public static List <string> GetAllHeaderLines(string vcfPath)
{
List <string> header;
using (var reader = new AlleleReader(vcfPath))
{
header = reader.HeaderLines;
}
return(header);
}
/// <summary>
/// Load a list of all variants in a file. This is memory-intensive; don't do this for whole-genome vcf files!
/// </summary>
public static List <CalledAllele> GetAllVariantsInFile(string vcfPath)
{
List <CalledAllele> allVariants = new List <CalledAllele>();
using (AlleleReader reader = new AlleleReader(vcfPath))
{
foreach (CalledAllele variant in reader.GetVariants())
{
allVariants.Add(variant);
}
}
return(allVariants);
}
public static Dictionary <string, List <CandidateAllele> > GetVariantsByChromosome(this AlleleReader reader,
bool variantsOnly = false, bool flagIsKnown = false, List <AlleleCategory> typeFilter = null, Func <CandidateAllele, bool> doSkipCandidate = null)
{
var lookup = new Dictionary <string, List <CandidateAllele> >();
var calledVariants = reader.GetVariants();
foreach (var calledVariant in calledVariants)
{
var candidate = BackToCandiate(calledVariant);
if (candidate.Type != AlleleCategory.Unsupported)
{
if (variantsOnly && candidate.Type == AlleleCategory.Reference)
{
continue;
}
if (typeFilter != null && !typeFilter.Contains(candidate.Type))
{
continue;
}
if (doSkipCandidate != null && doSkipCandidate(candidate))
{
continue;
}
if (flagIsKnown)
{
candidate.IsKnown = true;
}
if (!lookup.ContainsKey(candidate.Chromosome))
{
lookup[candidate.Chromosome] = new List <CandidateAllele>();
}
lookup[candidate.Chromosome].Add(candidate);
}
}
return(lookup);
}
/// <summary>
/// Take in a vcf, do stuff to it, write out a vcf. Streamed line by line, loci by loci, so as not to blow up your computer.
/// </summary>
/// <param name="vcfOut"> the output file name</param>
/// <param name="options"> all the parameters associated with writing out a vcf</param>
/// <param name="recalibrationData">the data you need for doing your "stuff" </param>
/// <param name="whatToDoWithSingleAllele">how you want to change each allele</param>
/// <param name="whatToDoWithCoLocatedAlleles">how you want to change each set of alleles, by loci</param>
/// <param name="canSkipLinesWithoutProcessing">when you can skip lines (saves CPU time)</param>
/// <param name="getVcfFileWriter">what your special vcf writer should be, includes special header lines, etc</param>
/// <param name="shouldTrimComplexAlleles">if ACGT-> ACCT is ok, or if you want it trimmed to G -> C. this might affect position and ordering. Generally turn if OFF for processing vcfs, post scylla. </param>
private static void UpdateVcf(string vcfOut, VcfConsumerAppOptions options, bool shouldTrimComplexAlleles, T recalibrationData,
UpdateSingleAlleleMethod whatToDoWithSingleAllele, UpdateCoLocatedAllelesMethod whatToDoWithCoLocatedAlleles,
CanSkipVcfLinesMethod canSkipLinesWithoutProcessing, GetVcfFileWriter getVcfFileWriter)
{
using (AlleleReader reader = new AlleleReader(options.VcfPath, shouldTrimComplexAlleles))
{
using (VcfFileWriter writer = getVcfFileWriter(options, vcfOut))
{
writer.WriteHeader();
writer.FlushBuffer();
var variantListFromFile = new List <CalledAllele>()
{
};
string incomingHangingLine = null;
string outgoingHangingLine = null;
while (true)
{
//get the next group to process
incomingHangingLine = outgoingHangingLine;
var coLocatedVcfLinesToProcess = reader.CloseColocatedLines(incomingHangingLine,
out outgoingHangingLine);
//how we know we are done
if (coLocatedVcfLinesToProcess.Count == 0)
{
break;
}
bool updateNeededForLocus = false;
TypeOfUpdateNeeded updatedNeededForLine = canSkipLinesWithoutProcessing(coLocatedVcfLinesToProcess);
switch (updatedNeededForLine)
{
case TypeOfUpdateNeeded.NoChangeNeeded:
writer.Write(coLocatedVcfLinesToProcess);
break;
case TypeOfUpdateNeeded.Modify:
//then we need to change them into alleles and do stuff to them
variantListFromFile = AlleleReader.VcfLinesToAlleles(coLocatedVcfLinesToProcess);
List <CalledAllele> modifiedVariantListToWrite = WhatToDoToAlleles(options, recalibrationData,
whatToDoWithSingleAllele, whatToDoWithCoLocatedAlleles, variantListFromFile, ref updateNeededForLocus);
if (updateNeededForLocus)
{
writer.Write(modifiedVariantListToWrite);
}
else
{
writer.Write(coLocatedVcfLinesToProcess);
}
break;
case TypeOfUpdateNeeded.DeleteCompletely:
default:
break;
}
}
}
}
}