def test_hashcode_changes_when_tx_mode_changes(self):
"""Test that a call to set_tx_mode will change the md5 hash for the datasource"""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hashcode_changes_when_tx_mode_changes_")
ds.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
dummy_seed = "dummy"
ds.set_hashcode(dummy_seed)
initial_hash = ds.get_hashcode()
self.assertTrue(initial_hash != dummy_seed)
ds.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
be_hash = ds.get_hashcode()
self.assertTrue(initial_hash != be_hash)
ds.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
test_hash = ds.get_hashcode()
self.assertTrue(test_hash == initial_hash)
new_dummy_seed = "new_dummy"
ds.set_hashcode(new_dummy_seed)
# MAke sure new_dummy changes the hash.
initial_hash2 = ds.get_hashcode()
self.assertTrue(initial_hash2 != initial_hash)
ds.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
be_hash2 = ds.get_hashcode()
self.assertTrue(initial_hash2 != be_hash2)
self.assertTrue(be_hash != be_hash2)
ds.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
test_hash = ds.get_hashcode()
self.assertTrue(test_hash == initial_hash2)
def test_canonical_tx_list(self):
"""Test that specifying the canonical list will actually change the transcript selected. """
ds = TestUtils._create_test_gencode_v19_ds(
"out/test_canonical_tx_list_")
m = MutationDataFactory.default_create()
m.chr = "22"
m.start = "22142650"
m.end = "22142650"
m.ref_allele = "T"
m.alt_allele = "A"
ds.set_custom_canonical_txs(["ENST00000544786"])
ds.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
# NOTE: tx list overrides best effect
m2 = ds.annotate_mutation(m)
self.assertTrue(
m2['annotation_transcript'].startswith("ENST00000544786"))
self.assertTrue(
m2['variant_classification'] == VariantClassification.INTRON)
ds.set_custom_canonical_txs([])
m2 = ds.annotate_mutation(m)
self.assertTrue(
m2['variant_classification'] == VariantClassification.MISSENSE)
self.assertFalse(
m2['annotation_transcript'].startswith("ENST00000544786"))
def test_check_for_missing_appris_tag(self):
"""Check that the correct value is returned for a site with no appris tag """
ds = TestUtils._create_test_gencode_v19_ds("out/appris_no_tag", )
txs = ds.get_overlapping_transcripts("16", 61556, 61556, padding=100)
self.assertTrue(len(txs) > 0)
self.assertEquals(ds._get_appris_rank(txs[0]),
TranscriptProviderUtils.NO_APPRIS_VALUE)
def test_small_positive_strand_transcript_change(self):
"""Test one location on a transcript and make sure that the transcript change rendered properly """
ds = TestUtils._create_test_gencode_v19_ds(
"out/small_positive_strand_")
# Now for a negative strand
m = MutationDataFactory.default_create()
m.chr = "22"
m.start = "22221730"
m.end = "22221730"
m.ref_allele = "T"
m.alt_allele = "G"
m2 = ds.annotate_mutation(m)
self.assertTrue(
m2['transcript_change'] == "c.1A>C",
"Incorrect transcript change: " + m2['transcript_change'])
# positive strand
m = MutationDataFactory.default_create()
m.chr = "3"
m.start = "178916614"
m.end = "178916614"
m.ref_allele = "G"
m.alt_allele = "T"
m2 = ds.annotate_mutation(m)
self.assertTrue(
m2['transcript_change'] == "c.1G>T",
"Incorrect transcript change: " + m2['transcript_change'])
def test_hashcode_changes_when_tx_mode_changes(self):
"""Test that a call to set_tx_mode will change the md5 hash for the datasource"""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hashcode_changes_when_tx_mode_changes_")
ds.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
dummy_seed = "dummy"
ds.set_hashcode(dummy_seed)
initial_hash = ds.get_hashcode()
self.assertTrue(initial_hash != dummy_seed)
ds.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
be_hash = ds.get_hashcode()
self.assertTrue(initial_hash != be_hash)
ds.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
test_hash = ds.get_hashcode()
self.assertTrue(test_hash == initial_hash)
new_dummy_seed = "new_dummy"
ds.set_hashcode(new_dummy_seed)
# MAke sure new_dummy changes the hash.
initial_hash2 = ds.get_hashcode()
self.assertTrue(initial_hash2 != initial_hash)
ds.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
be_hash2 = ds.get_hashcode()
self.assertTrue(initial_hash2 != be_hash2)
self.assertTrue(be_hash != be_hash2)
ds.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
test_hash = ds.get_hashcode()
self.assertTrue(test_hash == initial_hash2)
def test_check_for_appris_tag(self):
"""Test that a transcript with an appris tag returns the right rank"""
ds = TestUtils._create_test_gencode_v19_ds("out/appris_tag", )
txs = ds.get_overlapping_transcripts("22",
22222050,
22222050,
padding=100)
self.assertTrue(len(txs) == 1)
self.assertEquals(ds._get_appris_rank(txs[0]), 0)
def test_overlapping_gene_5flank(self):
"""Test that we can collect an overlapping gene on its 5' Flank """
ds = TestUtils._create_test_gencode_v19_ds("out/overlapping_genes_flank")
txs = ds.get_overlapping_transcripts("22", 22222050, 22222050, padding=100)
self.assertTrue( len(txs) == 1)
self.assertTrue(txs[0].get_transcript_id() == "ENST00000398822.3")
txs = ds.get_overlapping_transcripts("22", 22224920, 22224920)
self.assertTrue(len(txs) == 0)
def test_overlapping_gene_5flank(self):
"""Test that we can collect an overlapping gene on its 5' Flank """
ds = TestUtils._create_test_gencode_v19_ds("out/overlapping_genes_flank")
txs = ds.get_overlapping_transcripts("22", 22222050, 22222050, padding=100)
self.assertTrue( len(txs) == 1)
self.assertTrue(txs[0].get_transcript_id() == "ENST00000398822.3")
txs = ds.get_overlapping_transcripts("22", 22224920, 22224920)
self.assertTrue(len(txs) == 0)
def test_basic_tag_filtering(self):
"""Test several cases for the BasicTagTranscriptFilter"""
tx_filter = TranscriptFilterFactory.create_instance("basic")
ensembl_ds = TestUtils._create_test_gencode_v19_ds("out/basic_tag_filter_ensembl_ds")
tx_dict = ensembl_ds.getTranscriptDict()
tx = tx_dict["ENST00000215832.6"]
self.assertTrue(len(tx_filter.filter([tx])) == 1)
attrib_dict = tx.get_other_attributes()
attrib_dict.pop('tag', None)
self.assertTrue(len(tx_filter.filter([tx])) == 0)
def test_basic_tag_filtering(self):
"""Test several cases for the BasicTagTranscriptFilter"""
tx_filter = TranscriptFilterFactory.create_instance("basic")
ensembl_ds = TestUtils._create_test_gencode_v19_ds(
"out/basic_tag_filter_ensembl_ds")
tx_dict = ensembl_ds.getTranscriptDict()
tx = tx_dict["ENST00000215832.6"]
self.assertTrue(len(tx_filter.filter([tx])) == 1)
attrib_dict = tx.get_other_attributes()
attrib_dict.pop('tag', None)
self.assertTrue(len(tx_filter.filter([tx])) == 0)
def test_hgvs_annotations_simple_SNP(self):
"""Test that HGVS annotations appear (incl. protein change) in a mutation, so we believe that the Transcript objects are populated properly."""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hgvs_annotations_SNP_")
# Now for a negative strand
m = MutationData()
m.chr = "22"
m.start = "22221730"
m.end = "22221730"
m.ref_allele = "T"
m.alt_allele = "G"
m.build = "hg19"
m2 = ds.annotate_mutation(m)
self.assertEqual(m2.get('HGVS_genomic_change', None), 'chr22.hg19:g.22221730T>G')
self.assertEqual(m2.get('HGVS_coding_DNA_change', None), 'ENST00000215832.6:c.1A>C')
self.assertEqual(m2.get('HGVS_protein_change', None), 'ENSP00000215832:p.Met1Leu')
def test_hgvs_annotations_IGR(self):
"""Test that the HGVS annotations appear for IGR"""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hgvs_annotations_IGR_")
m = MutationData()
m.createAnnotation('variant_type', 'SNP')
m.createAnnotation('build', 'hg19')
m.createAnnotation('variant_classification', 'IGR')
m.createAnnotation('chr', '15')
m.createAnnotation('start', 30938316)
m.createAnnotation('end', 30938316)
m.createAnnotation('ref_allele', 'G')
m.createAnnotation('alt_allele', 'A')
m2 = ds.annotate_mutation(m)
self.assertEqual(m2.get('HGVS_genomic_change', None), 'chr15.hg19:g.30938316G>A')
self.assertEqual(m2.get('HGVS_coding_DNA_change', None), '')
self.assertEqual(m2.get('HGVS_protein_change', None), '')
def test_retrieve_transcript_by_gene(self):
"""Simple test of retrieve_transcript_by_gene """
gene = "MAPK1"
ds = TestUtils._create_test_gencode_v19_ds("out/test_retrieve_transcript_by_gene_")
txs = ds.retrieve_transcripts_by_gene(gene)
self.assertTrue(len(txs) > 2)
tx_ids = [tx.get_transcript_id() for tx in txs]
self.assertTrue("ENST00000398822.3" in tx_ids, "ENST00000398822.3 not in gene %s -- is the version number correct?" % gene)
self.assertTrue("ENST00000215832.6" in tx_ids, "ENST00000215832.6 not in gene %s -- is the version number correct?" % gene)
for tx in txs:
self.assertTrue(tx.get_gene() == gene)
def test_retrieve_transcript_by_gene(self):
"""Simple test of retrieve_transcript_by_gene """
gene = "MAPK1"
ds = TestUtils._create_test_gencode_v19_ds("out/test_retrieve_transcript_by_gene_")
txs = ds.retrieve_transcripts_by_gene(gene)
self.assertTrue(len(txs) > 2)
tx_ids = [tx.get_transcript_id() for tx in txs]
self.assertTrue("ENST00000398822.3" in tx_ids, "ENST00000398822.3 not in gene %s -- is the version number correct?" % gene)
self.assertTrue("ENST00000215832.6" in tx_ids, "ENST00000215832.6 not in gene %s -- is the version number correct?" % gene)
for tx in txs:
self.assertTrue(tx.get_gene() == gene)
def test_hgvs_annotations_IGR(self):
"""Test that the HGVS annotations appear for IGR"""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hgvs_annotations_IGR_")
m = MutationDataFactory.default_create()
m.createAnnotation('variant_type', 'SNP')
m.createAnnotation('build', 'hg19')
m.createAnnotation('variant_classification', 'IGR')
m.createAnnotation('chr', '15')
m.createAnnotation('start', 30938316)
m.createAnnotation('end', 30938316)
m.createAnnotation('ref_allele', 'G')
m.createAnnotation('alt_allele', 'A')
m2 = ds.annotate_mutation(m)
self.assertEqual(m2.get('HGVS_genomic_change', None), 'chr15.hg19:g.30938316G>A')
self.assertEqual(m2.get('HGVS_coding_DNA_change', None), '')
self.assertEqual(m2.get('HGVS_protein_change', None), '')
def test_hgvs_annotations_simple_SNP(self):
"""Test that HGVS annotations appear (incl. protein change) in a mutation, so we believe that the Transcript objects are populated properly."""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hgvs_annotations_SNP_")
# Now for a negative strand
m = MutationDataFactory.default_create()
m.chr = "22"
m.start = "22221730"
m.end = "22221730"
m.ref_allele = "T"
m.alt_allele = "G"
m.build = "hg19"
m2 = ds.annotate_mutation(m)
self.assertEqual(m2.get('HGVS_genomic_change', None), 'chr22.hg19:g.22221730T>G')
self.assertEqual(m2.get('HGVS_coding_DNA_change', None), 'ENST00000215832.6:c.1A>C')
self.assertEqual(m2.get('HGVS_protein_change', None), 'ENSP00000215832:p.Met1Leu')
def test_no_mapping_file(self):
"""Test that we can still create (from scratch) and instantiate a EnsemblDatasource when no protein mapping is specified (i.e. limited HGVS support)"""
"""Test that HGVS annotations appear (incl. protein change) in a mutation, so we believe that the Transcript objects are populated properly."""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hgvs_annotations_no_mapping_file_", protein_id_mapping_file=None)
# Now for a negative strand
m = MutationData()
m.chr = "22"
m.start = "22221730"
m.end = "22221730"
m.ref_allele = "T"
m.alt_allele = "G"
m.build = "hg19"
m2 = ds.annotate_mutation(m)
self.assertEqual(m2.get('HGVS_genomic_change', None), 'chr22.hg19:g.22221730T>G')
self.assertEqual(m2.get('HGVS_coding_DNA_change', None), 'ENST00000215832.6:c.1A>C')
self.assertEqual(m2.get('HGVS_protein_change', None), 'unknown_prot_seq_id:p.Met1Leu')
def test_no_mapping_file(self):
"""Test that we can still create (from scratch) and instantiate a EnsemblDatasource when no protein mapping is specified (i.e. limited HGVS support)"""
"""Test that HGVS annotations appear (incl. protein change) in a mutation, so we believe that the Transcript objects are populated properly."""
ds = TestUtils._create_test_gencode_v19_ds("out/test_hgvs_annotations_no_mapping_file_", protein_id_mapping_file=None)
# Now for a negative strand
m = MutationDataFactory.default_create()
m.chr = "22"
m.start = "22221730"
m.end = "22221730"
m.ref_allele = "T"
m.alt_allele = "G"
m.build = "hg19"
m2 = ds.annotate_mutation(m)
self.assertEqual(m2.get('HGVS_genomic_change', None), 'chr22.hg19:g.22221730T>G')
self.assertEqual(m2.get('HGVS_coding_DNA_change', None), 'ENST00000215832.6:c.1A>C')
self.assertEqual(m2.get('HGVS_protein_change', None), 'unknown_prot_seq_id:p.Met1Leu')
def test_canonical_tx_list_empty(self):
"""Test that not specifying the canonical list will do nothing."""
ds = TestUtils._create_test_gencode_v19_ds("out/test_canonical_tx_list_empty_")
m = MutationDataFactory.default_create()
m.chr = "22"
m.start = "22142650"
m.end = "22142650"
m.ref_allele = "T"
m.alt_allele = "A"
m2 = ds.annotate_mutation(m)
self.assertFalse(m2['annotation_transcript'].startswith("ENST00000544786"))
self.assertFalse(m2['variant_classification'] == VariantClassification.INTRON)
ds.set_custom_canonical_txs([])
m2 = ds.annotate_mutation(m)
self.assertTrue(m2['variant_classification'] == VariantClassification.MISSENSE)
self.assertFalse(m2['annotation_transcript'].startswith("ENST00000544786"))
def test_canonical_tx_list_empty(self):
"""Test that not specifying the canonical list will do nothing."""
ds = TestUtils._create_test_gencode_v19_ds("out/test_canonical_tx_list_")
m = MutationData()
m.chr = "22"
m.start = "22142650"
m.end = "22142650"
m.ref_allele = "T"
m.alt_allele = "A"
m2 = ds.annotate_mutation(m)
self.assertFalse(m2['annotation_transcript'].startswith("ENST00000544786"))
self.assertFalse(m2['variant_classification'] == VariantClassification.INTRON)
ds.set_custom_canonical_txs([])
m2 = ds.annotate_mutation(m)
self.assertTrue(m2['variant_classification'] == VariantClassification.MISSENSE)
self.assertFalse(m2['annotation_transcript'].startswith("ENST00000544786"))
def test_canonical_tx_list(self):
"""Test that specifying the canonical list will actually change the transcript selected. """
ds = TestUtils._create_test_gencode_v19_ds("out/test_canonical_tx_list_")
m = MutationDataFactory.default_create()
m.chr = "22"
m.start = "22142650"
m.end = "22142650"
m.ref_allele = "T"
m.alt_allele = "A"
ds.set_custom_canonical_txs(["ENST00000544786"])
ds.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
# NOTE: tx list overrides best effect
m2 = ds.annotate_mutation(m)
self.assertTrue(m2['annotation_transcript'].startswith("ENST00000544786"))
self.assertTrue(m2['variant_classification'] == VariantClassification.INTRON)
ds.set_custom_canonical_txs([])
m2 = ds.annotate_mutation(m)
self.assertTrue(m2['variant_classification'] == VariantClassification.MISSENSE)
self.assertFalse(m2['annotation_transcript'].startswith("ENST00000544786"))
def test_small_positive_strand_transcript_change(self):
"""Test one location on a transcript and make sure that the transcript change rendered properly """
ds = TestUtils._create_test_gencode_v19_ds("out/small_positive_strand_")
# Now for a negative strand
m = MutationDataFactory.default_create()
m.chr = "22"
m.start = "22221730"
m.end = "22221730"
m.ref_allele = "T"
m.alt_allele = "G"
m2 = ds.annotate_mutation(m)
self.assertTrue(m2['transcript_change'] == "c.1A>C", "Incorrect transcript change: " + m2['transcript_change'])
# positive strand
m = MutationDataFactory.default_create()
m.chr = "3"
m.start = "178916614"
m.end = "178916614"
m.ref_allele = "G"
m.alt_allele = "T"
m2 = ds.annotate_mutation(m)
self.assertTrue(m2['transcript_change'] == "c.1G>T", "Incorrect transcript change: " + m2['transcript_change'])
def test_check_for_appris_tag(self):
"""Test that a transcript with an appris tag returns the right rank"""
ds = TestUtils._create_test_gencode_v19_ds("out/appris_tag",)
txs = ds.get_overlapping_transcripts("22", 22222050, 22222050, padding=100)
self.assertTrue( len(txs) == 1)
self.assertEquals(ds._get_appris_rank(txs[0]),0)
def test_overlapping_gene(self):
"""Test that we can collect an overlapping gene """
ds = TestUtils._create_test_gencode_v19_ds("out/overlapping_genes_")
genes = ds.get_overlapping_genes("22", 22115000, 22120000)
self.assertTrue(len({"MAPK1"} - genes) == 0)
def test_overlapping_multiple_genes(self):
"""Test that we can collect multiple overlapping genes """
ds = TestUtils._create_test_gencode_v19_ds("out/overlapping_genes_multiple_")
genes = ds.get_overlapping_genes("22", 22080000, 22120000)
self.assertTrue(len({"MAPK1", "YPEL1"} - genes) ==0 )
def test_check_for_missing_appris_tag(self):
"""Check that the correct value is returned for a site with no appris tag """
ds = TestUtils._create_test_gencode_v19_ds("out/appris_no_tag",)
txs = ds.get_overlapping_transcripts("16", 61556, 61556, padding=100)
self.assertTrue( len(txs) > 0)
self.assertEquals(ds._get_appris_rank(txs[0]), TranscriptProviderUtils.NO_APPRIS_VALUE)
def test_overlapping_multiple_genes(self):
"""Test that we can collect multiple overlapping genes """
ds = TestUtils._create_test_gencode_v19_ds("out/overlapping_genes_multiple_")
genes = ds.get_overlapping_genes("22", 22080000, 22120000)
self.assertTrue(len({"MAPK1", "YPEL1"} - genes) ==0 )
def test_overlapping_gene(self):
"""Test that we can collect an overlapping gene """
ds = TestUtils._create_test_gencode_v19_ds("out/overlapping_genes_")
genes = ds.get_overlapping_genes("22", 22115000, 22120000)
self.assertTrue(len({"MAPK1"} - genes) == 0)
