def addReadGroupSet(self):
"""
Adds a new ReadGroupSet into this repo.
"""
self._openRepo()
dataset = self._repo.getDatasetByName(self._args.datasetName)
dataUrl = self._args.dataFile
indexFile = self._args.indexFile
parsed = urlparse.urlparse(dataUrl)
# TODO, add https support and others when they have been
# tested.
if parsed.scheme in ['http', 'ftp']:
if indexFile is None:
raise exceptions.MissingIndexException(dataUrl)
else:
if indexFile is None:
indexFile = dataUrl + ".bai"
dataUrl = self._getFilePath(self._args.dataFile,
self._args.relativePath)
indexFile = self._getFilePath(indexFile, self._args.relativePath)
name = self._args.name
if self._args.name is None:
name = getNameFromPath(dataUrl)
readGroupSet = reads.HtslibReadGroupSet(dataset, name)
readGroupSet.populateFromFile(dataUrl, indexFile)
referenceSetName = self._args.referenceSetName
if referenceSetName is None:
# Try to find a reference set name from the BAM header.
referenceSetName = readGroupSet.getBamHeaderReferenceSetName()
referenceSet = self._repo.getReferenceSetByName(referenceSetName)
readGroupSet.setReferenceSet(referenceSet)
readGroupSet.setAttributes(json.loads(self._args.attributes))
self._updateRepo(self._repo.insertReadGroupSet, readGroupSet)
def _readReadGroupSetTable(self):
for readGroupSetRecord in m.Readgroupset.select():
dataset = self.getDataset(readGroupSetRecord.datasetid.id)
readGroupSet = reads.HtslibReadGroupSet(
dataset, readGroupSetRecord.name)
referenceSet = self.getReferenceSet(
readGroupSetRecord.referencesetid.id)
readGroupSet.setReferenceSet(referenceSet)
readGroupSet.populateFromRow(readGroupSetRecord)
assert readGroupSet.getId() == readGroupSetRecord.id
# Insert the readGroupSet into the memory-based object model.
dataset.addReadGroupSet(readGroupSet)
def createRepo(self):
"""
Creates the repository for all the data we've just downloaded.
"""
repo = datarepo.SqlDataRepository(self.repoPath)
repo.open("w")
repo.initialise()
referenceSet = references.HtslibReferenceSet("GRCh37-subset")
referenceSet.populateFromFile(self.fastaFilePath)
referenceSet.setDescription("Subset of GRCh37 used for demonstration")
referenceSet.setSpeciesFromJson(
'{"id": "9606",' +
'"term": "H**o sapiens", "source_name": "NCBI"}')
for reference in referenceSet.getReferences():
reference.setSpeciesFromJson(
'{"id": "9606",' +
'"term": "H**o sapiens", "source_name": "NCBI"}')
reference.setSourceAccessions(
self.accessions[reference.getName()] + ".subset")
repo.insertReferenceSet(referenceSet)
dataset = datasets.Dataset("1kg-p3-subset")
dataset.setDescription("Sample data from 1000 Genomes phase 3")
repo.insertDataset(dataset)
variantSet = variants.HtslibVariantSet(dataset, "mvncall")
variantSet.setReferenceSet(referenceSet)
dataUrls = [vcfFile for vcfFile, _ in self.vcfFilePaths]
indexFiles = [indexFile for _, indexFile in self.vcfFilePaths]
variantSet.populateFromFile(dataUrls, indexFiles)
variantSet.checkConsistency()
repo.insertVariantSet(variantSet)
for sample, (bamFile, indexFile) in zip(self.samples,
self.bamFilePaths):
readGroupSet = reads.HtslibReadGroupSet(dataset, sample)
readGroupSet.populateFromFile(bamFile, indexFile)
readGroupSet.setReferenceSet(referenceSet)
repo.insertReadGroupSet(readGroupSet)
repo.commit()
repo.close()
self.log("Finished creating the repository; summary:\n")
repo.open("r")
repo.printSummary()
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(os.path.abspath(fastaFilePath))
referenceSet.setAssemblyId(refSetMetadata['assemblyId'])
referenceSet.setDescription(refSetMetadata['description'])
if refSetMetadata['species']:
speciesJson = json.dumps(refSetMetadata['species'])
referenceSet.setSpeciesFromJson(speciesJson) # needs a string
referenceSet.setIsDerived(refSetMetadata['isDerived'])
referenceSet.setSourceUri(refSetMetadata['sourceUri'])
referenceSet.setSourceAccessions(refSetMetadata['sourceAccessions'])
for reference in referenceSet.getReferences():
if refSetMetadata['species']:
speciesJsonStr = json.dumps(refMetadata['species'])
reference.setSpeciesFromJson(speciesJsonStr)
reference.setSourceAccessions(refMetadata['sourceAccessions'])
self.repo.insertReferenceSet(referenceSet)
dataset = datasets.Dataset("brca1")
# Some info is set, it isn't important what
dataset.setAttributes({"version": ga4gh.server.__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(
os.path.abspath(destFilePath),
os.path.abspath(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(os.path.abspath(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(os.path.abspath(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, os.path.abspath(outputG2PPath))
featuresetG2P.setOntology(sequenceOntology)
featuresetG2P.setReferenceSet(referenceSet)
featuresetG2P.populateFromFile(os.path.abspath(outputG2PPath))
self.repo.insertFeatureSet(featuresetG2P)
# add g2p phenotypeAssociationSet
phenotypeAssociationSet = \
g2p_associationset.RdfPhenotypeAssociationSet(
dataset, "cgd", os.path.abspath(outputG2PPath))
self.repo.insertPhenotypeAssociationSet(phenotypeAssociationSet)
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",
dataset=dataset,
featureSetNames="gencodev19",
biosampleId=hg00096Biosample.getId())
rnaQuantificationSet = rna_quantification.SqliteRnaQuantificationSet(
dataset, "rnaseq")
rnaQuantificationSet.setReferenceSet(referenceSet)
rnaQuantificationSet.populateFromFile(os.path.abspath(rnaDbName))
self.repo.insertRnaQuantificationSet(rnaQuantificationSet)
def getDataModelInstance(self, localId, dataPath):
readGroupSet = reads.HtslibReadGroupSet(self._dataset, localId)
readGroupSet.populateFromFile(dataPath)
readGroupSet.setPatientId("patient1")
readGroupSet.setSampleId("sample1")
return readGroupSet
def getDataModelInstance(self, localId, dataPath):
readGroupSet = reads.HtslibReadGroupSet(self._dataset, localId)
readGroupSet.populateFromFile(dataPath)
return readGroupSet