public HaploTypeReport(List <HaplotypeComparison> Comparisons, SimpleSample sample)
{
Sample = sample;
Comparisons.Sort((x, y) => - x.Rank.CompareTo(y.Rank));
BestHit = Comparisons [0];
NumberOfEquallyGoodBestHits = Comparisons.TakeWhile(x => x.Rank == BestHit.Rank).Count();
SecondBestHit = Comparisons [1];
}
public HaploTypeReport GetHaplotypeReport(Sequence toBuildReportFor, List <string> dataLines, string id = "Sample")
{
var delts = ReferenceGenome.GetDeltaAlignments(toBuildReportFor).SelectMany(x => x).ToList();
if (delts.Count != 1)
{
return(null);
throw new Exception("Final assembly had no or multiple delta alignments with the reference, whereas only 1 is expected");
}
var delt = delts [0];
var aln = delt.ConvertDeltaToSequences();
dataLines.Add("REF_OFFSET =" + delt.FirstSequenceStart.ToString());
dataLines.Add("ASSEMBLY_OFFSET =" + delt.SecondSequenceStart.ToString());
//dataLines.Add("ALN_SCORE=" + aln.Score.ToString());
var refseq = aln.FirstSequence as Sequence;
dataLines.Add("REF_SEQUENCE = " + refseq.ConvertToString());
var qseq = aln.SecondSequence as Sequence;
dataLines.Add("QUERY_SEQUENCE= " + qseq.ConvertToString());
//now get all polymorphisms, and sort haplotypes
//TODO: Don't call nucmer twice
var AllPolys = GenomeToHaploGrepConverter.FindPolymorphisms(toBuildReportFor);
PolymorphismFilter pf = new PolymorphismFilter();
var sample = new SimpleSample(id, AllPolys, pf);
var comparisons = treeRoot.GetAllChildren().Select(x => new HaplotypeComparison(x, sample)).ToList();
var report = new HaploTypeReport(comparisons, sample);
dataLines.AddRange(report.GetRowReportLines());
var passedFilters = sample.Polymorphisms.ToDictionary(x => x.position);
//noq get list of differences
dataLines.Add("#Data report below, (Note in rCRS positions, accounting for 'N' at position 3107 and in 1 based index)");
List <VariantAnnotation> reporters = new List <VariantAnnotation>()
{
new HaploTypeMatchAnnotation(report.BestHit, passedFilters),
new FrequencyFileAnnotation(), new BigTableAnnotation()
};
var headerLine = String.Join(VariantAnnotation.FIELD_DELIM, reporters.Select(x => x.GetHeaderLine()).ToArray());
dataLines.Add(headerLine);
var infoLines = AllPolys.Select(x => String.Join(VariantAnnotation.FIELD_DELIM, reporters.Select(y => y.GetAnnotation(x)).ToArray()));
dataLines.AddRange(infoLines);
return(report);
}
public HaploTypeReport GetHaplotypeReport(SimpleSample sample)
{
//verify filtering is the same
foreach (var filt in filtersApplied)
{
if (sample.Polymorphisms.Count() != filt.FilterPolys(sample.Polymorphisms).Count())
{
throw new HaploGrepException("It appears a different filter was used when constructing the tree as opposed to " +
"constructing the samples");
}
}
var comparisons = treeRoot.GetAllChildren().Select(x => new HaplotypeComparison(x, sample)).ToList();
var report = new HaploTypeReport(comparisons, sample);
return(report);
}
/// <summary>
/// Create a new class that compares a sample to a known haplotype
/// </summary>
public HaplotypeComparison(PhyloTreeNodev2 node, SimpleSample sample)
{
this.node = node;
var polysInNode = sample.Filter.FilterPolys(node.Mutations).ToList();
NumberOfPolymorhpismsInHaplotype = (ushort)polysInNode.Count;
haplotypeWeightPoly = polysInNode.Sum(x => x.getMutationRate());
MatchingPolymorphisms = polysInNode.Where(z => sample.Polymorphisms.Contains(z)).ToList();
NumberOfPolymorphismsMissingFromGenotype = (ushort)(polysInNode.Count - MatchingPolymorphisms.Count);
NumberOfPolymorphismsMissingFromHaplotype = (ushort)(sample.Polymorphisms.Count - MatchingPolymorphisms.Count);
matchingWeightPoly = MatchingPolymorphisms.Sum(x => x.getMutationRate());
var CorrectInHaplotypeRatio = haplotypeWeightPoly == 0 ? 1.0 : (matchingWeightPoly / haplotypeWeightPoly);
var CorrectInSampleRatio = sample.TotalSampleWeight == 0 ? 1.0 : (matchingWeightPoly / sample.TotalSampleWeight);
Rank = .5 * CorrectInHaplotypeRatio + .5 * CorrectInSampleRatio;
}