public void ConvertToMatrix(string inputPath, string outputPath, VariantFileType intputFileType,
VariantEncoding sourceEncoding,
VariantEncoding destinationEncoding, string snpPath = null, IProgressReporter progress = null)
{
/* string[] sampleIds = null;
var variants = new List<VariantCall>();
StringSplitter splitter = null;
var variantHeader = new StringBuilder();
using (var reader = _fileReaderFactory.GetFileReader(inputPath))
{
foreach (var line in reader)
{
if (intputFileType == VariantFileType.Vcf)
{
//Check for comment headers
if (line.Length < 2)
{
continue;
}
if (line.Substring(0, 2) == "##")
{
continue;
}
if (line[0] == '#')
{
var header = line.Split('\t');
sampleIds = new string[header.Length - 9];
for (var i = 9; i < header.Length; i++)
{
sampleIds[i - 9] = header[i];
}
variantHeader.Append("SampleId");
continue;
}
if (splitter == null)
{
var testSplit = line.Split('\t');
splitter = new StringSplitter(testSplit.Length);
}
//Split the line out
//var lineSplit = line.Split('\t');
splitter.Split(line, '\t');
var lineSplit = splitter.Results;
//Check if we passed QC
if (lineSplit[6] != "PASS")
{
continue;
}
//Check if id starts with rs
if (lineSplit[2][0] != 'r')
{
continue;
}
if (lineSplit[2][1] != 's')
{
continue;
}
//Create variant
var newVar = new VariantCall();
//newVar.Chromosome = lineSplit[0];
//newVar.ChromPos = Convert.ToInt32(lineSplit[1]);
newVar.Id = lineSplit[2];
newVar.dbSNPRef = lineSplit[3];
newVar.VcfAlt = lineSplit[4];
var tempBuilder = new StringBuilder();
var callA = lineSplit[9][0] == '0' ? newVar.dbSNPRef : newVar.VcfAlt;
var callB = lineSplit[9][2] == '0' ? newVar.dbSNPRef : newVar.VcfAlt;
tempBuilder.Append(callA);
tempBuilder.Append("_");
tempBuilder.Append(callB);
tempBuilder.Append("\t");
for (var i = 10; i < lineSplit.Length; i++)
{
callA = lineSplit[9][0] == '0' ? newVar.dbSNPRef : newVar.VcfAlt;
callB = lineSplit[9][2] == '0' ? newVar.dbSNPRef : newVar.VcfAlt;
tempBuilder.Append(callA);
tempBuilder.Append("_");
tempBuilder.Append(callB);
tempBuilder.Append("\t");
}
newVar.OutputCalls = tempBuilder.ToString();
variants.Add(newVar);
variantHeader.Append("\t");
variantHeader.Append(newVar.Id);
if (progress != null)
{
var perc = (double) reader.Stream.Position/reader.Stream.Length;
if (perc > 0.05)
{
break;
}
progress.Update(perc/2);
}
}
}
}
using (var writer = new StreamWriter(File.Open(outputPath, FileMode.Create)))
{
writer.WriteLine(variantHeader);
var totalSamples = sampleIds.Length;
var totalVariants = variants.Count;
splitter = new StringSplitter(totalSamples);
for (var i = 0; i < totalSamples; i++)
{
var lineBuilder = new StringBuilder();
lineBuilder.Append(sampleIds[i]);
for (var j = 0; j < totalVariants; j++)
{
splitter.Split(variants[j].OutputCalls, "\t");
lineBuilder.Append("\t");
lineBuilder.Append(splitter.Results[i]);
}
writer.WriteLine(lineBuilder);
if (progress != null)
{
var perc = (double)i / totalSamples;
progress.Update((perc / 2) + 0.5);
}
}
}*/
}
private VariantCall[] LoadVariant(VariantFileType fileType, string line)
{
if (fileType == VariantFileType.Vcf)
{
return LoadVcfVariant(line);
}
if (fileType == VariantFileType.IlluminiaFinalReport)
{
return LoadIlluminaFinalReportVariant(line);
}
if (fileType == VariantFileType.IlluminaMatrix)
{
return LoadIlluminaMatrixReportVariant(line);
}
return null;
}
public void AnalyseVcfFile(string variantPath, VariantFileType fileType, string snpPath = null,
ConsoleLogger logger = null)
{
ProgressReporter pr = null;
logger?.LogInfo(LogName, "Analysing variants");
//Do we need to load snps?
if (snpPath != null)
{
pr = null;
if (logger != null)
{
pr = new ProgressReporter(logger, 1);
}
logger?.StartTask(1, "Load SNP file");
LoadUcscSnps(snpPath, true, pr);
logger?.EndTask(1);
logger?.LogInfo(LogName, "Loaded " + Snps.Count + " SNPs");
}
//Init
_splitter = null;
var variantCount = 0;
var unmatchedCount = 0;
//Var vcf
var calledForwardOnPlus = 0;
var calledForwardOnNeg = 0;
//Var final report
var matchedFwdStrand = 0;
var unMatchedFwdStrand = 0;
//Logging
pr = null;
if (logger != null)
{
pr = new ProgressReporter(logger, 1);
}
logger?.StartTask(1, "Processing variants");
//Read file
using (var reader = _fileReaderFactory.GetFileReader(variantPath))
{
foreach (var line in reader)
{
//Load variants
var variants = LoadVariant(fileType, line);
//Continue if not a real variants
if (variants == null)
{
continue;
}
for (var i = 0; i < variants.Length; i++)
{
//Get current variant
var variant = variants[i];
//Try and match to snp's loads
Snp matchSnp;
Snps.TryGetValue(variant.Id, out matchSnp);
//Check we have a match
if (matchSnp != null)
{
if (fileType == VariantFileType.Vcf)
{
if (matchSnp.Strand == "+" && matchSnp.SnpRef == variant.VcfRef)
{
calledForwardOnPlus++;
}
else if (matchSnp.Strand == "-" && matchSnp.SnpRef == variant.VcfRef)
{
calledForwardOnNeg++;
}
}
else if (fileType == VariantFileType.IlluminiaFinalReport)
{
if (variant.AlleleA != "-" && variant.AlleleB != "-")
{
if ((variant.AlleleA == matchSnp.SnpRef || variant.AlleleA == matchSnp.SnpAlt) &&
(variant.AlleleB == matchSnp.SnpRef || variant.AlleleB == matchSnp.SnpAlt))
{
matchedFwdStrand++;
}
else
{
if (matchSnp.Observed.Length <= 3)
{
unMatchedFwdStrand++;
}
}
}
}
else if (fileType == VariantFileType.IlluminaMatrix)
{
if (variant.AlleleA != "-" && variant.AlleleB != "-")
{
if ((variant.AlleleA == matchSnp.SnpRef || variant.AlleleA == matchSnp.SnpAlt) &&
(variant.AlleleB == matchSnp.SnpRef || variant.AlleleB == matchSnp.SnpAlt))
{
matchedFwdStrand++;
}
else
{
if (matchSnp.Observed.Length <= 3)
{
unMatchedFwdStrand++;
}
}
}
}
}
else
{
unmatchedCount++;
}
variantCount++;
if (logger != null)
{
var perc = (double) reader.Stream.Position/reader.Stream.Length;
pr.Update(perc);
if (fileType == VariantFileType.Vcf)
{
pr.Update(
$" TV:{variantCount} UTV:{unmatchedCount} :CallFwd+:{calledForwardOnPlus} CallFwd-:{calledForwardOnNeg}");
}
else if (fileType == VariantFileType.IlluminiaFinalReport)
{
pr.Update(
$" TV:{variantCount} UTV:{unmatchedCount} MatchedFwd:{matchedFwdStrand} UnMatchedFwd:{unMatchedFwdStrand}");
}
else if (fileType == VariantFileType.IlluminaMatrix)
{
pr.Update(
$" TV:{variantCount} UTV:{unmatchedCount} MatchedFwd:{matchedFwdStrand} UnMatchedFwd:{unMatchedFwdStrand}");
}
}
}
}
}
logger?.EndTask(1);
}