public IEnumerable<PeptideGroupDocNode> Import(IProgressMonitor progressMonitor,
ColumnIndices indices,
IDictionary<string, FastaSequence> dictNameSeq,
out List<MeasuredRetentionTime> irtPeptides,
out List<SpectrumMzInfo> librarySpectra,
out List<TransitionImportErrorInfo> errorList)
{
irtPeptides = new List<MeasuredRetentionTime>();
librarySpectra = new List<SpectrumMzInfo>();
errorList = new List<TransitionImportErrorInfo>();
var status = new ProgressStatus(string.Empty);
// Get the lines used to guess the necessary columns and create the row reader
if (progressMonitor != null)
{
if (progressMonitor.IsCanceled)
return new PeptideGroupDocNode[0];
status = status.ChangeMessage(Resources.MassListImporter_Import_Reading_transition_list);
}
var lines = new List<string>(Inputs.ReadLines());
if (progressMonitor != null)
{
if (progressMonitor.IsCanceled)
return new PeptideGroupDocNode[0];
status = status.ChangeMessage(Resources.MassListImporter_Import_Inspecting_peptide_sequence_information);
}
if (indices != null)
{
_rowReader = new GeneralRowReader(FormatProvider, Separator, indices, Settings, lines);
}
else
{
// Check first line for validity
var line = lines.FirstOrDefault();
string[] fields = line.ParseDsvFields(Separator);
string[] headers = fields.All(field => GetColumnType(field.Trim(), FormatProvider) != typeof (double))
? fields
: null;
int decoyColumn = -1;
int irtColumn = -1;
int libraryColumn = -1;
var irtNames = new[] { "tr_recalibrated", "irt" }; // Not L10N
var libraryNames = new[] { "libraryintensity", "relativeintensity", "relative_intensity", "relativefragmentintensity", "library_intensity" }; // Not L10N
var decoyNames = new[] { "decoy" }; // Not L10N
if (headers != null)
{
lines.RemoveAt(0);
decoyColumn = headers.IndexOf(col => decoyNames.Contains(col.ToLowerInvariant()));
irtColumn = headers.IndexOf(col => irtNames.Contains(col.ToLowerInvariant()));
libraryColumn = headers.IndexOf(col => libraryNames.Contains(col.ToLowerInvariant()));
line = lines.FirstOrDefault();
fields = line != null ? line.ParseDsvFields(Separator) : new string[0];
}
if (fields.Length < 3)
throw new InvalidDataException(Resources.MassListImporter_Import_Invalid_transition_list_Transition_lists_must_contain_at_least_precursor_m_z_product_m_z_and_peptide_sequence);
// If no numeric columns in the first row
_rowReader = ExPeptideRowReader.Create(lines, decoyColumn, FormatProvider, Separator, Settings, irtColumn, libraryColumn);
if (_rowReader == null)
{
_rowReader = GeneralRowReader.Create(lines, headers, decoyColumn, FormatProvider, Separator, Settings, irtColumn, libraryColumn);
if (_rowReader == null && headers == null)
{
// Check for a possible header row
headers = lines[0].Split(Separator);
lines.RemoveAt(0);
_rowReader = GeneralRowReader.Create(lines, headers, decoyColumn, FormatProvider, Separator, Settings, irtColumn, libraryColumn);
}
if (_rowReader == null)
throw new LineColNumberedIoException(Resources.MassListImporter_Import_Failed_to_find_peptide_column, 1, -1);
}
}
// Set starting values for limit counters
_countPeptides = Document.PeptideCount;
_countIons = Document.PeptideTransitionCount;
List<PeptideGroupDocNode> peptideGroupsNew = new List<PeptideGroupDocNode>();
PeptideGroupBuilder seqBuilder = null;
// Process lines
long lineIndex = 0;
foreach (string row in lines)
{
lineIndex++;
var errorInfo = _rowReader.NextRow(row, lineIndex);
if (errorInfo != null)
{
errorList.Add(errorInfo);
continue;
}
if (progressMonitor != null)
{
if (progressMonitor.IsCanceled)
{
irtPeptides.Clear();
librarySpectra.Clear();
errorList.Clear();
return new PeptideGroupDocNode[0];
}
int percentComplete = (int)(lineIndex * 100 / lines.Count);
if (status.PercentComplete != percentComplete)
{
string message = string.Format(Resources.MassListImporter_Import_Importing__0__,
_rowReader.TransitionInfo.ProteinName ?? _rowReader.TransitionInfo.PeptideSequence);
status = status.ChangePercentComplete(percentComplete).ChangeMessage(message);
}
}
seqBuilder = AddRow(seqBuilder, _rowReader, dictNameSeq, peptideGroupsNew, lineIndex, irtPeptides, librarySpectra, errorList);
}
// Add last sequence.
if (seqBuilder != null)
AddPeptideGroup(peptideGroupsNew, seqBuilder, irtPeptides, librarySpectra, errorList);
return MergeEqualGroups(peptideGroupsNew);
}
private const string REGEX_PEPTIDE_FORMAT = @"^([^. ]+)\.([A-Z0-9_+\-\[\]]+)\..+\.(light|{0})$"; // Not L10N
#endregion Fields
#region Constructors
private ExPeptideRowReader(IFormatProvider provider,
char separator,
ColumnIndices indices,
Regex exPeptideRegex,
SrmSettings settings,
IEnumerable<string> lines)
: base(provider, separator, indices, settings, GetSequencesFromLines(lines, separator, indices, exPeptideRegex))
{
ExPeptideRegex = exPeptideRegex;
}
protected MassListRowReader(IFormatProvider provider,
char separator,
ColumnIndices indices,
SrmSettings settings,
IEnumerable<string> sequences)
{
FormatProvider = provider;
Separator = separator;
Indices = indices;
Settings = settings;
ModMatcher = CreateModificationMatcher(settings, sequences);
NodeDictionary = new Dictionary<string, PeptideDocNode>();
}
public IEnumerable<PeptideGroupDocNode> Import(IProgressMonitor progressMonitor,
ColumnIndices indices,
IDictionary<string, FastaSequence> dictNameSeq)
{
List<MeasuredRetentionTime> irtPeptides;
List<SpectrumMzInfo> librarySpectra;
List<TransitionImportErrorInfo> errorList;
return Import(progressMonitor, indices, dictNameSeq, out irtPeptides, out librarySpectra, out errorList);
}
private static IEnumerable<string> GetSequencesFromLines(IEnumerable<string> lines, char separator, ColumnIndices indices)
{
return lines.Select(line => RemoveModifiedSequenceNotes(line.ParseDsvFields(separator)[indices.PeptideColumn]));
}
public static GeneralRowReader Create(IList<string> lines, IList<string> headers, int iDecoy,
IFormatProvider provider, char separator, SrmSettings settings, int iirt, int iLibrary)
{
// Split the first line into fields.
Assume.IsTrue(lines.Count > 0);
// Look for sequence column
string[] fieldsFirstRow = null;
PrecursorCandidate[] sequenceCandidates = null;
int bestCandidateIndex = -1;
int iLabelType = -1;
double tolerance = settings.TransitionSettings.Instrument.MzMatchTolerance;
foreach (var line in lines)
{
string[] fields = line.ParseDsvFields(separator);
if (fieldsFirstRow == null)
fieldsFirstRow = fields;
// Choose precursor field candidates from the first row
if (sequenceCandidates == null)
{
iLabelType = FindLabelType(fields, lines, separator);
// If no sequence column found, return null. After this, all errors throw.
var newSeqCandidates = FindSequenceCandidates(fields);
if (newSeqCandidates.Length == 0)
return null;
var listNewCandidates = new List<PrecursorCandidate>();
foreach (var candidateIndex in newSeqCandidates)
{
string sequence = RemoveSequenceNotes(fields[candidateIndex]);
string modifiedSequence = RemoveModifiedSequenceNotes(fields[candidateIndex]);
IsotopeLabelType labelType = IsotopeLabelType.light;
if (iLabelType != -1)
labelType = GetLabelType(fields[iLabelType]);
IList<TransitionExp> transitionExps;
int candidateMzIndex = FindPrecursor(fields, sequence, modifiedSequence, labelType, candidateIndex, iDecoy,
tolerance, provider, settings, out transitionExps);
// If no match, and no specific label type, then try heavy.
if (settings.PeptideSettings.Modifications.HasHeavyModifications &&
candidateMzIndex == -1 && iLabelType == -1)
{
var peptideMods = settings.PeptideSettings.Modifications;
foreach (var typeMods in peptideMods.GetHeavyModifications())
{
if (settings.TryGetPrecursorCalc(typeMods.LabelType, null) != null)
{
candidateMzIndex = FindPrecursor(fields, sequence, modifiedSequence, typeMods.LabelType, candidateIndex, iDecoy,
tolerance, provider, settings, out transitionExps);
if (candidateMzIndex != -1)
break;
}
}
}
if (candidateMzIndex != -1)
listNewCandidates.Add(new PrecursorCandidate(candidateIndex, candidateMzIndex, sequence, transitionExps));
}
if (listNewCandidates.Count == 0)
throw new MzMatchException(Resources.GeneralRowReader_Create_No_valid_precursor_m_z_column_found, 1, -1);
sequenceCandidates = listNewCandidates.ToArray();
}
bestCandidateIndex = FindBestCandidate(sequenceCandidates, fields);
// Break if a best candidate was found
if (bestCandidateIndex != -1)
break;
}
if (sequenceCandidates == null)
return null;
if (bestCandidateIndex == -1)
bestCandidateIndex = 0;
var prec = sequenceCandidates[bestCandidateIndex];
int iSequence = prec.SequenceIndex;
int iPrecursor = prec.PrecursorMzIdex;
int iProduct = FindProduct(fieldsFirstRow, prec.Sequence, prec.TransitionExps, prec.SequenceIndex, prec.PrecursorMzIdex,
tolerance, provider, settings);
if (iProduct == -1)
throw new MzMatchException(Resources.GeneralRowReader_Create_No_valid_product_m_z_column_found, 1, -1);
int iProtein = FindProtein(fieldsFirstRow, iSequence, lines, headers, provider, separator);
var indices = new ColumnIndices(iProtein, iSequence, iPrecursor, iProduct, iLabelType, iDecoy, iirt, iLibrary);
return new GeneralRowReader(provider, separator, indices, settings, lines);
}
private static readonly string[] EXCLUDE_PROTEIN_VALUES = { "true", "false", "heavy", "light", "unit" }; // Not L10N
#endregion Fields
#region Constructors
public GeneralRowReader(IFormatProvider provider,
char separator,
ColumnIndices indices,
SrmSettings settings,
IEnumerable<string> lines)
: base(provider, separator, indices, settings, GetSequencesFromLines(lines, separator, indices))
{
}
private static IEnumerable<string> GetSequencesFromLines(IEnumerable<string> lines, char separator, ColumnIndices indices, Regex exPeptideRegex)
{
return lines.Select(line => GetModifiedSequence(exPeptideRegex.Match(line.ParseDsvFields(separator)[indices.PeptideColumn])));
}
public static ExPeptideRowReader Create(IList<string> lines, int iDecoy,
IFormatProvider provider, char separator, SrmSettings settings, int iirt, int iLibrary)
{
// Split the first line into fields.
Debug.Assert(lines.Count > 0);
string[] fields = lines[0].ParseDsvFields(separator);
// Create the ExPeptide regular expression
var modSettings = settings.PeptideSettings.Modifications;
var heavyTypeNames = from typedMods in modSettings.GetHeavyModifications()
select typedMods.LabelType.Name;
string exPeptideFormat = string.Format(REGEX_PEPTIDE_FORMAT, string.Join("|", heavyTypeNames.ToArray())); // Not L10N
var exPeptideRegex = new Regex(exPeptideFormat);
// Look for sequence column
string sequence;
string modifiedSequence;
IsotopeLabelType labelType;
int iExPeptide = FindExPeptide(fields, exPeptideRegex, settings,
out sequence, out modifiedSequence, out labelType);
// If no sequence column found, return null. After this,
// all errors throw.
if (iExPeptide == -1)
return null;
if (!labelType.IsLight && !modSettings.HasHeavyImplicitModifications)
{
var message = TextUtil.LineSeparate(Resources.ExPeptideRowReader_Create_Isotope_labeled_entry_found_without_matching_settings,
Resources.ExPeptideRowReaderCreateCheck_the_Modifications_tab_in_Transition_Settings);
throw new LineColNumberedIoException(message, 1, iExPeptide);
}
double tolerance = settings.TransitionSettings.Instrument.MzMatchTolerance;
IList<TransitionExp> transitionExps;
int iPrecursor = FindPrecursor(fields, sequence, modifiedSequence, labelType, iExPeptide, iDecoy,
tolerance, provider, settings, out transitionExps);
if (iPrecursor == -1)
throw new MzMatchException(Resources.GeneralRowReader_Create_No_valid_precursor_m_z_column_found, 1, -1);
int iProduct = FindProduct(fields, sequence, transitionExps, iExPeptide, iPrecursor,
tolerance, provider, settings);
if (iProduct == -1)
throw new MzMatchException(Resources.GeneralRowReader_Create_No_valid_product_m_z_column_found, 1, -1);
var indices = new ColumnIndices(iExPeptide, iExPeptide, iPrecursor, iProduct, iExPeptide, iDecoy, iirt, iLibrary);
return new ExPeptideRowReader(provider, separator, indices, exPeptideRegex, settings, lines);
}
