/// <summary>
/// Tells this builder that the resulting VariantContext should use this genotypes GenotypeContext
///
/// Note that genotypes can be null -> meaning there are no genotypes
/// </summary>
/// <param name="genotypes"> </param>
public void SetGenotypes(GenotypesContext genotypes, bool requireValidation = true)
{
this.genotypes_Renamed = genotypes;
if (genotypes != null && requireValidation)
{
toValidate = toValidate | VariantContext.Validation.GENOTYPES;
}
//TODO: Perhaps if the genotypes are reset we should remove the validation flag if it was not required but was already present?
}
public VariantContextBuilder(VariantContextBuilder parent) : this()
{
if (parent == null)
{
throw new System.ArgumentException("BUG: VariantContext parent argument cannot be null in VariantContextBuilder");
}
this.alleles_Renamed = parent.alleles_Renamed;
this.attributesCanBeModified = false;
this.Contig = parent.Contig;
this.genotypes_Renamed = parent.genotypes_Renamed;
this.ID = parent.ID;
this.Log10PError = parent.Log10PError;
this.Source = parent.Source;
this.start_Renamed = parent.start_Renamed;
this.stop_Renamed = parent.stop_Renamed;
this.FullyDecoded = parent.FullyDecoded;
this.Attributes = parent.attributes_Renamed;
this.SetFilters(parent.filters_Renamed);
}
/// <summary>
/// Returns a new builder based on parent -- the new VC will have all fields initialized
/// to their corresponding values in parent. This is the best way to create a derived VariantContext
/// </summary>
/// <param name="parent"> Cannot be null </param>
public VariantContextBuilder(VariantContext parent) : this()
{
if (parent == null)
{
throw new System.ArgumentException("BUG: VariantContextBuilder parent argument cannot be null in VariantContextBuilder");
}
this.alleles_Renamed = parent.alleles;
this.attributes_Renamed = (IDictionary <string, object>)parent.Attributes;
this.attributesCanBeModified = false;
this.Contig = parent.contig;
this.filters_Renamed = (ISet <string>)parent.FiltersMaybeNull;
this.genotypes_Renamed = parent.genotypes;
this.ID = parent.ID;
this.Log10PError = parent.Log10PError;
this.Source = parent.Source;
this.start_Renamed = parent.Start;
this.stop_Renamed = parent.End;
this.FullyDecoded = parent.FullyDecoded;
}
/// <summary>
/// Return a freshly allocated subcontext of this context containing only the samples
/// listed in samples. Note that samples can contain names not in this context, they
/// will just be ignored.
/// </summary>
/// <param name="samples">
/// @return </param>
public GenotypesContext subsetToSamples(ISet <string> samples)
{
int nSamples = samples.Count;
if (nSamples == 0)
{
return(NO_GENOTYPES);
}
else // nGenotypes < nSamples
{
GenotypesContext subset = create(samples.Count);
foreach (String sample in samples)
{
Genotype g = Get(sample);
if (g != null)
{
subset.Add(g);
}
}
return(subset);
}
}
/// <summary>
/// This method subsets down to a set of samples.
///
/// At the same time returns the alleles to just those in use by the samples,
/// if rederiveAllelesFromGenotypes is true, otherwise the full set of alleles
/// in this VC is returned as the set of alleles in the subContext, even if
/// some of those alleles aren't in the samples
///
/// WARNING: BE CAREFUL WITH rederiveAllelesFromGenotypes UNLESS YOU KNOW WHAT YOU ARE DOING
/// </summary>
/// <param name="sampleNames"> the sample names </param>
/// <param name="rederiveAllelesFromGenotypes"> if true, returns the alleles to just those in use by the samples, true should be default </param>
/// <returns> new VariantContext subsetting to just the given samples </returns>
public VariantContext SubContextFromSamples(ISet <string> sampleNames, bool rederiveAllelesFromGenotypes)
{
if (sampleNames.SetEquals(SampleNames) && !rederiveAllelesFromGenotypes)
{
return(this); // fast path when you don't have any work to do
}
else
{
VariantContextBuilder builder = new VariantContextBuilder(this);
GenotypesContext newGenotypes = genotypes.subsetToSamples(sampleNames);
if (rederiveAllelesFromGenotypes)
{
builder.SetAlleles(allelesOfGenotypes(newGenotypes));
}
else
{
builder.SetAlleles(alleles);
}
builder.SetGenotypes(newGenotypes);
return(builder.make());
}
}
/// <summary>
/// Add a record to the file
/// </summary>
/// <param name="vc">The Variant Context object </param>
protected string getVariantLinetoWrite(VariantContext vc)
{
if (doNotWriteGenotypes)
{
vc = (new VariantContextBuilder(vc)).noGenotypes().make();
}
try
{
//Convert alleles to 1,2,3,etc. numbering
IDictionary <Allele, string> alleleMap = buildAlleleMap(vc);
// CHROM
StringBuilder lineToWrite = new StringBuilder();
//Add chr, pos, id, ref
lineToWrite.Append(String.Join(VCFConstants.FIELD_SEPARATOR, vc.Chr, vc.Start.ToString(), vc.ID, vc.Reference.DisplayString));
// ALT
if (vc.Variant)
{
Allele altAllele = vc.GetAlternateAllele(0);
string alt = altAllele.DisplayString;
lineToWrite.Append(alt);
for (int i = 1; i < vc.AlternateAlleles.Count; i++)
{
altAllele = vc.GetAlternateAllele(i);
alt = altAllele.DisplayString;
lineToWrite.Append(",");
lineToWrite.Append(alt);
}
}
else
{
lineToWrite.Append(VCFConstants.EMPTY_ALTERNATE_ALLELE_FIELD);
}
lineToWrite.Append(VCFConstants.FIELD_SEPARATOR);
// QUAL
if (!vc.HasLog10PError)
{
lineToWrite.Append(VCFConstants.MISSING_VALUE_v4);
}
else
{
lineToWrite.Append(formatQualValue(vc.PhredScaledQual));
}
lineToWrite.Append(VCFConstants.FIELD_SEPARATOR);
// FILTER
string filters = getFilterString(vc);
lineToWrite.Append(filters);
lineToWrite.Append(VCFConstants.FIELD_SEPARATOR);
// INFO
IDictionary <string, string> infoFields = new SortedDictionary <string, string>();
foreach (KeyValuePair <string, object> field in vc.Attributes)
{
string key = field.Key;
if (!mHeader.hasInfoLine(key))
{
fieldIsMissingFromHeaderError(vc, key, "INFO");
}
string outputValue = formatVCFField(field.Value);
if (outputValue != null)
{
infoFields[key] = outputValue;
}
}
lineToWrite.Append(getInfoString(infoFields));;
// FORMAT
GenotypesContext gc = vc.Genotypes;
if (gc.LazyWithData && ((LazyGenotypesContext)gc).UnparsedGenotypeData is string)
{
lineToWrite.Append(VCFConstants.FIELD_SEPARATOR);
lineToWrite.Append(((LazyGenotypesContext)gc).UnparsedGenotypeData.ToString());
}
else
{
IList <string> genotypeAttributeKeys = calcVCFGenotypeKeys(vc);
if (genotypeAttributeKeys.Count > 0)
{
foreach (String format in genotypeAttributeKeys)
{
if (!mHeader.hasFormatLine(format))
{
fieldIsMissingFromHeaderError(vc, format, "FORMAT");
}
}
string genotypeFormatString = String.Join(VCFConstants.GENOTYPE_FIELD_SEPARATOR, genotypeAttributeKeys);
lineToWrite.Append(VCFConstants.FIELD_SEPARATOR);
lineToWrite.Append(genotypeFormatString);
lineToWrite.Append(getGenotypeDataText(vc, alleleMap, genotypeAttributeKeys));
}
}
lineToWrite.Append("\n");
return(lineToWrite.ToString());
}
catch (IOException e)
{
throw new Exception("Unable to write the VCF object:\n " + vc.ToString() + "\n", e);
}
}
/// <summary>
/// Tells this builder that the resulting VariantContext should not contain any GenotypeContext
/// </summary>
public virtual VariantContextBuilder noGenotypes()
{
this.genotypes_Renamed = null;
return(this);
}
/// <summary>
/// Tells this builder that the resulting VariantContext should use a GenotypeContext containing genotypes </summary>
/// <param name="genotypes"> </param>
public void SetGenotypes(params Genotype[] genotypes)
{
this.SetGenotypes(GenotypesContext.copy(genotypes));
}
/// <summary>
/// Tells this builder that the resulting VariantContext should use a GenotypeContext containing genotypes
///
/// Note that genotypes can be null -> meaning there are no genotypes
/// </summary>
/// <param name="genotypes"> </param>
public void SetGenotypes(ICollection <Genotype> genotypes)
{
this.SetGenotypes(GenotypesContext.copy(genotypes));
}
/// <summary>
/// Create a freshly allocated GenotypeContext containing the genotypes in toCopy
/// </summary>
/// <param name="toCopy"> the GenotypesContext to copy </param>
/// <returns> an mutable GenotypeContext containing genotypes </returns>
public static GenotypesContext copy(GenotypesContext toCopy)
{
return(create(new List <Genotype>(toCopy.Genotypes)));
}