private static Ab1Metadata GetAb1Metadata(IParserContext context)
{
var metadata = new Ab1Metadata();
LoadPeakLocations(metadata, context);
LoadSequenceConfidence(metadata, context);
LoadColorWheelData(metadata, context);
return metadata;
}
private static void LoadPeakLocations(Ab1Metadata metadata, IParserContext context)
{
metadata.PeakLocations =
((ShortDataItem)context
.DataItems
.First(item =>
item.Entry.TagName == Constants.PeakLocationTagName)).Value;
}
private static void LoadPeakLocations(Ab1Metadata metadata, IParserContext context)
{
metadata.PeakLocations =
((ShortDataItem)context
.DataItems
.First(item =>
item.Entry.TagName == Constants.PeakLocationTagName)).Value;
}
/// <summary>
/// Converts the parsed ab1 raw data file into a sequence, with the associated ab1 metadata setup. I use
/// this method because only a subset of the data is needed so there is no point in persisting a large amount
/// of unused metadata.
/// </summary>
/// <param name="context"></param>
/// <returns></returns>
public static ISequence Convert(IParserContext context)
{
ISequence sequence = GetSequence(context);
Ab1Metadata metadata = GetAb1Metadata(context);
Ab1Metadata.SetAb1Data(sequence, metadata);
return(sequence);
}
private static Ab1Metadata GetAb1Metadata(IParserContext context)
{
var metadata = new Ab1Metadata();
LoadPeakLocations(metadata, context);
LoadSequenceConfidence(metadata, context);
LoadColorWheelData(metadata, context);
return(metadata);
}
private static void LoadSequenceConfidence(Ab1Metadata metadata, IParserContext context)
{
metadata.ConfidenceData =
(from c in ((CharDataItem)context
.DataItems
.First(item =>
item.Entry.TagName == Constants.SequenceConfidanceTagName)).Value
select(short) c).ToArray();
}
private static void LoadColorWheelData(Ab1Metadata metadata, IParserContext context)
{
List<KeyValuePair<byte, int>> nucleotideIndices = GetNucleotideDataIndex(context);
IEnumerable<ShortDataItem> dataItems = context.DataItems.OfType<ShortDataItem>();
nucleotideIndices.ForEach(
pair =>
{
ShortDataItem item =
dataItems.First(i => i.Entry.TagNumber == pair.Value && i.Entry.TagName == Constants.DataTagName);
metadata.SetColorData(
pair.Key,
new Ab1ColorData(metadata.PeakLocations, item.Value));
});
}
/// <summary>
/// Sets ab1 metadata on a sequence.
/// </summary>
/// <param name="sequence"></param>
/// <param name="metadata"></param>
public static void SetAb1Data(ISequence sequence, Ab1Metadata metadata)
{
if (sequence == null)
{
throw new ArgumentNullException("sequence");
}
if (sequence.Metadata.ContainsKey(MetadataKey))
{
sequence.Metadata[MetadataKey] = metadata;
}
else
{
sequence.Metadata.Add(MetadataKey, metadata);
}
}
private static void LoadColorWheelData(Ab1Metadata metadata, IParserContext context)
{
List <KeyValuePair <byte, int> > nucleotideIndices = GetNucleotideDataIndex(context);
IEnumerable <ShortDataItem> dataItems = context.DataItems.OfType <ShortDataItem>();
nucleotideIndices.ForEach(
pair =>
{
ShortDataItem item =
dataItems.First(i => i.Entry.TagNumber == pair.Value && i.Entry.TagName == Constants.DataTagName);
metadata.SetColorData(
pair.Key,
new Ab1ColorData(metadata.PeakLocations, item.Value));
});
}
public void TestColorDataTrim()
{
const int dataCount = 40;
// 10 points
var confidence = new short[] {22, 31, 50, 1, 61, 22, 10, 19, 23, 10};
var peakLocations = new short[] {2, 5, 9, 12, 15, 22, 26, 32, 35, 38};
short[] aData = GetTestData(dataCount).ToArray();
short[] cData = GetTestData(dataCount).ToArray();
short[] tData = GetTestData(dataCount).ToArray();
short[] gData = GetTestData(dataCount).ToArray();
var data = new Ab1Metadata {ConfidenceData = confidence.ToArray(), PeakLocations = peakLocations.ToArray()};
data.SetColorData(Alphabets.DNA.A, new Ab1ColorData(peakLocations, aData));
data.SetColorData(Alphabets.DNA.T, new Ab1ColorData(peakLocations, tData));
data.SetColorData(Alphabets.DNA.C, new Ab1ColorData(peakLocations, cData));
data.SetColorData(Alphabets.DNA.G, new Ab1ColorData(peakLocations, gData));
const int segmentStartIndex = 3;
const int segmentLength = 5;
data.Trim(segmentStartIndex, segmentLength);
var newPeakLocations = new List<short> {1, 4, 11, 15, 21};
Assert.AreEqual(segmentLength, data.ConfidenceData.Length);
Assert.AreEqual(segmentLength, data.PeakLocations.Length);
AssertAreEqual(confidence.GetRange(segmentStartIndex, segmentLength), data.ConfidenceData.ToList());
AssertAreEqual(newPeakLocations, data.PeakLocations.ToList());
data.AllColorData()
.ToList().ForEach(colorData =>
{
Assert.AreEqual(segmentLength, colorData.DataByResidue.Count);
int offset = 0;
for (int i = 0; i < colorData.DataByResidue.Count; i++)
{
Assert.AreEqual(newPeakLocations[i],
colorData.DataByResidue[i].PeakIndex + offset);
offset += colorData.DataByResidue[i].Data.Length;
}
});
}
private static void LoadSequenceConfidence(Ab1Metadata metadata, IParserContext context)
{
metadata.ConfidenceData =
((CharDataItem)context.DataItems.First(item => item.Entry.TagName == Constants.SequenceConfidanceTagName)).Value.Select(c => (short)c).ToArray();
}
/// <summary>
/// Sets ab1 metadata on a sequence.
/// </summary>
/// <param name="sequence"></param>
/// <param name="metadata"></param>
public static void SetAb1Data(ISequence sequence, Ab1Metadata metadata)
{
if (sequence == null) throw new ArgumentNullException("sequence");
if (sequence.Metadata.ContainsKey(MetadataKey))
sequence.Metadata[MetadataKey] = metadata;
else
sequence.Metadata.Add(MetadataKey, metadata);
}
