public void AddRemoveExplicitModTest()
{
SrmDocument docStudy7 = CreateStudy7Doc();
string transitionList = ExportCsv(docStudy7);
Assert.AreEqual(69, transitionList.Split('\n').Length); // Special test mode may add an extra doc node
var modifications = docStudy7.Settings.PeptideSettings.Modifications;
var listStaticMods = modifications.StaticModifications;
var listHeavyMods = modifications.AllHeavyModifications.ToList();
docStudy7 = docStudy7.ChangeSettings(docStudy7.Settings.ChangeTransitionFilter(filter => filter.ChangeAutoSelect(false)));
// Remove all modifications
int i = 0;
// But save them for later
var removedMods = new Dictionary <int, ExplicitMods>();
foreach (var peptide in docStudy7.Peptides)
{
if (peptide.HasExplicitMods)
{
removedMods.Add(i, peptide.ExplicitMods);
IdentityPath path = docStudy7.GetPathTo((int)SrmDocument.Level.Molecules, i);
docStudy7 = docStudy7.ChangePeptideMods(path, null, listStaticMods, listHeavyMods);
}
i++;
}
// Removes heavy from peptide with c-terminal P
AssertEx.IsDocumentState(docStudy7, 6, 7, 11, 21, 63);
modifications = docStudy7.Settings.PeptideSettings.Modifications;
Assert.AreEqual(2, modifications.AllHeavyModifications.Count());
Assert.AreEqual(0, modifications.AllHeavyModifications.Count(mod => mod.IsExplicit));
Assert.AreEqual(0, docStudy7.Peptides.Count(peptide => peptide.HasExplicitMods));
listHeavyMods = ATOMIC_HEAVY_MODS;
foreach (var pair in removedMods)
{
IdentityPath path = docStudy7.GetPathTo((int)SrmDocument.Level.Molecules, pair.Key);
docStudy7 = docStudy7.ChangePeptideMods(path, pair.Value, listStaticMods, listHeavyMods);
}
AssertEx.IsDocumentState(docStudy7, 11, 7, 11, 21, 63);
// Replace the heavy precursor that was removed
// TODO: Yuck. Would be nice to have a way to do this without duplicating
// so much of the logic in PeptideDocNode and PeptideTreeNode
var pepPath = docStudy7.GetPathTo((int)SrmDocument.Level.Molecules, 10);
var nodePep = (PeptideDocNode)docStudy7.FindNode(pepPath);
var mods = nodePep.ExplicitMods;
var nodeGroupLight = (TransitionGroupDocNode)nodePep.Children[0];
var settings = docStudy7.Settings;
foreach (var tranGroup in nodePep.GetTransitionGroups(settings, mods, false))
{
if (tranGroup.PrecursorAdduct.Equals(nodeGroupLight.TransitionGroup.PrecursorAdduct) &&
!tranGroup.LabelType.IsLight)
{
TransitionDocNode[] transitions = nodePep.GetMatchingTransitions(tranGroup, settings, mods);
var nodeGroup = new TransitionGroupDocNode(tranGroup, transitions);
nodeGroup = nodeGroup.ChangeSettings(settings, nodePep, mods, SrmSettingsDiff.ALL);
docStudy7 = (SrmDocument)docStudy7.Add(pepPath, nodeGroup);
break;
}
}
AssertEx.IsDocumentState(docStudy7, 12, 7, 11, 22, 66);
modifications = docStudy7.Settings.PeptideSettings.Modifications;
Assert.AreEqual(2, modifications.AllHeavyModifications.Count(mod => mod.IsExplicit && mod.Label13C));
Assert.AreEqual(2, modifications.AllHeavyModifications.Count(mod => mod.Formula != null));
Assert.AreEqual(3, docStudy7.Peptides.Count(peptide => peptide.HasExplicitMods));
Assert.AreEqual(2, docStudy7.Peptides.Count(peptide => peptide.HasExplicitMods &&
peptide.ExplicitMods.StaticModifications.Count > 0 &&
peptide.ExplicitMods.StaticModifications[0].Modification.AAs[0] == 'C'));
Assert.AreEqual(2, docStudy7.Peptides.Count(peptide => peptide.HasExplicitMods &&
peptide.ExplicitMods.HeavyModifications[0].Modification.AAs[0] == 'V' &&
peptide.ExplicitMods.HeavyModifications[0].Modification.Label13C));
Assert.AreEqual(1, docStudy7.Peptides.Count(peptide => peptide.HasExplicitMods &&
peptide.ExplicitMods.HeavyModifications[0].Modification.AAs[0] == 'L' &&
peptide.ExplicitMods.HeavyModifications[0].Modification.Label13C));
AssertEx.NoDiff(transitionList, ExportCsv(docStudy7));
}
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;
}
private static SrmDocument GenerateDecoysFunc(SrmDocument document, int numDecoys, bool multiCycle,
Func<SequenceMods, SequenceMods> genDecoySequence)
{
// Loop through the existing tree in random order creating decoys
var settings = document.Settings;
var enzyme = settings.PeptideSettings.Enzyme;
var decoyNodePepList = new List<PeptideDocNode>();
var setDecoyKeys = new HashSet<PeptideModKey>();
while (numDecoys > 0)
{
int startDecoys = numDecoys;
foreach (var nodePep in document.Peptides.ToArray().RandomOrder())
{
if (numDecoys == 0)
break;
// Decoys should not be based on standard peptides
if (nodePep.GlobalStandardType != null)
continue;
// If the non-terminal end of the peptide sequence is all a single character, skip this peptide,
// since it can't support decoy generation.
var sequence = nodePep.Peptide.Sequence;
if (genDecoySequence != null && sequence.Substring(0, sequence.Length - 1).Distinct().Count() == 1)
continue;
var seqMods = new SequenceMods(nodePep);
if (genDecoySequence != null)
{
seqMods = genDecoySequence(seqMods);
}
var peptide = nodePep.Peptide;
var decoyPeptide = new Peptide(null, seqMods.Sequence, null, null, enzyme.CountCleavagePoints(seqMods.Sequence), true);
if (seqMods.Mods != null)
seqMods.Mods = seqMods.Mods.ChangePeptide(decoyPeptide);
foreach (var comparableGroups in PeakFeatureEnumerator.ComparableGroups(nodePep))
{
var decoyNodeTranGroupList = GetDecoyGroups(nodePep, decoyPeptide, seqMods.Mods, comparableGroups, document,
Equals(seqMods.Sequence, peptide.Sequence));
if (decoyNodeTranGroupList.Count == 0)
continue;
var nodePepNew = new PeptideDocNode(decoyPeptide, settings, seqMods.Mods,
null, nodePep.ExplicitRetentionTime, decoyNodeTranGroupList.ToArray(), false);
if (!Equals(nodePep.ModifiedSequence, nodePepNew.ModifiedSequence))
{
var sourceKey = new ModifiedSequenceMods(nodePep.ModifiedSequence, nodePep.ExplicitMods);
nodePepNew = nodePepNew.ChangeSourceKey(sourceKey);
}
// Avoid adding duplicate peptides
if (setDecoyKeys.Contains(nodePepNew.Key))
continue;
setDecoyKeys.Add(nodePepNew.Key);
decoyNodePepList.Add(nodePepNew);
numDecoys--;
}
}
// Stop if not multi-cycle or the number of decoys has not changed.
if (!multiCycle || startDecoys == numDecoys)
break;
}
var decoyNodePepGroup = new PeptideGroupDocNode(new PeptideGroup(true), Annotations.EMPTY, PeptideGroup.DECOYS,
null, decoyNodePepList.ToArray(), false);
decoyNodePepGroup = decoyNodePepGroup.ChangeSettings(document.Settings, SrmSettingsDiff.ALL);
return (SrmDocument)document.Add(decoyNodePepGroup);
}
public SrmDocument ConvertToSmallMolecules(SrmDocument document,
ConvertToSmallMoleculesMode mode = ConvertToSmallMoleculesMode.formulas,
bool invertCharges = false,
bool ignoreDecoys=false)
{
if (mode == ConvertToSmallMoleculesMode.none)
return document;
var newdoc = new SrmDocument(document.Settings);
var note = new Annotations(TestingConvertedFromProteomic, null, 1); // Mark this as a testing node so we don't sort it
newdoc = (SrmDocument)newdoc.ChangeIgnoreChangingChildren(true); // Retain copied results
foreach (var peptideGroupDocNode in document.MoleculeGroups)
{
if (!peptideGroupDocNode.IsProteomic)
{
newdoc = (SrmDocument)newdoc.Add(peptideGroupDocNode); // Already a small molecule
}
else
{
var newPeptideGroup = new PeptideGroup();
var newPeptideGroupDocNode = new PeptideGroupDocNode(newPeptideGroup,
peptideGroupDocNode.Annotations.Merge(note), peptideGroupDocNode.Name,
peptideGroupDocNode.Description, new PeptideDocNode[0],
peptideGroupDocNode.AutoManageChildren);
foreach (var mol in peptideGroupDocNode.Molecules)
{
var peptideSequence = mol.Peptide.Sequence;
// Create a PeptideDocNode with the presumably baseline charge and label
var precursorCharge = (mol.TransitionGroups.Any() ? mol.TransitionGroups.First().TransitionGroup.PrecursorCharge : 0) * (invertCharges ? -1 : 1);
var isotopeLabelType = mol.TransitionGroups.Any() ? mol.TransitionGroups.First().TransitionGroup.LabelType : IsotopeLabelType.light;
var moleculeCustomIon = ConvertToSmallMolecule(mode, document, mol, precursorCharge, isotopeLabelType);
var precursorCustomIon = moleculeCustomIon;
var newPeptide = new Peptide(moleculeCustomIon);
var newPeptideDocNode = new PeptideDocNode(newPeptide, newdoc.Settings, null, null,
null, null, mol.ExplicitRetentionTime, note, mol.Results, new TransitionGroupDocNode[0],
mol.AutoManageChildren);
foreach (var transitionGroupDocNode in mol.TransitionGroups)
{
if (transitionGroupDocNode.IsDecoy)
{
if (ignoreDecoys)
continue;
throw new Exception("There is no translation from decoy to small molecules"); // Not L10N
}
if (transitionGroupDocNode.TransitionGroup.PrecursorCharge != Math.Abs(precursorCharge) ||
!Equals(isotopeLabelType, transitionGroupDocNode.TransitionGroup.LabelType))
{
// Different charges or labels mean different ion formulas
precursorCharge = transitionGroupDocNode.TransitionGroup.PrecursorCharge * (invertCharges ? -1 : 1);
isotopeLabelType = transitionGroupDocNode.TransitionGroup.LabelType;
precursorCustomIon = ConvertToSmallMolecule(mode, document, mol, precursorCharge, isotopeLabelType);
}
var newTransitionGroup = new TransitionGroup(newPeptide, precursorCustomIon, precursorCharge, isotopeLabelType);
// Remove any library info, since for the moment at least small molecules don't support this and it won't roundtrip
var resultsNew = RemoveTransitionGroupChromInfoLibraryInfo(transitionGroupDocNode);
var newTransitionGroupDocNode = new TransitionGroupDocNode(newTransitionGroup,
transitionGroupDocNode.Annotations.Merge(note), document.Settings,
null, null, transitionGroupDocNode.ExplicitValues, resultsNew, null,
transitionGroupDocNode.AutoManageChildren);
var mzShift = invertCharges ? 2.0 * BioMassCalc.MassProton : 0; // We removed hydrogen rather than added
Assume.IsTrue((Math.Abs(newTransitionGroupDocNode.PrecursorMz + mzShift - transitionGroupDocNode.PrecursorMz) - Math.Abs(transitionGroupDocNode.TransitionGroup.PrecursorCharge * BioMassCalc.MassElectron)) <= 1E-5);
foreach (var transition in transitionGroupDocNode.Transitions)
{
double mass = 0;
var transitionCharge = transition.Transition.Charge * (invertCharges ? -1 : 1);
var ionType = IonType.custom;
CustomIon transitionCustomIon;
double mzShiftTransition = 0;
if (transition.Transition.IonType == IonType.precursor)
{
ionType = IonType.precursor;
transitionCustomIon = new DocNodeCustomIon(precursorCustomIon.Formula,
string.IsNullOrEmpty(precursorCustomIon.Formula) ? precursorCustomIon.MonoisotopicMass : (double?) null,
string.IsNullOrEmpty(precursorCustomIon.Formula) ? precursorCustomIon.AverageMass : (double?) null,
SmallMoleculeNameFromPeptide(peptideSequence, transitionCharge));
mzShiftTransition = invertCharges ? 2.0 * BioMassCalc.MassProton : 0; // We removed hydrogen rather than added
}
else if (transition.Transition.IonType == IonType.custom)
{
transitionCustomIon = transition.Transition.CustomIon;
mass = transitionCustomIon.MonoisotopicMass;
}
else
{
// TODO - try to get fragment formula?
mass = BioMassCalc.CalculateIonMassFromMz(transition.Mz, transition.Transition.Charge);
transitionCustomIon = new DocNodeCustomIon(mass, mass,// We can't really get at mono vs average mass from m/z, but for test purposes this is fine
transition.Transition.FragmentIonName);
}
if (mode == ConvertToSmallMoleculesMode.masses_and_names)
{
// Discard the formula if we're testing the use of mass-with-names (for matching in ratio calcs) target specification
transitionCustomIon = new DocNodeCustomIon(transitionCustomIon.MonoisotopicMass, transitionCustomIon.AverageMass,
transition.Transition.FragmentIonName);
}
else if (mode == ConvertToSmallMoleculesMode.masses_only)
{
// Discard the formula and name if we're testing the use of mass-only target specification
transitionCustomIon = new DocNodeCustomIon(transitionCustomIon.MonoisotopicMass, transitionCustomIon.AverageMass);
}
var newTransition = new Transition(newTransitionGroup, ionType,
null, transition.Transition.MassIndex, transition.Transition.Charge * (invertCharges ? -1 : 1), null,
transitionCustomIon);
if (ionType == IonType.precursor)
{
mass = document.Settings.GetFragmentMass(transitionGroupDocNode.TransitionGroup.LabelType, null, newTransition, newTransitionGroupDocNode.IsotopeDist);
}
var newTransitionDocNode = new TransitionDocNode(newTransition, transition.Annotations.Merge(note),
null, mass, transition.IsotopeDistInfo, null,
transition.Results);
Assume.IsTrue((Math.Abs(newTransitionDocNode.Mz + mzShiftTransition - transition.Mz) - Math.Abs(transitionGroupDocNode.TransitionGroup.PrecursorCharge * BioMassCalc.MassElectron)) <= 1E-5, String.Format("unexpected mz difference {0}-{1}={2}", newTransitionDocNode.Mz , transition.Mz, newTransitionDocNode.Mz - transition.Mz)); // Not L10N
newTransitionGroupDocNode =
(TransitionGroupDocNode)newTransitionGroupDocNode.Add(newTransitionDocNode);
}
if (newPeptideDocNode != null)
newPeptideDocNode = (PeptideDocNode)newPeptideDocNode.Add(newTransitionGroupDocNode);
}
newPeptideGroupDocNode =
(PeptideGroupDocNode)newPeptideGroupDocNode.Add(newPeptideDocNode);
}
newdoc = (SrmDocument)newdoc.Add(newPeptideGroupDocNode);
}
}
// No retention time prediction for small molecules (yet?)
newdoc = newdoc.ChangeSettings(newdoc.Settings.ChangePeptideSettings(newdoc.Settings.PeptideSettings.ChangePrediction(
newdoc.Settings.PeptideSettings.Prediction.ChangeRetentionTime(null))));
return newdoc;
}