public SrmDocument Import(TextReader reader, IProgressMonitor progressMonitor, long lineCount, bool isMinutes, bool removeMissing = false, bool changePeaks = true)
{
IProgressStatus status = new ProgressStatus(Resources.PeakBoundaryImporter_Import_Importing_Peak_Boundaries);
double timeConversionFactor = isMinutes ? 1.0 : 60.0;
int linesRead = 0;
int progressPercent = 0;
var docNew = (SrmDocument)Document.ChangeIgnoreChangingChildren(true);
var docReference = docNew;
var sequenceToNode = MakeSequenceDictionary(Document);
var fileNameToFileMatch = new Dictionary <string, ChromSetFileMatch>();
var trackAdjustedResults = new HashSet <ResultsKey>();
var modMatcher = new ModificationMatcher();
var canonicalSequenceDict = new Dictionary <string, string>();
// Add annotations as possible columns
var allFieldNames = new List <string[]>(FIELD_NAMES);
allFieldNames.AddRange(from def in Document.Settings.DataSettings.AnnotationDefs
where def.AnnotationTargets.Contains(AnnotationDef.AnnotationTarget.precursor_result)
select new[] { def.Name });
string line = reader.ReadLine();
linesRead++;
int[] fieldIndices;
int fieldsTotal;
// If we aren't changing peaks, allow start and end time to be missing
var requiredFields = changePeaks ? REQUIRED_FIELDS : REQUIRED_NO_CHROM;
char correctSeparator = ReadFirstLine(line, allFieldNames, requiredFields, out fieldIndices, out fieldsTotal);
while ((line = reader.ReadLine()) != null)
{
linesRead++;
if (progressMonitor != null)
{
if (progressMonitor.IsCanceled)
{
return(Document);
}
int progressNew = (int)(linesRead * 100 / lineCount);
if (progressPercent != progressNew)
{
progressMonitor.UpdateProgress(status = status.ChangePercentComplete(progressNew));
progressPercent = progressNew;
}
}
var dataFields = new DataFields(fieldIndices, line.ParseDsvFields(correctSeparator), allFieldNames);
if (dataFields.Length != fieldsTotal)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Line__0__field_count__1__differs_from_the_first_line__which_has__2_,
linesRead, dataFields.Length, fieldsTotal));
}
string modifiedPeptideString = dataFields.GetField(Field.modified_peptide);
string fileName = dataFields.GetField(Field.filename);
bool isDecoy = dataFields.IsDecoy(linesRead);
IList <IdentityPath> pepPaths;
if (!sequenceToNode.TryGetValue(Tuple.Create(modifiedPeptideString, isDecoy), out pepPaths))
{
string canonicalSequence;
if (!canonicalSequenceDict.TryGetValue(modifiedPeptideString, out canonicalSequence))
{
if (modifiedPeptideString.Any(c => c < 'A' || c > 'Z'))
{
modMatcher.CreateMatches(Document.Settings,
new List <string> {
modifiedPeptideString
},
Settings.Default.StaticModList,
Settings.Default.HeavyModList);
var nodeForModPep = modMatcher.GetModifiedNode(modifiedPeptideString);
if (nodeForModPep == null)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Peptide_has_unrecognized_modifications__0__at_line__1_, modifiedPeptideString, linesRead));
}
nodeForModPep = nodeForModPep.ChangeSettings(Document.Settings, SrmSettingsDiff.ALL);
// Convert the modified peptide string into a standardized form that
// converts unimod, names, etc, into masses, eg [+57.0]
canonicalSequence = nodeForModPep.ModifiedTarget.Sequence;
canonicalSequenceDict.Add(modifiedPeptideString, canonicalSequence);
}
}
if (null != canonicalSequence)
{
sequenceToNode.TryGetValue(Tuple.Create(canonicalSequence, isDecoy), out pepPaths);
}
}
if (null == pepPaths)
{
UnrecognizedPeptides.Add(modifiedPeptideString);
continue;
}
Adduct charge;
bool chargeSpecified = dataFields.TryGetCharge(linesRead, out charge);
string sampleName = dataFields.GetField(Field.sample_name);
double?apexTime = dataFields.GetTime(Field.apex_time, timeConversionFactor,
Resources.PeakBoundaryImporter_Import_The_value___0___on_line__1__is_not_a_valid_time_, linesRead);
double?startTime = dataFields.GetTime(Field.start_time, timeConversionFactor,
Resources.PeakBoundaryImporter_Import_The_value___0___on_line__1__is_not_a_valid_start_time_, linesRead);
double?endTime = dataFields.GetTime(Field.end_time, timeConversionFactor,
Resources.PeakBoundaryImporter_Import_The_value___0___on_line__1__is_not_a_valid_end_time_, linesRead);
// Error if only one of startTime and endTime is null
if (startTime == null && endTime != null)
{
if (changePeaks)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Missing_start_time_on_line__0_, linesRead));
}
endTime = null;
}
if (startTime != null && endTime == null)
{
if (changePeaks)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Missing_end_time_on_line__0_, linesRead));
}
startTime = null;
}
// Add filename to second dictionary if not yet encountered
ChromSetFileMatch fileMatch;
if (!fileNameToFileMatch.TryGetValue(fileName, out fileMatch))
{
fileMatch = Document.Settings.MeasuredResults.FindMatchingMSDataFile(MsDataFileUri.Parse(fileName));
fileNameToFileMatch.Add(fileName, fileMatch);
}
if (fileMatch == null)
{
UnrecognizedFiles.Add(fileName);
continue;
}
var chromSet = fileMatch.Chromatograms;
string nameSet = chromSet.Name;
ChromFileInfoId[] fileIds;
if (sampleName == null)
{
fileIds = chromSet.MSDataFileInfos.Select(x => x.FileId).ToArray();
}
else
{
var sampleFile = chromSet.MSDataFileInfos.FirstOrDefault(info => Equals(sampleName, info.FilePath.GetSampleName()));
if (sampleFile == null)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Sample__0__on_line__1__does_not_match_the_file__2__, sampleName, linesRead, fileName));
}
fileIds = new[] { sampleFile.FileId };
}
// Define the annotations to be added
var annotations = dataFields.GetAnnotations();
if (!changePeaks)
{
if (apexTime.HasValue)
{
annotations.Add(ComparePeakBoundaries.APEX_ANNOTATION, dataFields.GetField(Field.apex_time));
}
if (startTime.HasValue && endTime.HasValue)
{
annotations.Add(ComparePeakBoundaries.START_TIME_ANNOTATION, dataFields.GetField(Field.start_time));
annotations.Add(ComparePeakBoundaries.END_TIME_ANNOTATION, dataFields.GetField(Field.end_time));
}
}
AnnotationsAdded = annotations.Keys.ToList();
// Loop over all the transition groups in that peptide to find matching charge,
// or use all transition groups if charge not specified
bool foundSample = false;
foreach (var pepPath in pepPaths)
{
var nodePep = (PeptideDocNode)docNew.FindNode(pepPath);
foreach (TransitionGroupDocNode groupNode in nodePep.Children)
{
if (chargeSpecified && charge != groupNode.TransitionGroup.PrecursorAdduct)
{
continue;
}
// Loop over the files in this groupNode to find the correct sample
// Change peak boundaries for the transition group
foreach (var fileId in GetApplicableFiles(fileIds, groupNode))
{
var groupPath = new IdentityPath(pepPath, groupNode.Id);
// Attach annotations
if (annotations.Any())
{
docNew = docNew.AddPrecursorResultsAnnotations(groupPath, fileId, annotations);
}
// Change peak
var filePath = chromSet.GetFileInfo(fileId).FilePath;
if (changePeaks)
{
docNew = docNew.ChangePeak(groupPath, nameSet, filePath,
null, startTime, endTime, UserSet.IMPORTED, null, false);
}
// For removing peaks that are not in the file, if removeMissing = true
trackAdjustedResults.Add(new ResultsKey(fileId.GlobalIndex, groupNode.Id));
foundSample = true;
}
}
}
if (!foundSample)
{
UnrecognizedChargeStates.Add(new UnrecognizedChargeState(charge, fileName, modifiedPeptideString));
}
}
// Remove peaks from the document that weren't in the file.
if (removeMissing)
{
docNew = RemoveMissing(docNew, trackAdjustedResults, changePeaks);
}
// If nothing has changed, return the old Document before ChangeIgnoreChangingChildren was turned off
if (!ReferenceEquals(docNew, docReference))
{
Document = (SrmDocument)Document.ChangeIgnoreChangingChildren(false).ChangeChildrenChecked(docNew.Children);
}
return(Document);
}
public SrmDocument Import(TextReader reader, IProgressMonitor progressMonitor, long lineCount, bool isMinutes, bool removeMissing = false, bool changePeaks = true)
{
var status = new ProgressStatus(Resources.PeakBoundaryImporter_Import_Importing_Peak_Boundaries);
double timeConversionFactor = isMinutes ? 1.0 : 60.0;
int linesRead = 0;
int progressPercent = 0;
var docNew = (SrmDocument) Document.ChangeIgnoreChangingChildren(true);
var docReference = docNew;
var sequenceToNode = new Dictionary<Tuple<string, bool>, IList<IdentityPath>>();
var fileNameToFileMatch = new Dictionary<string, ChromSetFileMatch>();
var trackAdjustedResults = new HashSet<ResultsKey>();
var modMatcher = new ModificationMatcher();
// Make the dictionary of modified peptide strings to doc nodes and paths
for (int i = 0; i < Document.MoleculeCount; ++i)
{
IdentityPath peptidePath = Document.GetPathTo((int) SrmDocument.Level.Molecules, i);
PeptideDocNode peptideNode = (PeptideDocNode) Document.FindNode(peptidePath);
var peptidePair = new Tuple<string, bool>(peptideNode.RawTextId, peptideNode.IsDecoy);
IList<IdentityPath> idPathList;
// Each (sequence, isDecoy) pair can be associated with more than one peptide,
// to handle the case of duplicate peptides in the doucment.
if (sequenceToNode.TryGetValue(peptidePair, out idPathList))
{
idPathList.Add(peptidePath);
sequenceToNode[peptidePair] = idPathList;
}
else
{
idPathList = new List<IdentityPath> { peptidePath };
sequenceToNode.Add(peptidePair, idPathList);
}
}
// Add annotations as possible columns
var allFieldNames = new List<string[]>(FIELD_NAMES);
allFieldNames.AddRange(from def in Document.Settings.DataSettings.AnnotationDefs
where def.AnnotationTargets.Contains(AnnotationDef.AnnotationTarget.precursor_result)
select new[] { def.Name });
string line = reader.ReadLine();
linesRead++;
int[] fieldIndices;
int fieldsTotal;
// If we aren't changing peaks, allow start and end time to be missing
var requiredFields = changePeaks ? REQUIRED_FIELDS : REQUIRED_NO_CHROM;
char correctSeparator = ReadFirstLine(line, allFieldNames, requiredFields, out fieldIndices, out fieldsTotal);
while ((line = reader.ReadLine()) != null)
{
linesRead++;
if (progressMonitor != null)
{
if (progressMonitor.IsCanceled)
return Document;
int progressNew = (int) (linesRead*100/lineCount);
if (progressPercent != progressNew)
{
progressMonitor.UpdateProgress(status = status.ChangePercentComplete(progressNew));
progressPercent = progressNew;
}
}
var dataFields = new DataFields(fieldIndices, line.ParseDsvFields(correctSeparator), allFieldNames);
if (dataFields.Length != fieldsTotal)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Line__0__field_count__1__differs_from_the_first_line__which_has__2_,
linesRead, dataFields.Length, fieldsTotal));
}
string modifiedPeptideString = dataFields.GetField(Field.modified_peptide);
modMatcher.CreateMatches(Document.Settings,
new List<string> {modifiedPeptideString},
Settings.Default.StaticModList,
Settings.Default.HeavyModList);
// Convert the modified peptide string into a standardized form that
// converts unimod, names, etc, into masses, eg [+57.0]
var nodeForModPep = modMatcher.GetModifiedNode(modifiedPeptideString);
if (nodeForModPep == null)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Peptide_has_unrecognized_modifications__0__at_line__1_, modifiedPeptideString, linesRead));
}
nodeForModPep = nodeForModPep.ChangeSettings(Document.Settings, SrmSettingsDiff.ALL);
modifiedPeptideString = nodeForModPep.RawTextId; // Modified sequence, or custom ion name
string fileName = dataFields.GetField(Field.filename);
bool isDecoy = dataFields.IsDecoy(linesRead);
var peptideIdentifier = new Tuple<string, bool>(modifiedPeptideString, isDecoy);
int charge;
bool chargeSpecified = dataFields.TryGetCharge(linesRead, out charge);
string sampleName = dataFields.GetField(Field.sample_name);
double? startTime = null;
double? endTime = null;
if (changePeaks)
{
startTime = dataFields.GetTime(Field.start_time, Resources.PeakBoundaryImporter_Import_The_value___0___on_line__1__is_not_a_valid_start_time_, linesRead);
if (startTime.HasValue)
startTime = startTime / timeConversionFactor;
endTime = dataFields.GetTime(Field.end_time, Resources.PeakBoundaryImporter_Import_The_value___0___on_line__1__is_not_a_valid_end_time_, linesRead);
if (endTime.HasValue)
endTime = endTime / timeConversionFactor;
}
// Error if only one of startTime and endTime is null
if (startTime == null && endTime != null)
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Missing_start_time_on_line__0_, linesRead));
if (startTime != null && endTime == null)
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Missing_end_time_on_line__0_, linesRead));
// Add filename to second dictionary if not yet encountered
ChromSetFileMatch fileMatch;
if (!fileNameToFileMatch.TryGetValue(fileName, out fileMatch))
{
fileMatch = Document.Settings.MeasuredResults.FindMatchingMSDataFile(MsDataFileUri.Parse(fileName));
fileNameToFileMatch.Add(fileName, fileMatch);
}
if (fileMatch == null)
{
UnrecognizedFiles.Add(fileName);
continue;
}
var chromSet = fileMatch.Chromatograms;
string nameSet = chromSet.Name;
ChromFileInfoId[] fileIds;
if (sampleName == null)
{
fileIds = chromSet.MSDataFileInfos.Select(x => x.FileId).ToArray();
}
else
{
var sampleFile = chromSet.MSDataFileInfos.FirstOrDefault(info => Equals(sampleName, info.FilePath.GetSampleName()));
if (sampleFile == null)
{
throw new IOException(string.Format(Resources.PeakBoundaryImporter_Import_Sample__0__on_line__1__does_not_match_the_file__2__, sampleName, linesRead, fileName));
}
fileIds = new[] {sampleFile.FileId};
}
// Look up the IdentityPath of peptide in first dictionary
IList<IdentityPath> pepPaths;
if (!sequenceToNode.TryGetValue(peptideIdentifier, out pepPaths))
{
UnrecognizedPeptides.Add(modifiedPeptideString);
continue;
}
// Define the annotations to be added
var annotations = dataFields.GetAnnotations();
AnnotationsAdded = annotations.Keys.ToList();
// Loop over all the transition groups in that peptide to find matching charge,
// or use all transition groups if charge not specified
bool foundSample = false;
foreach (var pepPath in pepPaths)
{
var nodePep = (PeptideDocNode)docNew.FindNode(pepPath);
for(int i = 0; i < nodePep.Children.Count; ++i)
{
var groupRelPath = nodePep.GetPathTo(i);
var groupNode = (TransitionGroupDocNode) nodePep.FindNode(groupRelPath);
if (!chargeSpecified || charge == groupNode.TransitionGroup.PrecursorCharge)
{
var groupFileIndices =
new HashSet<int>(groupNode.ChromInfos.Select(x => x.FileId.GlobalIndex));
// Loop over the files in this groupNode to find the correct sample
// Change peak boundaries for the transition group
foreach (var fileId in fileIds)
{
if (groupFileIndices.Contains(fileId.GlobalIndex))
{
var groupPath = new IdentityPath(pepPath, groupNode.Id);
// Attach annotations
docNew = docNew.AddPrecursorResultsAnnotations(groupPath, fileId, annotations);
// Change peak
var filePath = chromSet.GetFileInfo(fileId).FilePath;
if (changePeaks)
{
docNew = docNew.ChangePeak(groupPath, nameSet, filePath,
null, startTime, endTime, UserSet.IMPORTED, null, false);
}
// For removing peaks that are not in the file, if removeMissing = true
trackAdjustedResults.Add(new ResultsKey(fileId.GlobalIndex, groupNode.Id));
foundSample = true;
}
}
}
}
}
if (!foundSample)
{
UnrecognizedChargeStates.Add(new UnrecognizedChargeState(charge, fileName, modifiedPeptideString));
}
}
// Remove peaks from the document that weren't in the file.
if (removeMissing)
docNew = RemoveMissing(docNew, trackAdjustedResults, changePeaks);
// If nothing has changed, return the old Document before ChangeIgnoreChangingChildren was turned off
if (!ReferenceEquals(docNew, docReference))
Document = (SrmDocument) Document.ChangeIgnoreChangingChildren(false).ChangeChildrenChecked(docNew.Children);
return Document;
}