/// <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>
/// 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();
}
}
private void fullyDecodeGenotypes (VariantContextBuilder builder, VCFHeader header)
{
GenotypesContext gc = new GenotypesContext ();
foreach (Genotype g in Genotypes) {
gc.Add (fullyDecodeGenotypes (g, header));
}
builder.SetGenotypes (gc, false);
}
/// <summary>
/// Parses a line from a VCF File
/// </summary>
/// <param name="parts">An array of length >8 where the 9th element contains unsplit genotype data (if present)</param>
/// <param name="includeGenotypes"> Whether or not to also parse the genotype data </param>
/// <returns></returns>
private VariantContext parseVCFLine(string[] parts, bool includeGenotypes)
{
VariantContextBuilder builder = new VariantContextBuilder();
builder.Source = Name;
// increment the line count
lineNo++;
// parse out the required fields
string chr = GetCachedString(parts [0]);
builder.Contig = chr;
int pos = -1;
try {
pos = Convert.ToInt32(parts [1]);
} catch (FormatException e) {
generateException(parts [1] + " is not a valid start position in the VCF format");
}
builder.Start = pos;
if (parts [2].Length == 0)
{
generateException("The VCF specification requires a valid ID field");
}
else if (parts [2].Equals(VCFConstants.EMPTY_ID_FIELD))
{
builder.ID = VCFConstants.EMPTY_ID_FIELD;
}
else
{
builder.ID = parts [2];
}
string refe = GetCachedString(parts [3].ToUpper());
string alts = GetCachedString(parts [4].ToUpper());
builder.Log10PError = parseQual(parts [5]);
string filterStr = GetCachedString(parts [6]);
var filters = filterHash [filterStr];
if (filters != null) //means filter data present
{
builder.SetFilters(filters.Hash);
}
IDictionary <string, object> attrs = parseInfo(parts [7]);
builder.Attributes = attrs;
if (attrs.ContainsKey(VCFConstants.END_KEY))
{
// update stop with the end key if provided
try {
builder.Stop = Convert.ToInt32(attrs [VCFConstants.END_KEY].ToString());
} catch (Exception e) {
generateException("the END value in the INFO field is not valid");
}
}
else
{
builder.Stop = (pos + refe.Length - 1);
}
// get our alleles, filters, and setup an attribute map
IList <Allele> alleles = parseAlleles(refe, alts, lineNo);
builder.SetAlleles(alleles);
// do we have genotyping data
if (parts.Length > NUM_STANDARD_FIELDS && includeGenotypes)
{
int nGenotypes = header.NGenotypeSamples;
LazyGenotypesContext lazy = new LazyGenotypesContext(this, alleles, chr, pos, parts [8], nGenotypes);
// did we resort the sample names? If so, we need to load the genotype data
if (!header.SamplesWereAlreadySorted)
{
lazy.Decode();
}
builder.SetGenotypes(lazy, false);
}
VariantContext vc = null;
try {
vc = builder.make();
} catch (Exception e) {
generateException(e.Message);
}
return(vc);
}
/// <summary>
/// Parses a line from a VCF File
/// </summary>
/// <param name="parts">An array of length >8 where the 9th element contains unsplit genotype data (if present)</param>
/// <param name="includeGenotypes"> Whether or not to also parse the genotype data </param>
/// <returns></returns>
private VariantContext parseVCFLine (string[] parts, bool includeGenotypes)
{
VariantContextBuilder builder = new VariantContextBuilder ();
builder.Source = Name;
// increment the line count
lineNo++;
// parse out the required fields
string chr = GetCachedString (parts [0]);
builder.Contig = chr;
int pos = -1;
try {
pos = Convert.ToInt32 (parts [1]);
} catch (FormatException e) {
generateException (parts [1] + " is not a valid start position in the VCF format");
}
builder.Start = pos;
if (parts [2].Length == 0) {
generateException ("The VCF specification requires a valid ID field");
} else if (parts [2].Equals (VCFConstants.EMPTY_ID_FIELD)) {
builder.ID = VCFConstants.EMPTY_ID_FIELD;
} else {
builder.ID = parts [2];
}
string refe = GetCachedString (parts [3].ToUpper ());
string alts = GetCachedString (parts [4].ToUpper ());
builder.Log10PError = parseQual (parts [5]);
string filterStr = GetCachedString (parts [6]);
var filters = filterHash [filterStr];
if (filters != null) {//means filter data present
builder.SetFilters (filters.Hash);
}
IDictionary<string, object> attrs = parseInfo (parts [7]);
builder.Attributes = attrs;
if (attrs.ContainsKey (VCFConstants.END_KEY)) {
// update stop with the end key if provided
try {
builder.Stop = Convert.ToInt32 (attrs [VCFConstants.END_KEY].ToString ());
} catch (Exception e) {
generateException ("the END value in the INFO field is not valid");
}
} else {
builder.Stop = (pos + refe.Length - 1);
}
// get our alleles, filters, and setup an attribute map
IList<Allele> alleles = parseAlleles (refe, alts, lineNo);
builder.SetAlleles (alleles);
// do we have genotyping data
if (parts.Length > NUM_STANDARD_FIELDS && includeGenotypes) {
int nGenotypes = header.NGenotypeSamples;
LazyGenotypesContext lazy = new LazyGenotypesContext (this, alleles, chr, pos, parts [8], nGenotypes);
// did we resort the sample names? If so, we need to load the genotype data
if (!header.SamplesWereAlreadySorted) {
lazy.Decode ();
}
builder.SetGenotypes (lazy, false);
}
VariantContext vc = null;
try {
vc = builder.make ();
} catch (Exception e) {
generateException (e.Message);
}
return vc;
}