private void LoadRowMetadata(Match headerline)
{
if (headerline.Groups["Statusin"].Value.Equals("in"))
{
SequenceMetadata metadata = new SequenceMetadata();
metadata.AlignmentRowName = headerline.Groups["Rowname"].Value;
_alignmentHeaderIndex.Add(metadata.AlignmentRowName, metadata);
}
}
private void LoadSequenceMetadata()
{
//Associate available metadata with Sequence objects.
var seqmetadata = from e in _manifestRoot.Descendants(SequenceMetadata.SequenceMetadataLabel)
join s in _alignment.Sequences
on e.Element(SequenceMetadata.AlignmentRowNameLabel).Value equals s.ID
select new { e, s };
//Load the metadata
foreach (var sm in seqmetadata)
{
sm.s.Metadata.Add(SequenceMetadata.SequenceMetadataLabel, new SequenceMetadata());
SequenceMetadata metadata = (SequenceMetadata)sm.s.Metadata[SequenceMetadata.SequenceMetadataLabel];
if (sm.e.Element(SequenceMetadata.ScientificNameLabel) != null)
{
metadata.ScientificName = sm.e.Element(SequenceMetadata.ScientificNameLabel).Value;
}
if (sm.e.Element(SequenceMetadata.TaxIDLabel) != null)
{
int taxID = 0;
Int32.TryParse(sm.e.Element(SequenceMetadata.TaxIDLabel).Value, out taxID);
metadata.TaxID = taxID;
}
if (sm.e.Element(SequenceMetadata.LineageLabel) != null)
{
metadata.Lineage = sm.e.Element(SequenceMetadata.LineageLabel).Value;
}
if (sm.e.Element(SequenceMetadata.SequenceLengthLabel) != null)
{
int seqLength = 0;
Int32.TryParse(sm.e.Element(SequenceMetadata.SequenceLengthLabel).Value, out seqLength);
metadata.SequenceLength = seqLength;
}
if (sm.e.Element(SequenceMetadata.LocationDescriptionLabel) != null)
{
metadata.LocationDescription = sm.e.Element(SequenceMetadata.LocationDescriptionLabel).Value;
}
if (sm.e.Element(SequenceMetadata.AlignmentRowNameLabel) != null)
{
metadata.AlignmentRowName = sm.e.Element(SequenceMetadata.AlignmentRowNameLabel).Value;
}
if (sm.e.Element(SequenceMetadata.AccessionsLabel) != null)
{
var accessions = from accession in sm.e.Element(SequenceMetadata.AccessionsLabel).Descendants(SequenceMetadata.GenbankAccessionLabel)
select new GenBankVersion
{
Accession = accession.Element(SequenceMetadata.GenbankAccessionIDLabel).Value,
Version = accession.Element(SequenceMetadata.GenbankAccessionVersionLabel).Value
};
foreach (var accession in accessions)
{
metadata.Accessions.Add(accession);
}
}
if (sm.e.Element(SequenceMetadata.StructureModelLabel) != null)
{
StructureModel strmodel = new StructureModel(metadata.SequenceLength);
var pairs = from pair in sm.e.Element(SequenceMetadata.StructureModelLabel).Descendants(SequenceMetadata.StructureModelPairLabel)
select pair;
foreach (var pair in pairs)
{
int fivePrimeIndex, threePrimeIndex;
if (Int32.TryParse(pair.Element(SequenceMetadata.StructureModelPairFivePrimeIndexLabel).Value, out fivePrimeIndex) &&
Int32.TryParse(pair.Element(SequenceMetadata.StructureModelPairThreePrimeIndexLabel).Value, out threePrimeIndex))
{
strmodel.Pairs.Add(fivePrimeIndex, threePrimeIndex);
}
}
strmodel.DecomposeStructure();
metadata.StructureModel = strmodel;
}
}
}
private XElement WriteSequenceMetadata(ISequence sequence)
{
XElement output = new XElement(SequenceMetadata.SequenceMetadataLabel);
if (sequence.Metadata.ContainsKey(SequenceMetadata.SequenceMetadataLabel))
{
SequenceMetadata seqMetadata = (SequenceMetadata)sequence.Metadata[SequenceMetadata.SequenceMetadataLabel];
if (seqMetadata.ScientificName != null)
{
output.Add(new XElement(SequenceMetadata.ScientificNameLabel, seqMetadata.ScientificName));
}
output.Add(new XElement(SequenceMetadata.SequenceLengthLabel, seqMetadata.SequenceLength));
output.Add(new XElement(SequenceMetadata.LineageLabel, seqMetadata.Lineage));
if (seqMetadata.AlignmentRowName != null)
{
output.Add(new XElement(SequenceMetadata.AlignmentRowNameLabel, seqMetadata.AlignmentRowName));
}
if (seqMetadata.LocationDescription != null)
{
output.Add(new XElement(SequenceMetadata.LocationDescriptionLabel, seqMetadata.LocationDescription));
}
if (seqMetadata.Accessions.Count() > 0)
{
XElement accession = new XElement(SequenceMetadata.AccessionsLabel);
foreach (var acc in seqMetadata.Accessions)
{
XElement gbacc = new XElement(SequenceMetadata.GenbankAccessionLabel);
gbacc.Add(new XElement(SequenceMetadata.GenbankAccessionIDLabel, acc.Accession));
gbacc.Add(new XElement(SequenceMetadata.GenbankAccessionVersionLabel, acc.Version));
accession.Add(gbacc);
}
output.Add(accession);
}
if (seqMetadata.StructureModel != null)
{
var model = from pair in seqMetadata.StructureModel.Pairs
select new XElement(SequenceMetadata.StructureModelPairLabel,
new XElement(SequenceMetadata.StructureModelPairFivePrimeIndexLabel, pair.Key),
new XElement(SequenceMetadata.StructureModelPairThreePrimeIndexLabel, pair.Value));
XElement strmodelxml = new XElement(SequenceMetadata.StructureModelLabel);
strmodelxml.Add(model.ToArray());
output.Add(strmodelxml);
}
/*if (seqMetadata.StructureModels.Count() > 0)
* {
* var models = from strmodel in seqMetadata.StructureModels
* select new XElement(SequenceMetadata.StructureModelLabel,
* from pair in strmodel.Pairs
* select new XElement(SequenceMetadata.StructureModelPairLabel,
* new XElement(SequenceMetadata.StructureModelPairFivePrimeIndexLabel, pair.Key),
* new XElement(SequenceMetadata.StructureModelPairThreePrimeIndexLabel, pair.Value)));
*
* XElement strmodels = new XElement(SequenceMetadata.StructureModelsLabel);
* strmodels.Add(models.ToArray());
* output.Add(strmodels);
* }*/
return(output);
}
else
{
return(null);
}
}
private void LoadSequenceData(Match seqdataline)
{
if (_alignmentHeaderIndex.ContainsKey(seqdataline.Groups["Rowname"].Value))
{
SequenceMetadata metadata = _alignmentHeaderIndex[seqdataline.Groups["Rowname"].Value];
ISequence sequence;
int startIndex = Int32.Parse(seqdataline.Groups["Startindex"].Value);
if (startIndex == 0)
{
sequence = new Sequence(RnaAlphabet.Instance)
{
ID = metadata.AlignmentRowName,
DisplayID = metadata.AlignmentRowName
};
sequence.Metadata.Add(SequenceMetadata.SequenceMetadataLabel, metadata);
_alignmentSequenceIndex.Add(metadata.AlignmentRowName, sequence);
_sequences.Add(sequence);
}
else
{
sequence = _alignmentSequenceIndex[seqdataline.Groups["Rowname"].Value];
}
if (sequence.Count < startIndex)
{
//Becuase AE2 format has blanks, we might have to pad
for (int i = 0; i < (startIndex - sequence.Count); i++)
{
sequence.Add(RnaAlphabet.Instance.Gap);
}
}
int lineidx = 0;
ISequenceItem nextElement;
char[] octalholder = new char[3];
string seqdata = seqdataline.Groups["Seqdata"].Value;
while (lineidx < seqdata.Length)
{
if (seqdata[lineidx] == '\\')
{
octalholder[0] = seqdata[lineidx + 1];
octalholder[1] = seqdata[lineidx + 2];
octalholder[2] = seqdata[lineidx + 3];
nextElement = sequence.Alphabet.LookupBySymbol(ConvertOctal(octalholder));
lineidx = lineidx + 4;
}
else
{
nextElement = sequence.Alphabet.LookupBySymbol(seqdata[lineidx]);
lineidx = lineidx + 1;
}
if (nextElement == null)
{
sequence.Add(RnaAlphabet.Instance.Gap);
}
else
{
if (!nextElement.IsGap)
{
metadata.SequenceLength++;
}
sequence.Add(nextElement);
}
}
}
}
