public void WhenIOpenTheFile()
{
//ScenarioContext.Current.Pending();
Vcp = new VCFParser(VcfFile);
Assert.IsNotNull(Vcp);
}
private bool tmpLoadVCF(string fileName)
{
if (!ValidateFileName(fileName, out fileName))
{
return(false);
}
_parserSTW.Restart();
VCFParser <Variant, VariantData> vcfParser = new VCFParser <Variant, VariantData>(source: fileName, species: Genomes.HomoSapiens, assembly: Assemblies.hg19);
try { Repository.parsedVariants = vcfParser.Parse(); }
catch (Exception e)
{
if (Path.GetDirectoryName(fileName) + Path.DirectorySeparatorChar == _workingDirectory &&
Path.GetExtension(fileName) == _logFile)
{
Herald.Announce(Herald.MessageType.Error, string.Format("The requested extension should not have same extension as the log file."));
}
else
{
Herald.Announce(Herald.MessageType.Error, string.Format("{0}", e.Message));
}
return(false);
}
_parserSTW.Stop();
_accumulatedLoadET += _parserSTW.Elapsed.TotalSeconds;
Herald.AnnounceExeReport("Loaded", new ExecutionReport(Repository.parsedVariants.intervalsCount, _parserSTW.Elapsed));
return(true);
}
public void OneTranscriptOneHeterozygousSynonymous()
{
Genome genome = new Genome(Path.Combine(TestContext.CurrentContext.TestDirectory, "TestData", "chr1_sample.fa"));
VCFParser vcf = new VCFParser(Path.Combine(TestContext.CurrentContext.TestDirectory, "TestVcfs", "chr_1_one_heterozygous_synonymous.vcf"));
List <Variant> variants = vcf.Select(x => new Variant(null, x, genome)).ToList();
Assert.AreEqual(1, variants.Count);
GeneModel geneModel = new GeneModel(genome, Path.Combine(TestContext.CurrentContext.TestDirectory, "TestData", "chr1_one_transcript.gtf"));
List <Protein> proteins_wo_variant = geneModel.Translate(true).ToList();
List <Transcript> transcripts = geneModel.ApplyVariants(variants);
List <Protein> proteins = transcripts.Select(t => t.Protein()).ToList();
Assert.AreEqual(1, geneModel.Genes.Count);
Assert.AreEqual(1, proteins.Count);
Assert.AreEqual(1, proteins_wo_variant.Count);
Assert.AreEqual(1, new HashSet <string> {
proteins[0].BaseSequence, proteins_wo_variant[0].BaseSequence
}.Count);
Assert.IsTrue(proteins.Any(p => p.FullName.Contains(FunctionalClass.SILENT.ToString()))); // synonymous
Assert.IsTrue(proteins.Any(p => p.FullName.Contains(GenotypeType.HETEROZYGOUS.ToString()))); // synonymous
Assert.IsTrue(proteins.Any(p => p.FullName.Contains("1:69666")));
string proteinFasta = Path.Combine(TestContext.CurrentContext.TestDirectory, "TestVcfs", "chr_1_one_heterozygous_synonymous.fasta");
ProteinDbWriter.WriteFastaDatabase(proteins, proteinFasta, " ");
string[] proteinFastaLines = File.ReadLines(proteinFasta).ToArray();
Assert.IsTrue(proteinFastaLines[0].Contains(FunctionalClass.SILENT.ToString())); // synonymous
Assert.IsTrue(proteinFastaLines[0].Contains("1:69666"));
}
public void Chr19VariantTranscript()
{
Genome genome = new Genome(Path.Combine(TestContext.CurrentContext.TestDirectory, "Homo_sapiens.GRCh38.dna.chromosome.19.fa"));
GeneModel geneModel = new GeneModel(genome, Path.Combine(TestContext.CurrentContext.TestDirectory, "ProblematicChr19", "chr19variantTranscript.gff3"));
var variants = new VCFParser(Path.Combine(TestContext.CurrentContext.TestDirectory, "ProblematicChr19", "chr19problematic.vcf"))
.Select(v => new Variant(null, v, genome.Chromosomes[0]))
.Where(v => v.SecondAlleleString.Length == 1 && v.ReferenceAlleleString.Length == 1).ToList();
List <Transcript> transcripts = geneModel.ApplyVariants(variants).ToList();
List <Protein> proteins = transcripts.Select(t => t.Protein(null)).ToList();
}
public void ParseVcf([Values("testData/NA12878.knowledgebase.snapshot.20131119.b37.vcf.gz")] string fname)
{
Assert.IsTrue(File.Exists(fname));
VCFParser vcp = new VCFParser(fname);
var j = vcp.First();
var ii = j.Genotypes;
var i = vcp.Select(x => x.NoCallCount).Count();
Console.WriteLine("Count: {0}", i);
}
static void Main(string[] args)
{
fname = "testData/NA12878.knowledgebase.snapshot.20131119.b37.vcf.gz";
System.Diagnostics.Stopwatch sw = new System.Diagnostics.Stopwatch();
sw.Start();
VCFParser vcp = new VCFParser(fname);
var j = vcp.First();
var ii = j.Genotypes;
var i = vcp.Select(x => x.NoCallCount).Count();
sw.Stop();
Console.WriteLine(sw.Elapsed.ToString());
}
static void Main (string[] args)
{
var fname = "testData/NA12878.knowledgebase.snapshot.20131119.b37.vcf.gz";
System.Diagnostics.Stopwatch sw = new System.Diagnostics.Stopwatch ();
sw.Start ();
VCFParser vcp = new VCFParser (fname);
var j = vcp.First ();
var ii = j.Genotypes.Count;
var i = vcp.Select (x => x.NoCallCount).Count ();
sw.Stop ();
Console.WriteLine("Count: {0}/{1}", ii,i);
Console.WriteLine ("Elapsed: {0}", sw.Elapsed);
var anyKey = Console.ReadKey();
}
public void TestMissenseMutation()
{
// Make a transcript
Sequence seq = new Sequence(Alphabets.DNA, "AAA".Select(cc => (byte)cc).ToArray(), false);
seq.ID = "1";
Chromosome c = new Chromosome(seq, null);
Gene g = new Gene("", c, "", "+", 1, 3, null);
Transcript t = new Transcript("", g, "", "+", 1, 3, "", null, null);
Exon x = new Exon(t, seq, "", 1, 3, seq.ID, "+", null, null);
t.Exons = new List <Exon> {
x
};
CDS cds = new CDS(t, seq.ID, "", "+", 1, 3, null, 0);
t.CodingDomainSequences = new List <CDS> {
cds
};
// Make a missense mutation
// ugh.vcf has a homozygous variation that should change the codon from AAA to AGA, which code for K and R
// # CHROM POS ID REF ALT QUAL FILTER INFO FORMAT sample
// 1 2 . A G 64.77 . info GT:AD:DP:GQ:PL 1/1:2,3:5:69:93,0,69
List <Variant> variants = new VCFParser(Path.Combine(TestContext.CurrentContext.TestDirectory, "TestVcfs", "ugh.vcf")).Select(v => new Variant(null, v, new Chromosome(seq, null))).ToList();
// Make sure it makes it into the DNA sequence
t.Variants = new HashSet <Variant>(variants);
List <Transcript> variantTranscripts = GeneModel.ApplyVariantsCombinitorially(t);
Assert.AreEqual("AAA", SequenceExtensions.ConvertToString(t.Exons[0].Sequence));
Assert.AreEqual("K", t.Protein().BaseSequence);
Assert.AreEqual("AGA", SequenceExtensions.ConvertToString(variantTranscripts[0].Exons[0].Sequence));
Assert.AreEqual("R", variantTranscripts[0].Protein().BaseSequence);
// Make sure it gets annotated as a missense mutation
Assert.IsTrue(variantTranscripts[0].VariantAnnotations.Any(str => str.Contains(FunctionalClass.MISSENSE.ToString())));
}
public override void LoadData(string filePath)
{
var vcfParser = new VCFParser(filePath);
var header = vcfParser.Header;
var collectionInformationList = new List<patient_variant_information>();
var patient = new patient();
// We pull out all of the metadata from the header (all lines) and write them as information
// lines associated with this result.
foreach (var headerItem in header.MetaDataInInputOrder)
{
if (headerItem.Key == "individual-id")
{
var individualParts = headerItem.Value.Replace("<", "").Replace(">", "").Split(new char[] { ',' });
var individualData = individualParts.Select(x => x.Split(new char[] { '=' })).ToArray();
var mrnParts = individualData.FirstOrDefault(x => x[0] == "Dbxref")[1].Split(':');
patient = patientRepo.AddPatient(mrnParts[1], mrnParts[0],
individualData.FirstOrDefault(x => x[0] == "First_name")[1],
individualData.FirstOrDefault(x => x[0] == "Last_name")[1],
DateTime.Parse(individualData.FirstOrDefault(x => x[0] == "DOB")[1]));
}
else if (headerItem.GetType() == typeof(VCFInfoHeaderLine))
{
var info = headerItem as VCFInfoHeaderLine;
collectionInformationList.Add(AddHeaderInformation(string.Format("VCF:{0}", headerItem.Key), CleanHeaderValue("INFO", info.ToString())));
}
else if (headerItem.GetType() == typeof(VCFFilterHeaderLine))
{
var filter = headerItem as VCFFilterHeaderLine;
collectionInformationList.Add(AddHeaderInformation(string.Format("VCF:{0}", headerItem.Key), CleanHeaderValue("FILTER", filter.ToString())));
}
else if (headerItem.GetType() == typeof(VCFFormatHeaderLine))
{
var format = headerItem as VCFFormatHeaderLine;
collectionInformationList.Add(AddHeaderInformation(string.Format("VCF:{0}", headerItem.Key), CleanHeaderValue("FORMAT", format.ToString())));
}
else
{
collectionInformationList.Add(AddHeaderInformation(string.Format("VCF:{0}", headerItem.Key), headerItem.Value));
}
}
var reference = header.MetaDataInInputOrder.First(x => x.Key == "reference").Value;
DateTime? resultDate = DateTime.ParseExact(header.MetaDataInInputOrder.First(x => x.Key == "fileDate").Value,
"yyyyMMdd", CultureInfo.InvariantCulture, DateTimeStyles.None);
var patientVariants = new List<patient_variants>();
var featureInformationList = new Dictionary<patient_variants, List<patient_variant_information>>();
while (vcfParser.MoveNext())
{
var current = vcfParser.Current;
var variant = variantRepo.AddVariant(null, current.ID, "dbSNP",
current.Chr, current.Start, current.End,
reference, current.Reference.BaseString);
var patientVariant = new patient_variants()
{
patient_id = patient.id,
reference_id = variant.id,
resulted_on = resultDate,
variant_type = Enums.PatientVariantType.SNP
};
SetVariantValues(patientVariant, current);
patientVariants.Add(patientVariant);
var attributeList = new List<patient_variant_information>();
foreach (var attribute in current.Attributes)
{
attributeList.Add(AddVariantInformation(string.Format("VCF:{0}", attribute.Key), attribute.Value.ToString()));
}
if (current.FiltersMaybeNull != null)
{
foreach (var filter in current.FiltersMaybeNull)
{
attributeList.Add(AddVariantInformation("VCF:Filter", filter));
}
}
foreach (var genotype in current.Genotypes)
{
attributeList.Add(AddVariantInformation("VCF:Genotype", genotype.ToMHGRString()));
}
attributeList.Add(AddVariantInformation("VCF:Quality", current.PhredScaledQual.ToString()));
attributeList.Add(AddVariantInformation("VCF:Filter", string.Join(",", current.Filters.ToArray())));
featureInformationList.Add(patientVariant, attributeList);
}
// Save the collection to get its ID
var source = sourceRepo.AddSource("VCF", "VCF file");
var file = AddResultFile(filePath, source);
var collection = patientRepo.AddCollection(patient, file);
// Save the collection-level header data
collectionInformationList.ForEach(x => x.item_id = collection.id);
variantRepo.AddPatientVariantInformationList(collectionInformationList);
variantRepo.AddPatientVariants(patientVariants);
// Save the individual attributes associated with each feature.
// Must be done after the patient variants are written to DB (above), since we
// rely on the ID being set.
foreach (var pair in featureInformationList)
{
foreach (var attribute in pair.Value)
{
attribute.item_id = pair.Key.id;
}
variantRepo.AddPatientVariantInformationList(pair.Value);
}
variantRepo.AddPatientVariantsToCollection(collection, patientVariants);
featureInformationList.Clear();
collectionInformationList.Clear();
patientVariants.Clear();
}
public void test()
{
VCFParser parser = new VCFParser(Path.Combine(TestContext.CurrentContext.TestDirectory, @"testData", @"NA12878.knowledgebase.snapshot.20131119.b37.vcf.gz"));
List <VariantContext> context = parser.Select(x => x).ToList();
}
public override void LoadData(string filePath)
{
var vcfParser = new VCFParser(filePath);
var header = vcfParser.Header;
var patient = new patient();
var source = sourceRepo.AddSource("VCF", "VCF file");
var file = AddResultFile(filePath, source);
// Process the file-level pragmas
result_entities rootEntity = new result_entities()
{
attribute_id = EntityRepository.GetAttribute(null, null, "Variant Call Format result", null).id,
result_file_id = file.id
};
// We pull out all of the metadata from the header (all lines) and write them as information
// lines associated with this result.
var headerEntities = new List<result_entities>();
foreach (var headerItem in header.MetaDataInInputOrder)
{
if (headerItem.Key == "individual-id")
{
var individualParts = headerItem.Value.Replace("<", "").Replace(">", "").Split(new char[] { ',' });
var individualData = individualParts.Select(x => x.Split(new char[] { '=' })).ToArray();
var mrnParts = individualData.FirstOrDefault(x => x[0] == "Dbxref")[1].Split(':');
patient = patientRepo.AddPatient(mrnParts[1], mrnParts[0],
individualData.FirstOrDefault(x => x[0] == "First_name")[1],
individualData.FirstOrDefault(x => x[0] == "Last_name")[1],
DateTime.Parse(individualData.FirstOrDefault(x => x[0] == "DOB")[1]));
}
else if (headerItem.Key == "fileDate")
{
DateTime resultDate = DateTime.ParseExact(headerItem.Value, "yyyyMMdd", CultureInfo.InvariantCulture, DateTimeStyles.None);
headerEntities.Add(CreateEntityAttribute("Resulted on", 0, file.id, rootEntity, resultDate.ToShortDateString()));
}
else if (headerItem.GetType() == typeof(VCFInfoHeaderLine))
{
var info = headerItem as VCFInfoHeaderLine;
var infoEntity = CreateEntityAttribute("INFO", 0, file.id, rootEntity, null);
headerEntities.Add(infoEntity);
headerEntities.Add(CreateEntityAttribute("ID", 0, file.id, infoEntity, info.ID));
headerEntities.Add(CreateEntityAttribute("Number", 0, file.id, infoEntity, info.CountType.ToString()));
headerEntities.Add(CreateEntityAttribute("Type", 0, file.id, infoEntity, info.Type.ToString()));
headerEntities.Add(CreateEntityAttribute("Description", 0, file.id, infoEntity, info.Description));
}
else if (headerItem.GetType() == typeof(VCFFilterHeaderLine))
{
var filter = headerItem as VCFFilterHeaderLine;
var filterEntity = CreateEntityAttribute("FILTER", 0, file.id, rootEntity, null);
headerEntities.Add(filterEntity);
foreach (var field in filter.GenericFields())
{
headerEntities.Add(CreateEntityAttribute(field.Key, 0, file.id, filterEntity, field.Value));
}
}
else if (headerItem.GetType() == typeof(VCFFormatHeaderLine))
{
var format = headerItem as VCFFormatHeaderLine;
var formatEntity = CreateEntityAttribute("FORMAT", 0, file.id, rootEntity, null);
headerEntities.Add(formatEntity);
headerEntities.Add(CreateEntityAttribute("ID", 0, file.id, formatEntity, format.ID));
headerEntities.Add(CreateEntityAttribute("Number", 0, file.id, formatEntity, format.CountType.ToString()));
headerEntities.Add(CreateEntityAttribute("Type", 0, file.id, formatEntity, format.Type.ToString()));
headerEntities.Add(CreateEntityAttribute("Description", 0, file.id, formatEntity, format.Description));
}
else
{
var headerEntity = CreateEntityAttribute(headerItem.Key, 0, file.id, rootEntity, headerItem.Value);
headerEntities.Add(headerEntity);
}
}
rootEntity.patient_id = patient.id;
headerEntities.ForEach(x => x.patient_id = patient.id);
var variantEntities = new List<result_entities>();
while (vcfParser.MoveNext())
{
var current = vcfParser.Current;
result_entities variantEntity = new result_entities()
{
attribute_id = EntityRepository.GetAttribute(null, null, "Variant Call Format variant", null).id,
result_file_id = file.id,
patient_id = patient.id,
parent = rootEntity
};
variantEntities.Add(variantEntity);
result_entities snpEntity = new result_entities()
{
attribute_id = EntityRepository.GetAttribute(current.ID, "dbSNP", null, null).id,
result_file_id = file.id,
patient_id = patient.id,
parent = variantEntity
};
variantEntities.Add(snpEntity);
SetVariantValues(current, patient.id, file.id, snpEntity, variantEntities);
variantEntities.Add(CreateEntityAttribute("Chromosome", patient.id, file.id, variantEntity, current.Chr));
variantEntities.Add(CreateEntityAttribute("Start position", patient.id, file.id, variantEntity, current.Start.ToString()));
variantEntities.Add(CreateEntityAttribute("End position", patient.id, file.id, variantEntity, current.End.ToString()));
variantEntities.Add(CreateEntityAttribute("Reference base", patient.id, file.id, variantEntity, current.Reference.BaseString));
variantEntities.Add(CreateEntityAttribute("Quality", patient.id, file.id, variantEntity, current.PhredScaledQual.ToString()));
foreach (var attr in current.Attributes)
{
variantEntities.Add(CreateEntityAttribute(string.Format("INFO:{0}", attr.Key), patient.id, file.id, variantEntity, attr.Value.ToString()));
}
if (current.FiltersMaybeNull != null)
{
foreach (var filter in current.FiltersMaybeNull)
{
variantEntities.Add(CreateEntityAttribute(string.Format("FILTER:{0}", filter), patient.id, file.id, variantEntity, string.Empty));
}
}
//foreach (var genotype in current.Genotypes)
//{
// attributeList.Add(AddVariantInformation("VCF:Genotype", genotype.ToMHGRString()));
//}
//attributeList.Add(AddVariantInformation("VCF:Quality", current.PhredScaledQual.ToString()));
//attributeList.Add(AddVariantInformation("VCF:Filter", string.Join(",", current.Filters.ToArray())));
//featureInformationList.Add(patientVariant, attributeList);
}
entityRepo.AddVCF(rootEntity, headerEntities, variantEntities);
}
