new SimpleIdentity();
IdentityPath pathSimple = new IdentityPath(id);
Assert.AreEqual(IdentityPath.ROOT, pathSimple.Parent);
Assert.AreEqual(IdentityPath.ROOT, pathSimple.GetPathTo(-1));
Assert.AreEqual(id, pathSimple.Child);
Assert.AreEqual(1, pathSimple.Length);
Assert.AreEqual(0, pathSimple.Depth);
AssertEx.ThrowsException <IndexOutOfRangeException>(() => pathSimple.GetPathTo(-2));
AssertEx.ThrowsException <IndexOutOfRangeException>(() => pathSimple.GetPathTo(1));
IdentityPath pathRoot = new IdentityPath(new Identity[0]);
Assert.AreEqual(IdentityPath.ROOT, pathRoot);
Assert.AreEqual(0, pathRoot.Length);
Assert.AreEqual(-1, pathRoot.Depth);
AssertEx.ThrowsException <IndexOutOfRangeException>(() => pathRoot.Child);
AssertEx.ThrowsException <IndexOutOfRangeException>(() => pathRoot.Parent);
const int count = 5;
List <Identity> listId1 = new List <Identity>();
List <Identity> listId2 = new List <Identity>();
HashSet <IdentityPath> setPaths = new HashSet <IdentityPath>();
IdentityPath last = IdentityPath.ROOT;
for (int i = 0; i < count; i++)
{
listId1.Add(new NumericIdentity(i));
listId2.Add(new NumericIdentity(i));
IdentityPath path = new IdentityPath(listId1);
Assert.AreEqual(path, new IdentityPath(listId1));
Assert.AreNotEqual(path, new IdentityPath(listId2));
Assert.AreEqual(last, path.Parent);
Assert.AreSame(listId1[i], path.Child);
Assert.AreEqual(path, new IdentityPath(path.Parent, path.Child));
Assert.AreEqual(i, path.Depth);
Assert.AreEqual(listId1.Count, path.Length);
Assert.AreSame(path.Child, path.GetIdentity(path.Depth));
Assert.AreEqual(path.Parent, path.GetPathTo(path.Depth - 1));
Assert.AreEqual("/" + listId1.ToString("/"), path.ToString());
for (int j = 0; j < i; j++)
{
Assert.IsTrue(setPaths.Contains(path.GetPathTo(j)));
}
setPaths.Add(path);
last = path;
}
}
protected static bool StartsWith(IdentityPath child, IdentityPath ancestor)
{
if (child.Length < ancestor.Length)
{
return(false);
}
for (int i = 0; i < ancestor.Length; i++)
{
if (!ReferenceEquals(ancestor.GetIdentity(i), child.GetIdentity(i)))
{
return(false);
}
}
return(true);
}
private SrmDocument AddPeptides(SrmDocument document, bool validating, ref IdentityPath selectedPath)
{
if (tabControl1.SelectedTab != tabPagePeptideList)
return document;
var matcher = new ModificationMatcher();
var listPeptideSequences = ListPeptideSequences();
if (listPeptideSequences == null)
return null;
try
{
matcher.CreateMatches(document.Settings, listPeptideSequences, Settings.Default.StaticModList,
Settings.Default.HeavyModList);
}
catch (FormatException e)
{
MessageDlg.ShowException(this, e);
ShowPeptideError(new PasteError
{
Column = colPeptideSequence.Index,
Message = Resources.PasteDlg_AddPeptides_Unable_to_interpret_peptide_modifications
});
return null;
}
var strNameMatches = matcher.FoundMatches;
if (!validating && !string.IsNullOrEmpty(strNameMatches))
{
string message = TextUtil.LineSeparate(Resources.PasteDlg_AddPeptides_Would_you_like_to_use_the_Unimod_definitions_for_the_following_modifications,
string.Empty, strNameMatches);
if (MultiButtonMsgDlg.Show(this, message, Resources.PasteDlg_AddPeptides_OK) == DialogResult.Cancel)
return null;
}
var backgroundProteome = GetBackgroundProteome(document);
// Insert last to first so that proteins get inserted on top of each other
// in the order they are added. Peptide insertion into peptide lists needs
// to be carefully tracked to insert them in the order they are listed in
// the grid.
int lastGroupGlobalIndex = 0, lastPeptideIndex = -1;
for (int i = gridViewPeptides.Rows.Count - 1; i >= 0; i--)
{
PeptideGroupDocNode peptideGroupDocNode;
var row = gridViewPeptides.Rows[i];
var pepModSequence = Convert.ToString(row.Cells[colPeptideSequence.Index].Value);
pepModSequence = FastaSequence.NormalizeNTerminalMod(pepModSequence);
var proteinName = Convert.ToString(row.Cells[colPeptideProtein.Index].Value);
if (string.IsNullOrEmpty(pepModSequence) && string.IsNullOrEmpty(proteinName))
continue;
if (string.IsNullOrEmpty(proteinName))
{
peptideGroupDocNode = GetSelectedPeptideGroupDocNode(document, selectedPath);
if (!IsPeptideListDocNode(peptideGroupDocNode))
{
peptideGroupDocNode = null;
}
}
else
{
peptideGroupDocNode = FindPeptideGroupDocNode(document, proteinName);
}
if (peptideGroupDocNode == null)
{
if (string.IsNullOrEmpty(proteinName))
{
peptideGroupDocNode = new PeptideGroupDocNode(new PeptideGroup(),
document.GetPeptideGroupId(true), null,
new PeptideDocNode[0]);
}
else
{
ProteinMetadata metadata = null;
PeptideGroup peptideGroup = backgroundProteome.IsNone ? new PeptideGroup()
: (backgroundProteome.GetFastaSequence(proteinName, out metadata) ??
new PeptideGroup());
if (metadata != null)
peptideGroupDocNode = new PeptideGroupDocNode(peptideGroup, metadata, new PeptideDocNode[0]);
else
peptideGroupDocNode = new PeptideGroupDocNode(peptideGroup, proteinName,
peptideGroup.Description, new PeptideDocNode[0]);
}
// Add to the end, if no insert node
var to = selectedPath;
if (to == null || to.Depth < (int)SrmDocument.Level.MoleculeGroups)
document = (SrmDocument)document.Add(peptideGroupDocNode);
else
{
Identity toId = selectedPath.GetIdentity((int) SrmDocument.Level.MoleculeGroups);
document = (SrmDocument) document.Insert(toId, peptideGroupDocNode);
}
selectedPath = new IdentityPath(peptideGroupDocNode.Id);
}
var peptides = new List<PeptideDocNode>();
foreach (PeptideDocNode peptideDocNode in peptideGroupDocNode.Children)
{
peptides.Add(peptideDocNode);
}
var fastaSequence = peptideGroupDocNode.PeptideGroup as FastaSequence;
PeptideDocNode nodePepNew;
if (fastaSequence != null)
{
// Attempt to create node for error checking.
nodePepNew = fastaSequence.CreateFullPeptideDocNode(document.Settings,
FastaSequence.StripModifications(pepModSequence));
if (nodePepNew == null)
{
ShowPeptideError(new PasteError
{
Column = colPeptideSequence.Index,
Line = i,
Message = Resources.PasteDlg_AddPeptides_This_peptide_sequence_was_not_found_in_the_protein_sequence
});
return null;
}
}
// Create node using ModificationMatcher.
nodePepNew = matcher.GetModifiedNode(pepModSequence, fastaSequence).ChangeSettings(document.Settings,
SrmSettingsDiff.ALL);
// Avoid adding an existing peptide a second time.
if (!peptides.Contains(nodePep => Equals(nodePep.Key, nodePepNew.Key)))
{
if (nodePepNew.Peptide.FastaSequence != null)
{
peptides.Add(nodePepNew);
peptides.Sort(FastaSequence.ComparePeptides);
}
else
{
int groupGlobalIndex = peptideGroupDocNode.PeptideGroup.GlobalIndex;
if (groupGlobalIndex == lastGroupGlobalIndex && lastPeptideIndex != -1)
{
peptides.Insert(lastPeptideIndex, nodePepNew);
}
else
{
lastPeptideIndex = peptides.Count;
peptides.Add(nodePepNew);
}
lastGroupGlobalIndex = groupGlobalIndex;
}
var newPeptideGroupDocNode = new PeptideGroupDocNode(peptideGroupDocNode.PeptideGroup, peptideGroupDocNode.Annotations, peptideGroupDocNode.Name, peptideGroupDocNode.Description, peptides.ToArray(), false);
document = (SrmDocument)document.ReplaceChild(newPeptideGroupDocNode);
}
}
if (!validating && listPeptideSequences.Count > 0)
{
var pepModsNew = matcher.GetDocModifications(document);
document = document.ChangeSettings(document.Settings.ChangePeptideModifications(mods => pepModsNew));
document.Settings.UpdateDefaultModifications(false);
}
return document;
}
private SrmDocument AddTransitionList(SrmDocument document, ref IdentityPath selectedPath)
{
if (tabControl1.SelectedTab != tabPageTransitionList)
return document;
if (IsMolecule)
{
// Save the current column order to settings
var active = new List<string>();
for (int order = 0; order < gridViewTransitionList.Columns.Count; order++)
{
for (int gridcol = 0; gridcol < gridViewTransitionList.Columns.Count; gridcol++)
{
var dataGridViewColumn = gridViewTransitionList.Columns[gridcol];
if (dataGridViewColumn.DisplayIndex == order)
{
if (dataGridViewColumn.Visible)
active.Add(dataGridViewColumn.Name);
break;
}
}
}
Settings.Default.CustomMoleculeTransitionInsertColumnsList = active;
// We will accept a completely empty product list as meaning
// "these are all precursor transitions"
var requireProductInfo = false;
for (var i = 0; i < gridViewTransitionList.RowCount - 1; i++)
{
var row = gridViewTransitionList.Rows[i];
var productMz = row.Cells[INDEX_PRODUCT_MZ].Value;
var productFormula = row.Cells[INDEX_PRODUCT_FORMULA].Value;
var productCharge = row.Cells[INDEX_PRODUCT_CHARGE].Value;
if ((productMz != null && productMz.ToString().Length > 0) ||
(productFormula != null && productFormula.ToString().Length > 0) ||
(productCharge != null && productCharge.ToString().Length > 0))
{
requireProductInfo = true; // Product list is not completely empty
break;
}
}
// For each row in the grid, add to or begin MoleculeGroup|Molecule|TransitionList tree
for(int i = 0; i < gridViewTransitionList.RowCount - 1; i ++)
{
DataGridViewRow row = gridViewTransitionList.Rows[i];
var precursor = ReadPrecursorOrProductColumns(document, row, true); // Get molecule values
if (precursor == null)
return null;
if (requireProductInfo && ReadPrecursorOrProductColumns(document, row, false) == null)
{
return null;
}
var charge = precursor.Charge;
var precursorMonoMz = BioMassCalc.CalculateIonMz(precursor.MonoMass, charge);
var precursorAverageMz = BioMassCalc.CalculateIonMz(precursor.AverageMass, charge);
// Preexisting molecule group?
bool pepGroupFound = false;
foreach (var pepGroup in document.MoleculeGroups)
{
var pathPepGroup = new IdentityPath(pepGroup.Id);
if (Equals(pepGroup.Name, Convert.ToString(row.Cells[INDEX_MOLECULE_GROUP].Value)))
{
// Found a molecule group with the same name - can we find an existing transition group to which we can add a transition?
pepGroupFound = true;
bool pepFound = false;
foreach (var pep in pepGroup.SmallMolecules)
{
var pepPath = new IdentityPath(pathPepGroup, pep.Id);
var ionMonoMz = BioMassCalc.CalculateIonMz(pep.CustomIon.MonoisotopicMass, charge);
var ionAverageMz = BioMassCalc.CalculateIonMz(pep.CustomIon.AverageMass, charge);
// Match existing molecule if same name (if any) and same formula (if any) and similar m/z at the precursor charge
// (we don't just check mass since we don't have a tolerance value for that)
// Or same name If any) and identical formula when stripped of labels
// Or same name, no formula, and different isotope labels
if (Equals(pep.CustomIon.Name, precursor.Name) &&
((Equals(pep.CustomIon.Formula, precursor.Formula) &&
Math.Abs(ionMonoMz - precursorMonoMz) <= document.Settings.TransitionSettings.Instrument.MzMatchTolerance &&
Math.Abs(ionAverageMz - precursorAverageMz) <= document.Settings.TransitionSettings.Instrument.MzMatchTolerance) ||
(!Equals(pep.CustomIon.Formula, precursor.Formula) &&
Equals(pep.CustomIon.UnlabeledFormula, BioMassCalc.MONOISOTOPIC.StripLabelsFromFormula(precursor.Formula))) ||
(string.IsNullOrEmpty(pep.CustomIon.Formula) && string.IsNullOrEmpty(precursor.Formula) &&
!pep.TransitionGroups.Any(t => Equals(t.TransitionGroup.LabelType, precursor.IsotopeLabelType??IsotopeLabelType.light))) ))
{
pepFound = true;
bool tranGroupFound = false;
foreach (var tranGroup in pep.TransitionGroups)
{
var pathGroup = new IdentityPath(pepPath, tranGroup.Id);
if (Math.Abs(tranGroup.PrecursorMz - precursor.Mz) <= document.Settings.TransitionSettings.Instrument.MzMatchTolerance)
{
tranGroupFound = true;
var tranFound = false;
try
{
var tranNode = GetMoleculeTransition(document, row, pep.Peptide, tranGroup.TransitionGroup, requireProductInfo);
if (tranNode == null)
return null;
foreach (var tran in tranGroup.Transitions)
{
if (Equals(tranNode.Transition.CustomIon,tran.Transition.CustomIon))
{
tranFound = true;
break;
}
}
if (!tranFound)
{
document = (SrmDocument) document.Add(pathGroup, tranNode);
}
}
catch (InvalidDataException e)
{
// Some error we didn't catch in the basic checks
ShowTransitionError(new PasteError
{
Column = 0,
Line = row.Index,
Message = e.Message
});
return null;
}
break;
}
}
if (!tranGroupFound)
{
var node = GetMoleculeTransitionGroup(document, row, pep.Peptide, requireProductInfo);
if (node == null)
return null;
document =
(SrmDocument)
document.Add(pepPath, node);
}
break;
}
}
if (!pepFound)
{
var node = GetMoleculePeptide(document, row, pepGroup.PeptideGroup, requireProductInfo);
if (node == null)
return null;
document =
(SrmDocument)
document.Add(pathPepGroup,node);
}
break;
}
}
if (!pepGroupFound)
{
var node = GetMoleculePeptideGroup(document, row, requireProductInfo);
if (node == null)
return null;
IdentityPath first;
IdentityPath next;
document =
document.AddPeptideGroups(new[] {node}, false,null
, out first,out next);
}
}
}
else
{
var backgroundProteome = GetBackgroundProteome(document);
var sbTransitionList = new StringBuilder();
var dictNameSeq = new Dictionary<string, FastaSequence>();
// Add all existing FASTA sequences in the document to the name to seq dictionary
// Including named peptide lists would cause the import code to give matching names
// in this list new names (e.g. with 1, 2, 3 appended). In this code, the names
// are intended to be merged.
foreach (var nodePepGroup in document.Children.Cast<PeptideGroupDocNode>().Where(n => !n.IsPeptideList))
{
if (!dictNameSeq.ContainsKey(nodePepGroup.Name))
dictNameSeq.Add(nodePepGroup.Name, (FastaSequence) nodePepGroup.PeptideGroup);
}
// Check for simple errors and build strings for import
for (int i = 0; i < gridViewTransitionList.Rows.Count; i++)
{
var row = gridViewTransitionList.Rows[i];
var peptideSequence = Convert.ToString(row.Cells[colTransitionPeptide.Index].Value);
var proteinName = Convert.ToString(row.Cells[colTransitionProteinName.Index].Value);
var precursorMzText = Convert.ToString(row.Cells[colTransitionPrecursorMz.Index].Value);
var productMzText = Convert.ToString(row.Cells[colTransitionProductMz.Index].Value);
if (string.IsNullOrEmpty(peptideSequence) && string.IsNullOrEmpty(proteinName))
{
continue;
}
if (string.IsNullOrEmpty(peptideSequence))
{
ShowTransitionError(new PasteError
{
Column = colTransitionPeptide.Index,
Line = i,
Message = Resources.PasteDlg_ListPeptideSequences_The_peptide_sequence_cannot_be_blank
});
return null;
}
if (!FastaSequence.IsExSequence(peptideSequence))
{
ShowTransitionError(new PasteError
{
Column = colTransitionPeptide.Index,
Line = i,
Message = Resources.PasteDlg_ListPeptideSequences_This_peptide_sequence_contains_invalid_characters
});
return null;
}
double mz;
if (!double.TryParse(precursorMzText, out mz))
{
ShowTransitionError(new PasteError
{
Column = colTransitionPrecursorMz.Index,
Line = i,
Message = Resources.PasteDlg_AddTransitionList_The_precursor_m_z_must_be_a_number_
});
return null;
}
if (!double.TryParse(productMzText, out mz))
{
ShowTransitionError(new PasteError
{
Column = colTransitionProductMz.Index,
Line = i,
Message = Resources.PasteDlg_AddTransitionList_The_product_m_z_must_be_a_number_
});
return null;
}
const char sep = TRANSITION_LIST_SEPARATOR;
// Add columns in order specified by TRANSITION_LIST_COL_INDICES
sbTransitionList
.Append(proteinName).Append(sep)
.Append(peptideSequence).Append(sep)
.Append(precursorMzText).Append(sep)
.Append(productMzText).AppendLine();
// Build FASTA sequence text in cases where it is known
if (!dictNameSeq.ContainsKey(proteinName))
{
var fastaSeq = backgroundProteome.GetFastaSequence(proteinName);
if (fastaSeq != null)
dictNameSeq.Add(proteinName, fastaSeq);
}
}
if (sbTransitionList.Length == 0)
return document;
// Do the actual import into PeptideGroupDocNodes
IEnumerable<PeptideGroupDocNode> peptideGroupDocNodes;
try
{
List<TransitionImportErrorInfo> errorList;
List<MeasuredRetentionTime> irtPeptides;
List<SpectrumMzInfo> librarySpectra;
var inputs = new MassListInputs(sbTransitionList.ToString(), LocalizationHelper.CurrentCulture, TRANSITION_LIST_SEPARATOR);
var importer = new MassListImporter(document, inputs);
// TODO: support long-wait broker
peptideGroupDocNodes = importer.Import(null,
TRANSITION_LIST_COL_INDICES,
dictNameSeq,
out irtPeptides,
out librarySpectra,
out errorList);
if (errorList.Any())
{
var firstError = errorList[0];
if (firstError.Row.HasValue)
{
throw new LineColNumberedIoException(firstError.ErrorMessage, firstError.Row.Value, firstError.Column ?? -1);
}
else
{
throw new InvalidDataException(firstError.ErrorMessage);
}
}
}
catch (LineColNumberedIoException x)
{
var columns = new[]
{
colTransitionProteinName,
colPeptideSequence,
colTransitionPrecursorMz,
colTransitionProductMz
};
ShowTransitionError(new PasteError
{
Column = x.ColumnIndex >= 0 ? columns[x.ColumnIndex].Index : 0,
Line = (int) x.LineNumber - 1,
Message = x.PlainMessage
});
return null;
}
catch (InvalidDataException x)
{
ShowTransitionError(new PasteError
{
Message = x.Message
});
return null;
}
// Insert the resulting nodes into the document tree, merging when possible
bool after = false;
foreach (var nodePepGroup in peptideGroupDocNodes)
{
PeptideGroupDocNode nodePepGroupExist = FindPeptideGroupDocNode(document, nodePepGroup);
if (nodePepGroupExist != null)
{
var nodePepGroupNew = nodePepGroupExist.Merge(nodePepGroup);
if (!ReferenceEquals(nodePepGroupExist, nodePepGroupNew))
document = (SrmDocument) document.ReplaceChild(nodePepGroupNew);
}
else
{
// Add to the end, if no insert node
var to = selectedPath;
if (to == null || to.Depth < (int) SrmDocument.Level.MoleculeGroups)
document = (SrmDocument) document.Add(nodePepGroup);
else
{
Identity toId = selectedPath.GetIdentity((int) SrmDocument.Level.MoleculeGroups);
document = (SrmDocument) document.Insert(toId, nodePepGroup, after);
}
selectedPath = new IdentityPath(nodePepGroup.Id);
// All future insertions should be after, to avoid reversing the list
after = true;
}
}
}
return document;
}
private SrmDocument AddProteins(SrmDocument document, ref IdentityPath selectedPath)
{
if (tabControl1.SelectedTab != tabPageProteinList)
return document;
var backgroundProteome = GetBackgroundProteome(document);
for (int i = gridViewProteins.Rows.Count - 1; i >= 0; i--)
{
var row = gridViewProteins.Rows[i];
var proteinName = Convert.ToString(row.Cells[colProteinName.Index].Value);
if (String.IsNullOrEmpty(proteinName))
{
continue;
}
var pastedMetadata = new ProteinMetadata(proteinName,
Convert.ToString(row.Cells[colProteinDescription.Index].Value),
NullForEmpty(Convert.ToString(row.Cells[colProteinPreferredName.Index].Value)),
NullForEmpty(Convert.ToString(row.Cells[colProteinAccession.Index].Value)),
NullForEmpty(Convert.ToString(row.Cells[colProteinGene.Index].Value)),
NullForEmpty(Convert.ToString(row.Cells[colProteinSpecies.Index].Value)));
FastaSequence fastaSequence = null;
if (!backgroundProteome.IsNone)
{
ProteinMetadata protdbMetadata;
fastaSequence = backgroundProteome.GetFastaSequence(proteinName, out protdbMetadata);
// Fill in any gaps in pasted metadata with that in protdb
pastedMetadata = pastedMetadata.Merge(protdbMetadata);
}
// Strip any whitespace (tab, newline etc) In case it was copied out of a FASTA file
var fastaSequenceString = new string(Convert.ToString(row.Cells[colProteinSequence.Index].Value).Where(c => !Char.IsWhiteSpace(c)).ToArray());
if (!string.IsNullOrEmpty(fastaSequenceString))
{
try
{
if (fastaSequence == null) // Didn't match anything in protdb
{
fastaSequence = new FastaSequence(pastedMetadata.Name, pastedMetadata.Description,
new ProteinMetadata[0], fastaSequenceString);
}
else
{
if (fastaSequence.Sequence != fastaSequenceString)
{
fastaSequence = new FastaSequence(pastedMetadata.Name, pastedMetadata.Description,
fastaSequence.Alternatives, fastaSequenceString);
}
}
}
catch (Exception exception)
{
ShowProteinError(new PasteError
{
Line = i,
Column = colProteinDescription.Index,
Message = string.Format(Resources.PasteDlg_AddProteins_Invalid_protein_sequence__0__, exception.Message)
});
return null;
}
}
if (fastaSequence == null)
{
ShowProteinError(
new PasteError
{
Line = i,
Message = backgroundProteome.IsNone
? Resources.PasteDlg_AddProteins_Missing_protein_sequence
: Resources.PasteDlg_AddProteins_This_protein_was_not_found_in_the_background_proteome_database
});
return null;
}
var description = pastedMetadata.Description;
if (!string.IsNullOrEmpty(description) && description != fastaSequence.Description)
{
fastaSequence = new FastaSequence(fastaSequence.Name, description, fastaSequence.Alternatives, fastaSequence.Sequence);
}
pastedMetadata = pastedMetadata.ChangeName(fastaSequence.Name).ChangeDescription(fastaSequence.Description); // Make sure these agree
var nodeGroupPep = new PeptideGroupDocNode(fastaSequence, pastedMetadata, new PeptideDocNode[0]);
nodeGroupPep = nodeGroupPep.ChangeSettings(document.Settings, SrmSettingsDiff.ALL);
var to = selectedPath;
if (to == null || to.Depth < (int)SrmDocument.Level.MoleculeGroups)
document = (SrmDocument)document.Add(nodeGroupPep);
else
{
Identity toId = selectedPath.GetIdentity((int)SrmDocument.Level.MoleculeGroups);
document = (SrmDocument)document.Insert(toId, nodeGroupPep);
}
selectedPath = new IdentityPath(nodeGroupPep.Id);
}
return document;
}
public void ImportFastaTest()
{
SrmDocument document = new SrmDocument(SrmSettingsList.GetDefault0_6());
IdentityPath path = IdentityPath.ROOT;
SrmDocument docFasta = document.ImportFasta(new StringReader(ExampleText.TEXT_FASTA_YEAST), false, path, out path);
AssertEx.IsDocumentState(docFasta, 1, 2, 98, 311);
Assert.AreEqual("YAL001C", ((PeptideGroupDocNode)docFasta.Children[0]).Name);
Assert.AreEqual("YAL002W", ((PeptideGroupDocNode)docFasta.Children[1]).Name);
Assert.AreEqual(1, path.Length);
Assert.IsInstanceOfType(path.GetIdentity(0), typeof(FastaSequence));
Assert.AreEqual("YAL001C", ((FastaSequence)path.GetIdentity(0)).Name);
int maxMz = docFasta.Settings.TransitionSettings.Instrument.MaxMz - 120;
foreach (PeptideGroupDocNode nodeGroup in docFasta.Children)
{
Assert.IsInstanceOfType(nodeGroup.Id, typeof(FastaSequence));
int lastEnd = docFasta.Settings.PeptideSettings.Filter.ExcludeNTermAAs - 1;
foreach (PeptideDocNode nodePeptide in nodeGroup.Children)
{
Peptide peptide = nodePeptide.Peptide;
char prev = peptide.PrevAA;
if (prev != 'K' && prev != 'R')
{
Assert.Fail("Unexpected preceding cleavage at {0}", prev);
}
string seq = peptide.Sequence;
char last = seq[seq.Length - 1];
if (last != 'K' && last != 'R' && peptide.NextAA != '-')
{
Assert.Fail("Unexpected cleavage at {0}", last);
}
Assert.IsNotNull(peptide.Begin);
Assert.IsNotNull(peptide.End);
// Make sure peptides are ordered, and not overlapping
if (peptide.Begin.Value < lastEnd)
{
Assert.Fail("Begin {0} less than last end {1}.", peptide.Begin.Value, lastEnd);
}
lastEnd = peptide.End.Value;
IList <DocNode> nodesTrans = ((DocNodeParent)nodePeptide.Children[0]).Children;
int trans = nodesTrans.Count;
if (trans < 3)
{
// Might have been cut off by the instrument limit.
if ((trans == 0 && ((TransitionGroupDocNode)nodePeptide.Children[0]).PrecursorMz < maxMz) ||
(trans > 0 && ((TransitionDocNode)nodesTrans[0]).Mz < maxMz))
{
Assert.Fail("Found {0} transitions, expecting 3.", trans);
}
}
// Might have extra proline transitions
else if (trans > 3 && peptide.Sequence.IndexOf('P') == -1)
{
Assert.Fail("Found {0} transitions, expecting 3.", trans);
}
// Make sure transitions are ordered correctly
IonType lastType = IonType.a;
int lastOffset = -1;
foreach (TransitionDocNode nodeTran in nodesTrans)
{
Transition transition = nodeTran.Transition;
if (lastType == transition.IonType)
{
Assert.IsTrue(transition.CleavageOffset > lastOffset);
}
else
{
Assert.IsTrue(((int)transition.IonType) > ((int)lastType));
}
lastType = transition.IonType;
lastOffset = transition.CleavageOffset;
}
}
}
// Make sure old document is unmodified.
Assert.AreEqual(0, document.RevisionIndex);
Assert.AreEqual(0, document.PeptideTransitionCount);
// Re-paste of fasta should have no impact.
// path = IdentityPath.ROOT; use null as substitute for Root
SrmDocument docFasta2 = docFasta.ImportFasta(new StringReader(ExampleText.TEXT_FASTA_YEAST), false, null, out path);
// Returns the original document to avoid adding undo record in running app
Assert.AreSame(docFasta, docFasta2);
Assert.IsNull(path);
// Discard double-insert document, and add peptides list into previous document
path = IdentityPath.ROOT;
SrmDocument docPeptides = docFasta.ImportFasta(new StringReader(TEXT_BOVINE_PEPTIDES1), true, path, out path);
AssertEx.IsDocumentState(docPeptides, 2, 3, 111, 352);
Assert.AreEqual(1, path.Length);
Assert.IsNotInstanceOfType(path.GetIdentity(0), typeof(FastaSequence));
Assert.AreEqual("Peptides1", ((PeptideGroupDocNode)docPeptides.FindNode(path)).Name);
PeptideGroupDocNode nodePepList = (PeptideGroupDocNode)docPeptides.Children[2];
Assert.IsNotInstanceOfType(nodePepList.Id, typeof(FastaSequence));
// Make sure other two nodes are unchanged
Assert.AreSame(docFasta.Children[0], docPeptides.Children[0]);
Assert.AreSame(docFasta.Children[1], docPeptides.Children[1]);
foreach (PeptideDocNode nodePeptide in nodePepList.Children)
{
char prev = nodePeptide.Peptide.PrevAA;
char next = nodePeptide.Peptide.NextAA;
if (prev != 'X' || next != 'X')
{
Assert.Fail("Expected amino acids X, but found {0} or {1}", prev, next);
}
string seq = nodePeptide.Peptide.Sequence;
char last = seq[seq.Length - 1];
// Just because they are tryptic peptides in the list
if (last != 'K' && last != 'R' && nodePeptide.Peptide.NextAA != '-')
{
Assert.Fail("Unexpected cleavage at {0}", last);
}
Assert.IsNull(nodePeptide.Peptide.Begin);
Assert.IsNull(nodePeptide.Peptide.End);
IList <DocNode> nodesTrans = ((DocNodeParent)nodePeptide.Children[0]).Children;
int trans = nodesTrans.Count;
if (trans < 3)
{
// Might have been cut off by the instrument limit.
if ((trans == 0 && ((TransitionGroupDocNode)nodePeptide.Children[0]).PrecursorMz < maxMz) ||
(trans > 0 && ((TransitionDocNode)nodesTrans[0]).Mz < maxMz))
{
Assert.Fail("Found {0} transitions, expecting 3.", trans);
}
}
// Might have extra proline transitions
else if (trans > 3 && nodePeptide.Peptide.Sequence.IndexOf('P') == -1)
{
Assert.Fail("Found {0} transitions, expecting 3.", trans);
}
}
// Make sure old documents are unmodified.
AssertEx.IsDocumentState(document, 0, 0, 0, 0);
AssertEx.IsDocumentState(docFasta, 1, 2, 98, 311);
AssertEx.IsDocumentState(docPeptides, 2, 3, 111, 352);
// Add peptides in all possible locations.
// 1. Root (already done)
// 1. Before another group
path = docPeptides.GetPathTo(0);
SrmDocument docPeptides2 = docPeptides.ImportFasta(new StringReader(TEXT_BOVINE_PEPTIDES1), true, path, out path);
AssertEx.IsDocumentState(docPeptides2, 3, 4, 124, 393);
Assert.IsNotInstanceOfType(docPeptides2.Children[0].Id, typeof(FastaSequence));
Assert.AreEqual(docPeptides2.Children[0].Id, path.GetIdentity(0));
Assert.IsInstanceOfType(docPeptides2.Children[1].Id, typeof(FastaSequence));
// Make sure previously existing groups are unchanged
Assert.AreSame(docPeptides.Children[0], docPeptides2.Children[1]);
Assert.AreSame(docPeptides.Children[1], docPeptides2.Children[2]);
Assert.AreSame(docPeptides.Children[2], docPeptides2.Children[3]);
// 2. Inside a FASTA group
path = docPeptides2.GetPathTo((int)SrmDocument.Level.Transitions, 100);
SrmDocument docPeptides3 = docPeptides2.ImportFasta(new StringReader(TEXT_BOVINE_PEPTIDES1), true, path, out path);
AssertEx.IsDocumentState(docPeptides3, 4, 5, 137, 434);
Assert.AreEqual(2, docPeptides3.FindNodeIndex(path));
// Make sure previously existing groups are unchanged
Assert.AreSame(docPeptides2.Children[1], docPeptides3.Children[1]);
Assert.AreSame(docPeptides2.Children[2], docPeptides3.Children[3]);
// 3. To a peptide list
// a. Same peptides
path = docPeptides2.GetPathTo(0);
docPeptides3 = docPeptides2.ImportFasta(new StringReader(TEXT_BOVINE_PEPTIDES1), true, path, out path);
// No longer filter repeated peptides, because they are useful for explicit modifictations.
Assert.AreNotSame(docPeptides2, docPeptides3);
Assert.IsNotNull(path);
// b. Different paptides
path = docPeptides2.GetPathTo(0);
IdentityPath pathFirstPep = docPeptides3.GetPathTo((int)SrmDocument.Level.Molecules, 0);
docPeptides3 = docPeptides2.ImportFasta(new StringReader(TEXT_BOVINE_PEPTIDES2), true, path, out path);
AssertEx.IsDocumentState(docPeptides3, 4, 4, 140, 448);
Assert.AreSame(docPeptides2.Children[0].Id, docPeptides3.Children[0].Id);
Assert.AreNotSame(docPeptides2.Children[0], docPeptides3.Children[0]);
Assert.AreEqual("LVTDLTK", ((PeptideDocNode)docPeptides3.FindNode(path)).Peptide.Sequence);
int index = docPeptides3.FindNodeIndex(path);
IdentityPath pathPreceding = docPeptides3.GetPathTo(path.Depth, index - 1);
Assert.AreEqual("IVGYLDEEGVLDQNR", ((PeptideDocNode)docPeptides3.FindNode(pathPreceding)).Peptide.Sequence);
Assert.AreEqual(0, docPeptides3.FindNodeIndex(pathFirstPep));
// 4. At a peptide in a peptide list
path = docPeptides2.GetPathTo((int)SrmDocument.Level.Molecules, 0);
docPeptides3 = docPeptides2.ImportFasta(new StringReader(TEXT_BOVINE_PEPTIDES2), true, path, out path);
AssertEx.IsDocumentState(docPeptides3, 4, 4, 140, 448);
Assert.AreSame(docPeptides2.Children[0].Id, docPeptides3.Children[0].Id);
Assert.AreNotSame(docPeptides2.Children[0], docPeptides3.Children[0]);
Assert.AreEqual(0, docPeptides3.FindNodeIndex(path));
Assert.AreEqual(16, docPeptides3.FindNodeIndex(pathFirstPep));
// 5. Inside a peptide in a peptide list
path = docPeptides2.GetPathTo((int)SrmDocument.Level.Transitions, 0);
docPeptides3 = docPeptides2.ImportFasta(new StringReader(TEXT_BOVINE_PEPTIDES2), true, path, out path);
AssertEx.IsDocumentState(docPeptides3, 4, 4, 140, 448);
Assert.AreSame(docPeptides2.Children[0].Id, docPeptides3.Children[0].Id);
Assert.AreNotSame(docPeptides2.Children[0], docPeptides3.Children[0]);
Assert.AreEqual(1, docPeptides3.FindNodeIndex(path));
Assert.AreEqual(0, docPeptides3.FindNodeIndex(pathFirstPep));
}
/// <summary>
/// Adds a list of PeptideDocNodes found in the library to the current document.
/// </summary>
public SrmDocument AddPeptides(SrmDocument document, ILongWaitBroker broker, IdentityPath toPath, out IdentityPath selectedPath)
{
if (toPath != null &&
toPath.Depth == (int)SrmDocument.Level.MoleculeGroups &&
ReferenceEquals(toPath.GetIdentity((int)SrmDocument.Level.MoleculeGroups), SequenceTree.NODE_INSERT_ID))
{
toPath = null;
}
SkippedPeptideCount = 0;
var dictCopy = new Dictionary <PeptideSequenceModKey, PeptideMatch>();
// Make heavy mods explicit
if (PeptideMatches.Values.Contains(match => match.NodePep.HasExplicitMods &&
match.NodePep.ExplicitMods.HeavyModifications != null))
{
_matcher.ConvertAllHeavyModsExplicit();
}
// Call ensure mods on all peptides to be added to the document.
var listDefStatMods = new MappedList <string, StaticMod>();
listDefStatMods.AddRange(Properties.Settings.Default.StaticModList);
listDefStatMods.AddRange(document.Settings.PeptideSettings.Modifications.StaticModifications);
var listDefHeavyMods = new MappedList <string, StaticMod>();
listDefHeavyMods.AddRange(Properties.Settings.Default.HeavyModList);
listDefHeavyMods.AddRange(document.Settings.PeptideSettings.Modifications.AllHeavyModifications);
foreach (var key in PeptideMatches.Keys)
{
var match = PeptideMatches[key];
var nodePepDocSet = match.NodePep;
if (_matcher.MatcherPepMods != null)
{
nodePepDocSet = match.NodePep.EnsureMods(_matcher.MatcherPepMods,
document.Settings.PeptideSettings.Modifications,
listDefStatMods, listDefHeavyMods);
}
if (!dictCopy.ContainsKey(nodePepDocSet.SequenceKey))
{
dictCopy.Add(nodePepDocSet.SequenceKey,
new PeptideMatch(nodePepDocSet, match.Proteins,
match.MatchesFilterSettings));
}
}
if (!Properties.Settings.Default.LibraryPeptidesKeepFiltered)
{
// TODO: This removes entire peptides where only a single
// precursor does not match. e.g. the library contains
// a singly charged precursor match, but also doubly charged
dictCopy = dictCopy.Where(match => match.Value.MatchesFilterSettings)
.ToDictionary(match => match.Key, match => match.Value);
}
SrmDocument newDocument = UpdateExistingPeptides(document, dictCopy, toPath, out selectedPath);
toPath = selectedPath;
// If there is an associated background proteome, add peptides that can be
// matched to the proteins from the background proteom.
if (_backgroundProteome != null)
{
newDocument = AddProteomePeptides(newDocument, dictCopy, broker,
toPath, out selectedPath);
}
toPath = selectedPath;
// Add all remaining peptides as a peptide list.
if (_backgroundProteome == null || Properties.Settings.Default.LibraryPeptidesAddUnmatched)
{
var listPeptidesToAdd = dictCopy.Values.ToList();
listPeptidesToAdd.RemoveAll(match => match.Proteins != null && match.Proteins.Count > 0);
if (listPeptidesToAdd.Count > 0)
{
newDocument = AddPeptidesToLibraryGroup(newDocument, listPeptidesToAdd, broker,
toPath, out selectedPath);
if (listPeptidesToAdd.Count > 1000)
{
selectedPath = selectedPath.Parent; // Don't force Skyline to open a massive peptide list, if it wouldn't otherwise
}
}
}
return(newDocument);
}
/// <summary>
/// Adds a list of PeptideDocNodes found in the library to the current document.
/// </summary>
public SrmDocument AddPeptides(SrmDocument document, ILongWaitBroker broker, IdentityPath toPath, out IdentityPath selectedPath)
{
if (toPath != null &&
toPath.Depth == (int)SrmDocument.Level.MoleculeGroups &&
ReferenceEquals(toPath.GetIdentity((int)SrmDocument.Level.MoleculeGroups), SequenceTree.NODE_INSERT_ID))
{
toPath = null;
}
SkippedPeptideCount = 0;
var dictCopy = new Dictionary<PeptideSequenceModKey, PeptideMatch>();
// Make heavy mods explicit
if (PeptideMatches.Values.Contains(match => match.NodePep.HasExplicitMods
&& match.NodePep.ExplicitMods.HeavyModifications != null))
{
_matcher.ConvertAllHeavyModsExplicit();
}
// Call ensure mods on all peptides to be added to the document.
var listDefStatMods = new MappedList<string, StaticMod>();
listDefStatMods.AddRange(Properties.Settings.Default.StaticModList);
listDefStatMods.AddRange(document.Settings.PeptideSettings.Modifications.StaticModifications);
var listDefHeavyMods = new MappedList<string, StaticMod>();
listDefHeavyMods.AddRange(Properties.Settings.Default.HeavyModList);
listDefHeavyMods.AddRange(document.Settings.PeptideSettings.Modifications.HeavyModifications);
foreach (var key in PeptideMatches.Keys)
{
var match = PeptideMatches[key];
var nodePepDocSet = match.NodePep;
if (_matcher.MatcherPepMods != null)
nodePepDocSet = match.NodePep.EnsureMods(_matcher.MatcherPepMods,
document.Settings.PeptideSettings.Modifications,
listDefStatMods, listDefHeavyMods);
if (!dictCopy.ContainsKey(nodePepDocSet.SequenceKey))
dictCopy.Add(nodePepDocSet.SequenceKey,
new PeptideMatch(nodePepDocSet, match.Proteins,
match.MatchesFilterSettings));
}
if (!Properties.Settings.Default.LibraryPeptidesKeepFiltered)
{
// TODO: This removes entire peptides where only a single
// precursor does not match. e.g. the library contains
// a singly charged precursor match, but also doubly charged
dictCopy = dictCopy.Where(match => match.Value.MatchesFilterSettings)
.ToDictionary(match => match.Key, match => match.Value);
}
SrmDocument newDocument = UpdateExistingPeptides(document, dictCopy, toPath, out selectedPath);
toPath = selectedPath;
// If there is an associated background proteome, add peptides that can be
// matched to the proteins from the background proteom.
if (_backgroundProteome != null)
{
newDocument = AddProteomePeptides(newDocument, dictCopy, broker,
toPath, out selectedPath);
}
toPath = selectedPath;
// Add all remaining peptides as a peptide list.
if (_backgroundProteome == null || Properties.Settings.Default.LibraryPeptidesAddUnmatched)
{
var listPeptidesToAdd = dictCopy.Values.ToList();
listPeptidesToAdd.RemoveAll(match => match.Proteins != null && match.Proteins.Count > 0);
if (listPeptidesToAdd.Count > 0)
{
newDocument = AddPeptidesToLibraryGroup(newDocument, listPeptidesToAdd, broker,
toPath, out selectedPath);
}
}
return newDocument;
}
public void IdentityPathTest()
{
SimpleIdentity id = new SimpleIdentity();
IdentityPath pathSimple = new IdentityPath(id);
Assert.AreEqual(IdentityPath.ROOT, pathSimple.Parent);
Assert.AreEqual(IdentityPath.ROOT, pathSimple.GetPathTo(-1));
Assert.AreEqual(id, pathSimple.Child);
Assert.AreEqual(1, pathSimple.Length);
Assert.AreEqual(0, pathSimple.Depth);
AssertEx.ThrowsException<IndexOutOfRangeException>(() => pathSimple.GetPathTo(-2));
AssertEx.ThrowsException<IndexOutOfRangeException>(() => pathSimple.GetPathTo(1));
IdentityPath pathRoot = new IdentityPath(new Identity[0]);
Assert.AreEqual(IdentityPath.ROOT, pathRoot);
Assert.AreEqual(0, pathRoot.Length);
Assert.AreEqual(-1, pathRoot.Depth);
AssertEx.ThrowsException<IndexOutOfRangeException>(() => pathRoot.Child);
AssertEx.ThrowsException<IndexOutOfRangeException>(() => pathRoot.Parent);
const int count = 5;
List<Identity> listId1 = new List<Identity>();
List<Identity> listId2 = new List<Identity>();
HashSet<IdentityPath> setPaths = new HashSet<IdentityPath>();
IdentityPath last = IdentityPath.ROOT;
for (int i = 0; i < count; i++)
{
listId1.Add(new NumericIdentity(i));
listId2.Add(new NumericIdentity(i));
IdentityPath path = new IdentityPath(listId1);
Assert.AreEqual(path, new IdentityPath(listId1));
Assert.AreNotEqual(path, new IdentityPath(listId2));
Assert.AreEqual(last, path.Parent);
Assert.AreSame(listId1[i], path.Child);
Assert.AreEqual(path, new IdentityPath(path.Parent, path.Child));
Assert.AreEqual(i, path.Depth);
Assert.AreEqual(listId1.Count, path.Length);
Assert.AreSame(path.Child, path.GetIdentity(path.Depth));
Assert.AreEqual(path.Parent, path.GetPathTo(path.Depth - 1));
Assert.AreEqual("/" + listId1.ToString("/"), path.ToString());
for (int j = 0; j < i; j++)
Assert.IsTrue(setPaths.Contains(path.GetPathTo(j)));
setPaths.Add(path);
last = path;
}
}