def testMixedAnnotation(self):
"""Test that the COSMIC datasource can retrieve entries by both gp and gpp."""
tabixDir = "testdata/small_cosmic_with_gp_and_gpp/"
cosmicDS = Cosmic(
src_file=tabixDir + "small_cosmic_trimmed_for_sorting.txt.tbi.gz",
title="Cosmic",
version="test",
gpp_tabix_file=tabixDir +
"small_cosmic_trimmed_for_sorting.txt.tbi.byAA.sorted.tsv.gz")
# These values are not taken from a real world scenario, but are cooked for this test.
# Line 9 should get picked up genomic coords
# Lines 7,8 should get picked up by the protein position
m = MutationDataFactory.default_create()
m.createAnnotation("gene", "A2M")
m.createAnnotation("transcript_protein_position_start", "1300")
m.createAnnotation("transcript_protein_position_end", "1400")
m.chr = '12'
m.start = '9227220'
m.end = '9227230'
m = cosmicDS.annotate_mutation(m)
self.assertTrue(m['COSMIC_n_overlapping_mutations'] == '3')
self.assertTrue(
m['COSMIC_overlapping_mutation_AAs'].find('1229') != -1,
"Could not find the entry specified by genomic coords.")
self.assertTrue(
m['COSMIC_overlapping_primary_sites'] == "lung(3)",
"Did not have the correct primary sites annotation (lung(3)): " +
m['COSMIC_overlapping_primary_sites'])
def testBasicAnnotate(self):
'''Test that the COSMIC datasource can be initialized with two index files (gp and gpp) and a simple annotation performed'''
tabixDir = "testdata/small_cosmic_with_gp_and_gpp/"
cosmicDS = Cosmic(src_file=tabixDir + "small_cosmic_trimmed_for_sorting.txt.tbi.gz", title="Cosmic", version="test", gpp_tabix_file= tabixDir + "small_cosmic_trimmed_for_sorting.txt.tbi.byAA.sorted.tsv.gz")
# These values are not taken from a real world scenario, but are cooked for this test.
m = MutationDataFactory.default_create()
m.createAnnotation("gene", "EGFR")
m.createAnnotation("transcript_protein_position_start", "747")
m.createAnnotation("transcript_protein_position_end", "747")
m.chr = '7'
m.start = '55259560'
m.end = '55259560'
m = cosmicDS.annotate_mutation(m)
self.assertTrue(m['COSMIC_n_overlapping_mutations'] == '2')
def testBasicAnnotate(self):
'''Test that the COSMIC datasource can be initialized with two index files (gp and gpp) and a simple annotation performed'''
tabixDir = "testdata/small_cosmic_with_gp_and_gpp/"
cosmicDS = Cosmic(src_file=tabixDir + "small_cosmic_trimmed_for_sorting.txt.tbi.gz", title="Cosmic", version="test", gpp_tabix_file= tabixDir + "small_cosmic_trimmed_for_sorting.txt.tbi.byAA.sorted.tsv.gz")
# These values are not taken from a real world scenario, but are cooked for this test.
m = MutationDataFactory.default_create()
m.createAnnotation("gene", "EGFR")
m.createAnnotation("transcript_protein_position_start", "747")
m.createAnnotation("transcript_protein_position_end", "747")
m.chr = '7'
m.start = '55259560'
m.end = '55259560'
m = cosmicDS.annotate_mutation(m)
self.assertTrue(m['COSMIC_n_overlapping_mutations'] == '2')
def testMixedAnnotation(self):
"""Test that the COSMIC datasource can retrieve entries by both gp and gpp."""
tabixDir = "testdata/small_cosmic_with_gp_and_gpp/"
cosmicDS = Cosmic(src_file=tabixDir + "small_cosmic_trimmed_for_sorting.txt.tbi.gz", title="Cosmic", version="test", gpp_tabix_file= tabixDir + "small_cosmic_trimmed_for_sorting.txt.tbi.byAA.sorted.tsv.gz")
# These values are not taken from a real world scenario, but are cooked for this test.
# Line 9 should get picked up genomic coords
# Lines 7,8 should get picked up by the protein position
m = MutationDataFactory.default_create()
m.createAnnotation("gene", "A2M")
m.createAnnotation("transcript_protein_position_start", "1300")
m.createAnnotation("transcript_protein_position_end", "1400")
m.chr = '12'
m.start = '9227220'
m.end = '9227230'
m = cosmicDS.annotate_mutation(m)
self.assertTrue(m['COSMIC_n_overlapping_mutations'] == '3')
self.assertTrue(m['COSMIC_overlapping_mutation_AAs'].find('1229') != -1, "Could not find the entry specified by genomic coords.")
self.assertTrue(m['COSMIC_overlapping_primary_sites'] == "lung(3)", "Did not have the correct primary sites annotation (lung(3)): " + m['COSMIC_overlapping_primary_sites'])
def createDatasourceFromConfigParser(configParser, leafDir):
"""
configParser -- config parser instance from the config file in the leafdir. For information on config file format/conventions see (TODO)
leafDir -- contains the file and necessary files (post indexing and install steps) to instantiate a datasource.
"""
result = None
# Determine the type
dsType = configParser.get("general", "type")
# TODO: Replace these if statements with something a bit more robust, such as a proper dependency injection framework
filePrefix = leafDir + "/"
if dsType == "gaf":
gaf_fname = filePrefix + configParser.get('general', 'gaf_fname')
gaf_transcript_sequences_fname = filePrefix + configParser.get('general', 'gaf_transcript_seqs_fname')
result = Gaf(gaf_fname, gaf_transcript_sequences_fname, title=configParser.get("general", "title"), version=configParser.get("general", "version"), protocol=configParser.get("general", "protocol"))
elif dsType == "dbsnp":
result = dbSNP(filePrefix + configParser.get('general', 'src_file'), title=configParser.get('general', 'title'), version=configParser.get('general', 'version'))
elif dsType == "ensembl":
result = EnsemblTranscriptDatasource(filePrefix + configParser.get('general', 'src_file'),
title=configParser.get('general', 'title'),
version=configParser.get('general', 'version'),
tx_filter=configParser.get('general', 'transcript_filter'))
elif dsType == "cosmic":
result = Cosmic(src_file=filePrefix + configParser.get('general', 'src_file'), version=configParser.get('general', 'version'), gpp_tabix_file=filePrefix + configParser.get('general', 'gpp_src_file'))
elif dsType == 'ref':
if configParser.has_option('general', 'windowSizeRef'):
window_size = configParser.get('general', 'windowSizeRef')
else:
window_size = 10
result = ReferenceDatasource(filePrefix, title=configParser.get("general", "title"), version=configParser.get('general', 'version'), windowSizeRef=window_size)
elif dsType == 'gene_tsv':
result = GenericGeneDatasource(src_file=filePrefix + configParser.get('general', 'src_file'), title=configParser.get("general", "title"), version=configParser.get('general', 'version'), geneColumnName=configParser.get('general', 'gene_col'))
elif dsType == 'transcript_tsv':
result = GenericTranscriptDatasource(src_file=filePrefix + configParser.get('general', 'src_file'), title=configParser.get("general", "title"), version=configParser.get('general', 'version'), geneColumnName=configParser.get('general', 'transcript_col'))
elif dsType == 'vc_tsv':
result = GenericVariantClassificationDatasource(src_file=filePrefix + configParser.get('general', 'src_file'), title=configParser.get("general", "title"), version=configParser.get('general', 'version'), geneColumnName=configParser.get('general', 'vc_col'))
elif dsType == 'gp_tsv':
result = GenericGenomicPositionDatasource(src_file=filePrefix + configParser.get('general', 'src_file'), title=configParser.get("general", "title"), version=configParser.get('general', 'version'), gpColumnNames=configParser.get('general', 'genomic_position_cols'))
elif dsType == 'gm_tsv':
result = GenericGenomicMutationDatasource(src_file=filePrefix + configParser.get('general', 'src_file'), title=configParser.get("general", "title"), version=configParser.get('general', 'version'), gpColumnNames=configParser.get('general', 'genomic_position_cols'))
elif dsType == 'gm_tsv_reverse_complement':
result = GenericGenomicMutationDatasource(src_file=filePrefix + configParser.get('general', 'src_file'), title=configParser.get("general", "title"), version=configParser.get('general', 'version'), gpColumnNames=configParser.get('general', 'genomic_position_cols'), use_complementary_strand_alleles_for_negative_strand_transcripts=True)
elif dsType == 'gpp_tsv':
result = GenericGeneProteinPositionDatasource(src_file=filePrefix + configParser.get('general', 'src_file'),title=configParser.get("general", "title"), version=configParser.get('general', 'version'), gpColumnNames=configParser.get('general', 'gene_protein_position_cols'))
elif dsType == "transcript_to_uniprot_aa":
result = TranscriptToUniProtProteinPositionTransformingDatasource(title=configParser.get("general", "title"),
version=configParser.get('general', 'version'),
src_file="file://" + filePrefix + configParser.get('general', 'src_file'), # three slashes for sqlite
inputPositionAnnotationName=configParser.get('general', 'inputPositionAnnotationName'),
outputPositionAnnotationName=configParser.get('general','outputPositionAnnotationName'))
elif dsType == "mock_exception":
result = MockExceptionThrowingDatasource(title=configParser.get("general", "title"), version=configParser.get('general', 'version'))
elif dsType == "indexed_vcf":
result = IndexedVcfDatasource(src_file=filePrefix + configParser.get('general', 'src_file'),
title=configParser.get("general", "title"),
version=configParser.get('general', 'version'),
match_mode=configParser.get('general', 'match_mode'))
elif dsType == "indexed_tsv":
columnNames = configParser.get("general", "column_names")
columnNames = columnNames.split(",")
annotationColumnNames = configParser.get("general", "annotation_column_names")
annotationColumnNames = annotationColumnNames.split(",")
indexColumnNames = configParser.get("general", "index_column_names")
indexColumnNames = indexColumnNames.split(",")
DatasourceFactory._log_missing_column_name_msg(columnNames, annotationColumnNames)
columnDataTypes = dict()
for columnName in annotationColumnNames:
if columnName.strip() == "":
continue
columnDataTypes[columnName] = configParser.get("data_types", columnName)
result = IndexedTsvDatasource(src_file=filePrefix + configParser.get("general", "src_file"),
title=configParser.get("general", "title"),
version=configParser.get("general", "version"),
colNames=columnNames,
annotationColNames=annotationColumnNames,
indexColNames=indexColumnNames,
match_mode=configParser.get("general", "match_mode"),
colDataTypes=columnDataTypes)
elif dsType == 'bigwig':
if not NGSLIB_INSTALLED:
raise RuntimeError("Bigwig datasource found in db-dir but ngslib library not installed.")
result = BigWigDatasource(src_file=filePrefix + configParser.get('general', 'src_file'), title=configParser.get("general", "title"), version=configParser.get('general', 'version'))
else:
raise RuntimeError('Unknown datasource type: %s' % dsType)
hashcode = DatasourceFactory._retrieve_hash_code(leafDir)
result.set_hashcode(hashcode)
return result