public bool Equals(ProteinSequence otherSequence)
{
return(Accession == otherSequence.Accession && GetSequence() == otherSequence.GetSequence());
}
public void ParseFile(string path, ProcessInfo processInfo)
{
processInfo.Status("Parsing " + path);
string accession = "";
int sequenceCounter = 0;
StringBuilder sequence = new StringBuilder();
ProteinSequence protein = new ProteinSequence();
try{
StreamReader file = new StreamReader(path);
string line;
while ((line = file.ReadLine()) != null) // valid line
{
if (sequenceCounter % 500 == 0)
{
processInfo.Status("Parsing " + path + ", " + (int)((float)file.BaseStream.Position / file.BaseStream.Length * 100) +
"%");
}
bool lineIsHeader = line.StartsWith(">");
// skip all lines until the first header is found
if (sequenceCounter == 0 && !lineIsHeader)
{
continue;
}
// line is a piece of a sequence
if (sequenceCounter > 0 && !lineIsHeader)
{
sequence.Append(line.Trim());
continue;
}
// line is a fasta header
if (lineIsHeader)
{
if (sequenceCounter > 0)
// this is not the first header, i.e. the previous sequence is now completely read in
{
// add the previous protein
protein.SetSequence(sequence.ToString());
entries.Add(accession, protein);
}
// initialize a new protein
protein = new ProteinSequence();
sequenceCounter++;
// then parse the new header
string header = line;
Match m = regexUniprotAccession.Match(header);
if (m.Success) // uniprot header
{
accession = m.Groups[1].Value;
protein.Accession = accession;
protein.Header = header;
}
else // fallback position: take entire header after the > as accession
{
accession = header.Substring(1).Trim();
protein.Accession = accession;
protein.Header = header;
}
sequence = new StringBuilder();
}
} //end while
file.Close();
//add the last protein
if (sequenceCounter > 0) // make sure there is at least one sequence in the file
{
protein.SetSequence(sequence.ToString());
entries.Add(accession, protein);
}
} catch (Exception) {
processInfo.ErrString =
"Something went wrong while parsing the fasta file.\nMake sure the path is correct and the " +
"file is not opened in another application.\nMake sure the fasta file is valid.";
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int proteinIdColumnInd = param.GetParam <int>("Protein IDs").Value;
string[][] proteinIds = new string[mdata.RowCount][];
string[][] leadingIds = new string[mdata.RowCount][];
List <string> allIds = new List <string>();
for (int row = 0; row < mdata.RowCount; row++)
{
proteinIds[row] = mdata.StringColumns[proteinIdColumnInd][row].Split(';');
leadingIds[row] = new[] { proteinIds[row][0] };
allIds.AddRange(proteinIds[row]);
}
string fastaFilePath = param.GetParam <string>("Fasta file").Value;
Fasta fasta = new Fasta();
fasta.ParseFile(fastaFilePath, processInfo);
// Text annotations
processInfo.Status("Adding fasta header annotations.");
int[] selection =
param.GetParamWithSubParams <int>("Fasta header annotations").GetSubParameters().GetParam <int[]>("Annotations").Value;
string[][] idsToBeAnnotated = param.GetParamWithSubParams <int>("Fasta header annotations").Value == 0
? proteinIds
: leadingIds;
ProteinSequence[][] fastaEntries = new ProteinSequence[mdata.RowCount][];
for (int row = 0; row < mdata.RowCount; row++)
{
List <ProteinSequence> rowEntries = new List <ProteinSequence>();
foreach (string id in idsToBeAnnotated[row])
{
ProteinSequence entry = fasta.GetEntry(id);
if (entry == null)
{
continue;
}
rowEntries.Add(entry);
}
fastaEntries[row] = rowEntries.ToArray();
}
if (ArrayUtils.Contains(selection, 0)) // Entry name
{
string[] annotationColumn = new string[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <string> rowAnnotations = new List <string>();
foreach (ProteinSequence entry in fastaEntries[row])
{
string entryName = entry.EntryName;
if (entryName != null && !ArrayUtils.Contains(rowAnnotations, entryName))
{
rowAnnotations.Add(entryName);
}
}
annotationColumn[row] = string.Join(";", rowAnnotations.ToArray());
}
mdata.AddStringColumn("Entry name", "", annotationColumn);
}
if (ArrayUtils.Contains(selection, 1)) // Gene name
{
string[] annotationColumn = new string[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <string> rowAnnotations = new List <string>();
foreach (ProteinSequence entry in fastaEntries[row])
{
string geneName = entry.GeneName;
if (geneName != null && !ArrayUtils.Contains(rowAnnotations, geneName))
{
rowAnnotations.Add(geneName);
}
}
annotationColumn[row] = string.Join(";", rowAnnotations.ToArray());
}
mdata.AddStringColumn("Gene name", "", annotationColumn);
}
if (ArrayUtils.Contains(selection, 2))
{
// Verbose protein name, i.e. all protein names annotated in all fasta headers, including the
//'Isoform x of...' prefixes and '(Fragment)' suffixes
string[] annotationColumn = new string[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <string> rowAnnotations = new List <string>();
foreach (ProteinSequence entry in fastaEntries[row])
{
string proteinName = entry.ProteinName;
if (proteinName != null && !ArrayUtils.Contains(rowAnnotations, proteinName))
{
rowAnnotations.Add(proteinName);
}
}
annotationColumn[row] = string.Join(";", rowAnnotations.ToArray());
}
mdata.AddStringColumn("Protein name (verbose)", "", annotationColumn);
}
if (ArrayUtils.Contains(selection, 3)) // Consensus protein name
{
string[] annotationColumn = new string[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <string> rowAnnotations = new List <string>();
foreach (ProteinSequence entry in fastaEntries[row])
{
string proteinName = entry.ConsensusProteinName;
if (proteinName != null && !ArrayUtils.Contains(rowAnnotations, proteinName))
{
rowAnnotations.Add(proteinName);
}
}
annotationColumn[row] = String.Join(";", rowAnnotations.ToArray());
}
mdata.AddStringColumn("Protein name", "", annotationColumn);
}
if (ArrayUtils.Contains(selection, 4)) // Species
{
string[] annotationColumn = new string[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <string> rowAnnotations = new List <string>();
foreach (ProteinSequence entry in fastaEntries[row])
{
string speciesName = entry.Species;
if (speciesName != null && !ArrayUtils.Contains(rowAnnotations, speciesName))
{
rowAnnotations.Add(speciesName);
}
}
annotationColumn[row] = String.Join(";", rowAnnotations.ToArray());
}
mdata.AddStringColumn("Species", "", annotationColumn);
}
// Numeric annotations
processInfo.Status("Adding numeric annotations.");
selection =
param.GetParamWithSubParams <int>("Numeric annotations").GetSubParameters().GetParam <int[]>("Annotations").Value;
bool annotateLeadingId = param.GetParamWithSubParams <int>("Numeric annotations").Value == 1;
if (ArrayUtils.Contains(selection, 0)) // Sequence length
{
double[] annotationColumn = new double[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <double> rowAnnotations = new List <double>();
foreach (ProteinSequence entry in fastaEntries[row])
{
double sequenceLength = entry.GetSequence().Length;
rowAnnotations.Add(sequenceLength);
if (annotateLeadingId && rowAnnotations.Count > 0)
{
break;
}
}
annotationColumn[row] = ArrayUtils.Median(rowAnnotations.ToArray());
}
mdata.AddNumericColumn("Sequence length", "", annotationColumn);
}
if (ArrayUtils.Contains(selection, 1)) // Monoisotopic molecular mass
{
double[] annotationColumn = new double[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <double> rowAnnotations = new List <double>();
foreach (ProteinSequence entry in fastaEntries[row])
{
double monoisotopicMass = entry.GetMonoisotopicMolecularMass();
rowAnnotations.Add(monoisotopicMass);
if (annotateLeadingId && rowAnnotations.Count > 0)
{
break;
}
}
annotationColumn[row] = ArrayUtils.Median(rowAnnotations.ToArray());
}
mdata.AddNumericColumn("Monoisotopic molecular mass", "", annotationColumn);
}
if (ArrayUtils.Contains(selection, 2)) // Average molecular mass
{
double[] annotationColumn = new double[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <double> rowAnnotations = new List <double>();
foreach (ProteinSequence entry in fastaEntries[row])
{
double averageMass = entry.GetAverageMolecularMass();
rowAnnotations.Add(averageMass);
if (annotateLeadingId && rowAnnotations.Count > 0)
{
break;
}
}
annotationColumn[row] = ArrayUtils.Median(rowAnnotations.ToArray());
}
mdata.AddNumericColumn("Average molecular mass", "", annotationColumn);
}
// Theoretical peptides
processInfo.Status("Calculating theoretical peptides.");
annotateLeadingId = param.GetParamWithSubParams <int>("Calculate theoretical peptides").Value == 1;
Protease[] proteases = ArrayUtils.SubArray(Constants.defaultProteases,
param.GetParamWithSubParams <int>("Calculate theoretical peptides").GetSubParameters().GetParam <int[]>("Proteases")
.Value);
double minLength =
param.GetParamWithSubParams <int>("Calculate theoretical peptides").GetSubParameters().GetParam <double>(
"Min. peptide length").Value;
double maxLength =
param.GetParamWithSubParams <int>("Calculate theoretical peptides").GetSubParameters().GetParam <double>(
"Max. peptide length").Value;
bool displayPeptideSequences = annotateLeadingId &&
param.GetParamWithSubParams <int>("Calculate theoretical peptides").GetSubParameters().GetParam <bool>(
"Show sequences").Value;
foreach (Protease protease in proteases)
{
double[] annotationColumn = new double[mdata.RowCount];
string[] peptideColumn = new string[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <double> rowAnnotations = new List <double>();
List <string> rowPeptides = new List <string>();
foreach (ProteinSequence entry in fastaEntries[row])
{
double nTheoreticalPeptides = entry.GetNumberOfTheoreticalPeptides(protease, (int)minLength, (int)maxLength);
rowAnnotations.Add(nTheoreticalPeptides);
if (displayPeptideSequences)
{
rowPeptides.AddRange(entry.GetTheoreticalPeptideSequences(protease, (int)minLength, (int)maxLength));
}
if (annotateLeadingId && rowAnnotations.Count > 0)
{
break;
}
}
annotationColumn[row] = ArrayUtils.Median(rowAnnotations.ToArray());
peptideColumn[row] = String.Join(";", rowPeptides);
}
mdata.AddNumericColumn(
"Number of theoretical peptides (" + protease.name + ", " + minLength + "-" + maxLength + ")", "", annotationColumn);
if (displayPeptideSequences)
{
mdata.AddStringColumn(
"Theoretical peptide sequences (" + protease.name + ", " + minLength + "-" + maxLength + ")", "", peptideColumn);
}
}
// Sequence features
processInfo.Status("Counting sequence features.");
annotateLeadingId = param.GetParamWithSubParams <int>("Count sequence features").Value == 1;
bool normalizeBySequenceLength =
param.GetParamWithSubParams <int>("Count sequence features").GetSubParameters().GetParam <bool>(
"Normalize by sequence length").Value;
if (param.GetParamWithSubParams <int>("Count sequence features").GetSubParameters().GetParam <string>("Regex").Value !=
"")
{
Regex regex;
try{
regex =
new Regex(
param.GetParamWithSubParams <int>("Count sequence features").GetSubParameters().GetParam <string>("Regex").Value);
} catch (ArgumentException) {
processInfo.ErrString = "The regular expression you provided has invalid syntax.";
return;
}
double[] sequenceFeatureColumn = new double[mdata.RowCount];
for (int row = 0; row < mdata.RowCount; row++)
{
List <double> featureCount = new List <double>();
foreach (ProteinSequence entry in fastaEntries[row])
{
double nFeatures = regex.Matches(entry.GetSequence()).Count;
featureCount.Add(normalizeBySequenceLength ? nFeatures / entry.GetLength() : nFeatures);
if (annotateLeadingId)
{
break;
}
}
sequenceFeatureColumn[row] = ArrayUtils.Median(featureCount.ToArray());
}
mdata.AddNumericColumn(
(normalizeBySequenceLength ? "Normalized feature count (" : "Feature count (") + regex + ")", "",
sequenceFeatureColumn);
}
processInfo.Status("Done.");
}
