/// <summary>
/// Adds a single genotype to this context.
///
/// There are many constraints on this input, and important
/// impacts on the performance of other functions provided by this
/// context.
///
/// First, the sample name of genotype must be unique within this
/// context. However, this is not enforced in the code itself, through
/// you will invalid the contract on this context if you add duplicate
/// samples and are running with CoFoJa enabled.
///
/// Second, adding genotype also updates the sample name -> index map,
/// so add() followed by containsSample and related function is an efficient
/// series of operations.
///
/// Third, adding the genotype invalidates the sorted list of sample names, to
/// add() followed by any of the SampleNamesInOrder operations is inefficient, as
/// each SampleNamesInOrder must rebuild the sorted list of sample names at
/// an O(n log n) cost.
/// </summary>
/// <param name="genotype">
/// @return </param>
public void Add(Genotype genotype)
{
checkImmutability();
invalidateSampleOrdering();
if (sampleNameToOffset != null)
{
// update the name map by adding entries
sampleNameToOffset[genotype.SampleName] = Count;
}
Genotypes.Add(genotype);
}
public SequenceVariantDescription(string description)
{
Description = description;
if (description == null)
{
return;
}
// Parse description into
string[] vcfFields = description.Split(new[] { @"\t" }, StringSplitOptions.None);
if (vcfFields.Length < 10)
{
return;
}
ReferenceAlleleString = vcfFields[3];
AlternateAlleleString = vcfFields[4];
Info = new SnpEffAnnotation(vcfFields[7]);
AlleleIndex = Info.Allele == null ? -1 : AlternateAlleleString.Split(',').ToList().IndexOf(Info.Allele) + 1; // reference is zero
Format = vcfFields[8];
string[] genotypes = Enumerable.Range(9, vcfFields.Length - 9).Select(i => vcfFields[i]).ToArray();
// loop through genotypes for this variant (e.g. tumor and normal)
for (int individual = 0; individual < genotypes.Length; individual++)
{
var genotypeFields = GenotypeDictionary(Format.Trim(), genotypes[individual].Trim());
// parse genotype
string[] gt = null;
if (genotypeFields.TryGetValue("GT", out string gtString))
{
gt = gtString.Split('/');
}
if (gt == null)
{
continue;
}
// parse allele depth (might be null, technically, but shouldn't be in most use cases)
string[] ad = null;
if (genotypeFields.TryGetValue("AD", out string adString))
{
ad = adString.Split(',');
}
Genotypes.Add(individual.ToString(), gt);
AlleleDepths.Add(individual.ToString(), ad);
Homozygous.Add(individual.ToString(), gt.Distinct().Count() == 1);
Heterozygous.Add(individual.ToString(), gt.Distinct().Count() > 1);
}
}
