def __init__(self,
localId,
referenceSet,
randomSeed=0,
numVariantSets=1,
numCalls=1,
variantDensity=0.5,
numReadGroupSets=1,
numReadGroupsPerReadGroupSet=1,
numAlignments=1,
numFeatureSets=1):
super(SimulatedDataset, self).__init__(localId)
self._description = "Simulated dataset {}".format(localId)
# TODO create a simulated Ontology
# Variants
for i in range(numVariantSets):
localId = "simVs{}".format(i)
seed = randomSeed + i
variantSet = variants.SimulatedVariantSet(self, referenceSet,
localId, seed, numCalls,
variantDensity)
callSets = variantSet.getCallSets()
# Add biosamples
for callSet in callSets:
bioSample = biodata.BioSample(self, callSet.getLocalId())
bioSample2 = biodata.BioSample(self,
callSet.getLocalId() + "2")
individual = biodata.Individual(self, callSet.getLocalId())
bioSample.setIndividualId(individual.getId())
bioSample2.setIndividualId(individual.getId())
self.addIndividual(individual)
self.addBioSample(bioSample)
self.addBioSample(bioSample2)
self.addVariantSet(variantSet)
variantAnnotationSet = variants.SimulatedVariantAnnotationSet(
variantSet, "simVas{}".format(i), seed)
variantSet.addVariantAnnotationSet(variantAnnotationSet)
# Reads
for i in range(numReadGroupSets):
localId = 'simRgs{}'.format(i)
seed = randomSeed + i
readGroupSet = reads.SimulatedReadGroupSet(
self, localId, referenceSet, seed,
numReadGroupsPerReadGroupSet, numAlignments)
for rg in readGroupSet.getReadGroups():
bioSample = biodata.BioSample(self, rg.getLocalId())
individual = biodata.Individual(self, rg.getLocalId())
bioSample.setIndividualId(individual.getId())
rg.setBioSampleId(bioSample.getId())
self.addIndividual(individual)
self.addBioSample(bioSample)
self.addReadGroupSet(readGroupSet)
# Features
for i in range(numFeatureSets):
localId = "simFs{}".format(i)
seed = randomSeed + i
featureSet = sequenceAnnotations.SimulatedFeatureSet(
self, localId, seed)
featureSet.setReferenceSet(referenceSet)
self.addFeatureSet(featureSet)
def addBioSample(self):
"""
Adds a new biosample into this repo
"""
self._openRepo()
dataset = self._repo.getDatasetByName(self._args.datasetName)
bioSample = bio_metadata.BioSample(
dataset, self._args.bioSampleName)
bioSample.populateFromJson(self._args.bioSample)
self._updateRepo(self._repo.insertBioSample, bioSample)
def _readBioSampleTable(self, cursor):
cursor.row_factory = sqlite3.Row
cursor.execute("SELECT * FROM BioSample;")
for row in cursor:
dataset = self.getDataset(row[b'datasetId'])
bioSample = biodata.BioSample(
dataset, row[b'name'])
bioSample.populateFromRow(row)
assert bioSample.getId() == row[b'id']
dataset.addBioSample(bioSample)
def testToProtocolElement(self):
dataset = datasets.Dataset('dataset1')
# Write out a valid input
validBioSample = protocol.BioSample(name="test",
created="2016-05-19T21:00:19Z",
updated="2016-05-19T21:00:19Z")
validBioSample.info['test'].values.add().string_value = 'test-info'
# pass through protocol creation
bioSample = bioMetadata.BioSample(dataset, "test")
bioSample.populateFromJson(protocol.toJson(validBioSample))
gaBioSample = bioSample.toProtocolElement()
# Verify elements exist
self.assertEqual(gaBioSample.created, validBioSample.created)
self.assertEqual(gaBioSample.updated, validBioSample.updated)
# Invalid input
invalidBioSample = '{"bad:", "json"}'
bioSample = bioMetadata.Individual(dataset, "test")
# Should fail
self.assertRaises(exceptions.InvalidJsonException,
bioSample.populateFromJson, invalidBioSample)
def run(self):
if not os.path.exists(self.outputDirectory):
os.makedirs(self.outputDirectory)
self.repo.open("w")
self.repo.initialise()
referenceFileName = "ref_brca1.fa"
inputRef = os.path.join(
self.inputDirectory, referenceFileName)
outputRef = os.path.join(
self.outputDirectory, referenceFileName)
shutil.copy(inputRef, outputRef)
fastaFilePath = os.path.join(
self.outputDirectory,
referenceFileName + '.gz')
pysam.tabix_compress(
outputRef, fastaFilePath)
with open(
os.path.join(
self.inputDirectory, "ref_brca1.json")) as refMetadataFile:
refMetadata = json.load(refMetadataFile)
with open(
os.path.join(
self.inputDirectory,
"referenceset_hg37.json")) as refMetadataFile:
refSetMetadata = json.load(refMetadataFile)
referenceSet = references.HtslibReferenceSet(
refSetMetadata['assemblyId'])
referenceSet.populateFromFile(fastaFilePath)
referenceSet.setAssemblyId(refSetMetadata['assemblyId'])
referenceSet.setDescription(refSetMetadata['description'])
referenceSet.setNcbiTaxonId(refSetMetadata['ncbiTaxonId'])
referenceSet.setIsDerived(refSetMetadata['isDerived'])
referenceSet.setSourceUri(refSetMetadata['sourceUri'])
referenceSet.setSourceAccessions(refSetMetadata['sourceAccessions'])
for reference in referenceSet.getReferences():
reference.setNcbiTaxonId(refMetadata['ncbiTaxonId'])
reference.setSourceAccessions(
refMetadata['sourceAccessions'])
self.repo.insertReferenceSet(referenceSet)
dataset = datasets.Dataset("brca1")
self.repo.insertDataset(dataset)
hg00096Individual = biodata.Individual(dataset, "HG00096")
with open(
os.path.join(
self.inputDirectory,
"individual_HG00096.json")) as jsonString:
hg00096Individual.populateFromJson(jsonString.read())
self.repo.insertIndividual(hg00096Individual)
hg00096BioSample = biodata.BioSample(dataset, "HG00096")
with open(
os.path.join(
self.inputDirectory,
"bioSample_HG00096.json")) as jsonString:
hg00096BioSample.populateFromJson(jsonString.read())
hg00096BioSample.setIndividualId(hg00096Individual.getId())
self.repo.insertBioSample(hg00096BioSample)
hg00099Individual = biodata.Individual(dataset, "HG00099")
with open(
os.path.join(
self.inputDirectory,
"individual_HG00099.json")) as jsonString:
hg00099Individual.populateFromJson(jsonString.read())
self.repo.insertIndividual(hg00099Individual)
hg00099BioSample = biodata.BioSample(dataset, "HG00099")
with open(
os.path.join(
self.inputDirectory,
"bioSample_HG00099.json")) as jsonString:
hg00099BioSample.populateFromJson(jsonString.read())
hg00099BioSample.setIndividualId(hg00099Individual.getId())
self.repo.insertBioSample(hg00099BioSample)
hg00101Individual = biodata.Individual(dataset, "HG00101")
with open(
os.path.join(
self.inputDirectory,
"individual_HG00101.json")) as jsonString:
hg00101Individual.populateFromJson(jsonString.read())
self.repo.insertIndividual(hg00101Individual)
hg00101BioSample = biodata.BioSample(dataset, "HG00101")
with open(
os.path.join(
self.inputDirectory,
"bioSample_HG00101.json")) as jsonString:
hg00101BioSample.populateFromJson(jsonString.read())
hg00101BioSample.setIndividualId(hg00101Individual.getId())
self.repo.insertBioSample(hg00101BioSample)
readFiles = [
"brca1_HG00096.sam",
"brca1_HG00099.sam",
"brca1_HG00101.sam"]
for readFile in readFiles:
name = readFile.split('_')[1].split('.')[0]
readSrc = pysam.AlignmentFile(
os.path.join(self.inputDirectory, readFile), "r")
readDest = pysam.AlignmentFile(
os.path.join(
self.outputDirectory,
name + ".bam"),
"wb", header=readSrc.header)
destFilePath = readDest.filename
for readData in readSrc:
readDest.write(readData)
readDest.close()
readSrc.close()
pysam.index(destFilePath)
readGroupSet = reads.HtslibReadGroupSet(dataset, name)
readGroupSet.populateFromFile(destFilePath, destFilePath + ".bai")
readGroupSet.setReferenceSet(referenceSet)
bioSamples = [hg00096BioSample, hg00099BioSample, hg00101BioSample]
for readGroup in readGroupSet.getReadGroups():
for bioSample in bioSamples:
if bioSample.getLocalId() == readGroup.getSampleName():
readGroup.setBioSampleId(bioSample.getId())
self.repo.insertReadGroupSet(readGroupSet)
ontologyMapFileName = "so-xp-simple.obo"
inputOntologyMap = os.path.join(
self.inputDirectory, ontologyMapFileName)
outputOntologyMap = os.path.join(
self.outputDirectory, ontologyMapFileName)
shutil.copy(inputOntologyMap, outputOntologyMap)
sequenceOntology = ontologies.Ontology("so-xp-simple")
sequenceOntology.populateFromFile(outputOntologyMap)
sequenceOntology._id = "so-xp-simple"
self.repo.insertOntology(sequenceOntology)
self.repo.addOntology(sequenceOntology)
vcfFiles = [
"brca1_1kgPhase3_variants.vcf",
"brca1_WASH7P_annotation.vcf",
"brca1_OR4F_annotation.vcf"]
for vcfFile in vcfFiles:
self.addVariantSet(
vcfFile,
dataset,
referenceSet,
sequenceOntology,
bioSamples)
seqAnnFile = "brca1_gencodev19.gff3"
seqAnnSrc = os.path.join(self.inputDirectory, seqAnnFile)
seqAnnDest = os.path.join(self.outputDirectory, "gencodev19.db")
dbgen = generate_gff3_db.Gff32Db(seqAnnSrc, seqAnnDest)
dbgen.run()
gencode = sequenceAnnotations.Gff3DbFeatureSet(dataset, "gencodev19")
gencode.setOntology(sequenceOntology)
gencode.populateFromFile(seqAnnDest)
gencode.setReferenceSet(referenceSet)
self.repo.insertFeatureSet(gencode)
self.repo.commit()
print("Done converting compliance data.", file=sys.stderr)
def run(self):
if not os.path.exists(self.outputDirectory):
os.makedirs(self.outputDirectory)
self.repo.open("w")
self.repo.initialise()
referenceFileName = "ref_brca1.fa"
inputRef = os.path.join(self.inputDirectory, referenceFileName)
outputRef = os.path.join(self.outputDirectory, referenceFileName)
shutil.copy(inputRef, outputRef)
fastaFilePath = os.path.join(self.outputDirectory,
referenceFileName + '.gz')
pysam.tabix_compress(outputRef, fastaFilePath)
with open(os.path.join(self.inputDirectory,
"ref_brca1.json")) as refMetadataFile:
refMetadata = json.load(refMetadataFile)
with open(os.path.join(self.inputDirectory,
"referenceset_hg37.json")) as refMetadataFile:
refSetMetadata = json.load(refMetadataFile)
referenceSet = references.HtslibReferenceSet(
refSetMetadata['assemblyId'])
referenceSet.populateFromFile(fastaFilePath)
referenceSet.setAssemblyId(refSetMetadata['assemblyId'])
referenceSet.setDescription(refSetMetadata['description'])
referenceSet.setNcbiTaxonId(refSetMetadata['ncbiTaxonId'])
referenceSet.setIsDerived(refSetMetadata['isDerived'])
referenceSet.setSourceUri(refSetMetadata['sourceUri'])
referenceSet.setSourceAccessions(refSetMetadata['sourceAccessions'])
for reference in referenceSet.getReferences():
reference.setNcbiTaxonId(refMetadata['ncbiTaxonId'])
reference.setSourceAccessions(refMetadata['sourceAccessions'])
self.repo.insertReferenceSet(referenceSet)
dataset = datasets.Dataset("brca1")
# Some info is set, it isn't important what
dataset.setInfo({"version": ga4gh.__version__})
self.repo.insertDataset(dataset)
hg00096Individual = biodata.Individual(dataset, "HG00096")
with open(os.path.join(self.inputDirectory,
"individual_HG00096.json")) as jsonString:
hg00096Individual.populateFromJson(jsonString.read())
self.repo.insertIndividual(hg00096Individual)
hg00096BioSample = biodata.BioSample(dataset, "HG00096")
with open(os.path.join(self.inputDirectory,
"bioSample_HG00096.json")) as jsonString:
hg00096BioSample.populateFromJson(jsonString.read())
hg00096BioSample.setIndividualId(hg00096Individual.getId())
self.repo.insertBioSample(hg00096BioSample)
hg00099Individual = biodata.Individual(dataset, "HG00099")
with open(os.path.join(self.inputDirectory,
"individual_HG00099.json")) as jsonString:
hg00099Individual.populateFromJson(jsonString.read())
self.repo.insertIndividual(hg00099Individual)
hg00099BioSample = biodata.BioSample(dataset, "HG00099")
with open(os.path.join(self.inputDirectory,
"bioSample_HG00099.json")) as jsonString:
hg00099BioSample.populateFromJson(jsonString.read())
hg00099BioSample.setIndividualId(hg00099Individual.getId())
self.repo.insertBioSample(hg00099BioSample)
hg00101Individual = biodata.Individual(dataset, "HG00101")
with open(os.path.join(self.inputDirectory,
"individual_HG00101.json")) as jsonString:
hg00101Individual.populateFromJson(jsonString.read())
self.repo.insertIndividual(hg00101Individual)
hg00101BioSample = biodata.BioSample(dataset, "HG00101")
with open(os.path.join(self.inputDirectory,
"bioSample_HG00101.json")) as jsonString:
hg00101BioSample.populateFromJson(jsonString.read())
hg00101BioSample.setIndividualId(hg00101Individual.getId())
self.repo.insertBioSample(hg00101BioSample)
readFiles = [
"brca1_HG00096.sam", "brca1_HG00099.sam", "brca1_HG00101.sam"
]
for readFile in readFiles:
name = readFile.split('_')[1].split('.')[0]
readSrc = pysam.AlignmentFile(
os.path.join(self.inputDirectory, readFile), "r")
readDest = pysam.AlignmentFile(os.path.join(
self.outputDirectory, name + ".bam"),
"wb",
header=readSrc.header)
destFilePath = readDest.filename
for readData in readSrc:
readDest.write(readData)
readDest.close()
readSrc.close()
pysam.index(destFilePath)
readGroupSet = reads.HtslibReadGroupSet(dataset, name)
readGroupSet.populateFromFile(destFilePath, destFilePath + ".bai")
readGroupSet.setReferenceSet(referenceSet)
dataset.addReadGroupSet(readGroupSet)
bioSamples = [hg00096BioSample, hg00099BioSample, hg00101BioSample]
for readGroup in readGroupSet.getReadGroups():
for bioSample in bioSamples:
if bioSample.getLocalId() == readGroup.getSampleName():
readGroup.setBioSampleId(bioSample.getId())
self.repo.insertReadGroupSet(readGroupSet)
ontologyMapFileName = "so-xp-simple.obo"
inputOntologyMap = os.path.join(self.inputDirectory,
ontologyMapFileName)
outputOntologyMap = os.path.join(self.outputDirectory,
ontologyMapFileName)
shutil.copy(inputOntologyMap, outputOntologyMap)
sequenceOntology = ontologies.Ontology("so-xp-simple")
sequenceOntology.populateFromFile(outputOntologyMap)
sequenceOntology._id = "so-xp-simple"
self.repo.insertOntology(sequenceOntology)
self.repo.addOntology(sequenceOntology)
vcfFiles = [
"brca1_1kgPhase3_variants.vcf", "brca1_WASH7P_annotation.vcf",
"brca1_OR4F_annotation.vcf"
]
for vcfFile in vcfFiles:
self.addVariantSet(vcfFile, dataset, referenceSet,
sequenceOntology, bioSamples)
# Sequence annotations
seqAnnFile = "brca1_gencodev19.gff3"
seqAnnSrc = os.path.join(self.inputDirectory, seqAnnFile)
seqAnnDest = os.path.join(self.outputDirectory, "gencodev19.db")
dbgen = generate_gff3_db.Gff32Db(seqAnnSrc, seqAnnDest)
dbgen.run()
gencode = sequence_annotations.Gff3DbFeatureSet(dataset, "gencodev19")
gencode.setOntology(sequenceOntology)
gencode.populateFromFile(seqAnnDest)
gencode.setReferenceSet(referenceSet)
self.repo.insertFeatureSet(gencode)
# add g2p featureSet
g2pPath = os.path.join(self.inputDirectory, "cgd")
# copy all files input directory to output path
outputG2PPath = os.path.join(self.outputDirectory, "cgd")
os.makedirs(outputG2PPath)
for filename in glob.glob(os.path.join(g2pPath, '*.*')):
shutil.copy(filename, outputG2PPath)
featuresetG2P = g2p_featureset.PhenotypeAssociationFeatureSet(
dataset, outputG2PPath)
featuresetG2P.setOntology(sequenceOntology)
featuresetG2P.setReferenceSet(referenceSet)
featuresetG2P.populateFromFile(outputG2PPath)
self.repo.insertFeatureSet(featuresetG2P)
# add g2p phenotypeAssociationSet
phenotypeAssociationSet = g2p_associationset\
.RdfPhenotypeAssociationSet(dataset, "cgd", outputG2PPath)
self.repo.insertPhenotypeAssociationSet(phenotypeAssociationSet)
self.repo.commit()
dataset.addFeatureSet(gencode)
# RNA Quantification
rnaDbName = os.path.join(self.outputDirectory, "rnaseq.db")
store = rnaseq2ga.RnaSqliteStore(rnaDbName)
store.createTables()
rnaseq2ga.rnaseq2ga(self.inputDirectory + "/rna_brca1.tsv",
rnaDbName,
"rna_brca1.tsv",
"rsem",
featureType="transcript",
readGroupSetNames="HG00096",
featureSetNames="gencodev19",
dataset=dataset)
rnaQuantificationSet = rna_quantification.SqliteRnaQuantificationSet(
dataset, "rnaseq")
rnaQuantificationSet.setReferenceSet(referenceSet)
rnaQuantificationSet.populateFromFile(rnaDbName)
self.repo.insertRnaQuantificationSet(rnaQuantificationSet)
self.repo.commit()