def test_effect_tx_mode(self):
gafDatasource = TestUtils.createTranscriptProviderDatasource(
self.config)
gafDatasource.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
# Canonical mutation was Intron
m = MutationData()
m.chr = '2'
m.start = '219137340'
m.end = '219137340'
m.ref_allele = 'G'
m.alt_allele = 'T'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['gene'] == "PNKD")
self.assertTrue(m['variant_classification'] == "Missense_Mutation")
gafDatasource.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
m = MutationData()
m.chr = '2'
m.start = '219137340'
m.end = '219137340'
m.ref_allele = 'G'
m.alt_allele = 'T'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['gene'] == "PNKD")
self.assertTrue(
m['variant_classification'] == "Intron",
"Canonical no longer is Intron. This test is no longer valid. This failure can come up when changing the GAF datasource."
)
def test_effect_tx_mode(self):
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
gafDatasource.set_tx_mode(TranscriptProvider.TX_MODE_BEST_EFFECT)
# Canonical mutation was Intron
m = MutationData()
m.chr = '2'
m.start = '219137340'
m.end = '219137340'
m.ref_allele = 'G'
m.alt_allele = 'T'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['gene'] == "PNKD")
self.assertTrue(m['variant_classification'] == "Missense_Mutation")
gafDatasource.set_tx_mode(TranscriptProvider.TX_MODE_CANONICAL)
m = MutationData()
m.chr = '2'
m.start = '219137340'
m.end = '219137340'
m.ref_allele = 'G'
m.alt_allele = 'T'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['gene'] == "PNKD")
self.assertTrue(m['variant_classification'] == "Intron", "Canonical no longer is Intron. This test is no longer valid. This failure can come up when changing the GAF datasource.")
def testRealWorld(self):
"""Test that the full COSMIC datasource can retrieve entries by both gp and gpp."""
gafDS = TestUtils.createTranscriptProviderDatasource(self.config)
cosmicDS = TestUtils.createCosmicDatasource(self.config)
# These values are not taken from a real world scenario, but are cooked for this test.
m = MutationData()
m.chr = '1'
m.start = '12941796'
m.end = '12941796'
m.ref_allele = "G"
m.alt_allele = "T"
m = gafDS.annotate_mutation(m)
m = cosmicDS.annotate_mutation(m)
self.assertTrue(m['COSMIC_n_overlapping_mutations'] == '0')
#1 150483621 150483621
m = MutationData()
m.chr = '1'
m.start = '150483621'
m.end = '150483621'
m.ref_allele = "G"
m.alt_allele = "T"
m = gafDS.annotate_mutation(m)
m = cosmicDS.annotate_mutation(m)
def test_denovo(self):
"""GAF de novo test """
gafDatasource = TestUtils.createTranscriptProviderDatasource(
self.config)
m = MutationData()
m.start = str(22221735)
m.end = str(22221737)
m.chr = "22"
m.ref_allele = ''
m.alt_allele = 'CAT'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(
m['variant_classification'] == 'De_novo_Start_OutOfFrame')
m = MutationData()
m.start = str(22221735)
m.end = str(22221740)
m.chr = "22"
m.ref_allele = ''
m.alt_allele = 'AACATAA'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(
m['variant_classification'] == 'De_novo_Start_OutOfFrame')
m = MutationData()
m.start = str(22221735)
m.end = str(22221739)
m.chr = "22"
m.ref_allele = ''
m.alt_allele = 'ACATAA'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['variant_classification'] == 'De_novo_Start_InFrame')
def testMulticoreAnnotate(self):
"""Test a (too) simple annotating exercise from GAF on 2 cores"""
gafDatasource = TestUtils.createGafDatasourceProxy(self.config)
# Test pickling
dump(gafDatasource, file('out/testGAFPickle.pkl', 'w'))
m1 = MutationData()
m1.chr = '3'
m1.start = '178866811'
m1.end = '178866811'
m1.ref_allele = "A"
m1.alt_allele = "C"
m1.build = "hg19"
m2 = MutationData()
m2.chr = '3'
m2.start = '178866812'
m2.end = '178866812'
m2.ref_allele = "A"
m2.alt_allele = "C"
m2.build = "hg19"
p = LoggingPool(processes=2)
result = p.map(annotate_mutation_global, [(gafDatasource, m1),
(gafDatasource, m2)])
p.close()
p.join()
for r in result:
self.assertTrue("transcript_id" in r.keys())
self.assertTrue("gene" in r.keys())
self.assertTrue(r["gene"] == "PIK3CA")
self.assertTrue(result[0].start != result[1].start)
def testRealWorld(self):
"""Test that the full COSMIC datasource can retrieve entries by both gp and gpp."""
gafDS = TestUtils.createTranscriptProviderDatasource(self.config)
cosmicDS = TestUtils.createCosmicDatasource(self.config)
# These values are not taken from a real world scenario, but are cooked for this test.
m = MutationData()
m.chr = '1'
m.start = '12941796'
m.end = '12941796'
m.ref_allele = "G"
m.alt_allele = "T"
m = gafDS.annotate_mutation(m)
m = cosmicDS.annotate_mutation(m)
self.assertTrue(m['COSMIC_n_overlapping_mutations'] == '0')
#1 150483621 150483621
m = MutationData()
m.chr = '1'
m.start = '150483621'
m.end = '150483621'
m.ref_allele = "G"
m.alt_allele = "T"
m = gafDS.annotate_mutation(m)
m = cosmicDS.annotate_mutation(m)
def _simple_annotate(self, is_skip_no_alts):
runSpec = RunSpecification()
runSpec.initialize(None,
None,
datasources=[],
is_skip_no_alts=is_skip_no_alts)
# Initialize the annotator with the runspec
annotator = Annotator()
annotator.initialize(runSpec)
m = MutationData()
m.chr = "1"
m.start = "12941796"
m.end = "12941796"
m.alt_allele = "G"
m.ref_allele = "T"
m.createAnnotation("alt_allele_seen", "False")
m2 = MutationData()
m2.chr = "1"
m2.start = "12941796"
m2.end = "12941796"
m2.alt_allele = "G"
m2.ref_allele = "T"
muts = [m, m2]
muts = annotator.annotate_mutations(muts)
ctr = 0
for m in muts:
ctr += 1
return ctr
def test_denovo(self):
"""GAF de novo test """
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
m = MutationData()
m.start = str(22221735)
m.end = str(22221737)
m.chr="22"
m.ref_allele = ''
m.alt_allele = 'CAT'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['variant_classification'] == 'De_novo_Start_OutOfFrame')
m = MutationData()
m.start = str(22221735)
m.end = str(22221740)
m.chr="22"
m.ref_allele = ''
m.alt_allele = 'AACATAA'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['variant_classification'] == 'De_novo_Start_OutOfFrame')
m = MutationData()
m.start = str(22221735)
m.end = str(22221739)
m.chr="22"
m.ref_allele = ''
m.alt_allele = 'ACATAA'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['variant_classification'] == 'De_novo_Start_InFrame')
def test_continuous_exons_in_segments(self):
"""Test that all exons are accounted when annotating adjacent segments that skip an exon. """
# SPECC1L 10+ 22 24734447 SPECC1L 10+ 41783674 TEF 1- 1215.0 -0.04975556624325125 hg19 CESC.TCGA.BI.A0VR.Tumor.SM.1RACM
# SPECC1L 8- 22 16282318 POTEH 2- 24730543 SPECC1L 8- 433.0 -0.00781166374668759 hg19 CESC.TCGA.BI.A0VR.Tumor.SM.1RACM
# SPECC1L-ADORA2A 22 24734447 SPECC1L 10+ 41783674 TEF 1- 1215.0 -0.04975556624325125 hg19 CESC.TCGA.BI.A0VR.Tumor.SM.1RACM
seg1 = MutationData()
seg1.chr = "22"
seg1.start = "24734447" # Just passed the exon 9 (0-based)
seg1.end = "41783674"
seg2 = MutationData()
seg2.chr = "22"
seg2.start = "16282318"
seg2.end = "24730543" # Just passed the exon 8 (0-based)
segs = [seg1, seg2]
# 'ENST00000314328.9' for GENCODE v19
chosen_tx, transcript_ds = self._get_chosen_tx_and_transcript_ds(seg1.chr, seg1.start)
result_tuple = transcript_ds._determine_exons_affected_by_start(seg1.start, chosen_tx)
self.assertTrue(result_tuple == (10, '+'))
result_tuple = transcript_ds._determine_exons_affected_by_end(seg2.end, chosen_tx)
self.assertTrue(result_tuple == (8, '-'))
def test_continuous_exons_in_segments(self):
"""Test that all exons are accounted when annotating adjacent segments that skip an exon. """
# SPECC1L 10+ 22 24734447 SPECC1L 10+ 41783674 TEF 1- 1215.0 -0.04975556624325125 hg19 CESC.TCGA.BI.A0VR.Tumor.SM.1RACM
# SPECC1L 8- 22 16282318 POTEH 2- 24730543 SPECC1L 8- 433.0 -0.00781166374668759 hg19 CESC.TCGA.BI.A0VR.Tumor.SM.1RACM
# SPECC1L-ADORA2A 22 24734447 SPECC1L 10+ 41783674 TEF 1- 1215.0 -0.04975556624325125 hg19 CESC.TCGA.BI.A0VR.Tumor.SM.1RACM
seg1 = MutationData()
seg1.chr = "22"
seg1.start = "24734447" # Just passed the exon 9 (0-based)
seg1.end = "41783674"
seg2 = MutationData()
seg2.chr = "22"
seg2.start = "16282318"
seg2.end = "24730543" # Just passed the exon 8 (0-based)
segs = [seg1, seg2]
# 'ENST00000314328.9' for GENCODE v19
chosen_tx, transcript_ds = self._get_chosen_tx_and_transcript_ds(seg1.chr, seg1.start)
result_tuple = transcript_ds._determine_exons_affected_by_start(seg1.start, chosen_tx)
self.assertTrue(result_tuple == (10, '+'))
result_tuple = transcript_ds._determine_exons_affected_by_end(seg2.end, chosen_tx)
self.assertTrue(result_tuple == (8, '-'))
def _simple_annotate(self, is_skip_no_alts):
runSpec = RunSpecification()
runSpec.initialize(None, None, datasources=[], is_skip_no_alts=is_skip_no_alts)
# Initialize the annotator with the runspec
annotator = Annotator()
annotator.initialize(runSpec)
m = MutationData()
m.chr = "1"
m.start = "12941796"
m.end = "12941796"
m.alt_allele = "G"
m.ref_allele = "T"
m.createAnnotation("alt_allele_seen", "False")
m2 = MutationData()
m2.chr = "1"
m2.start = "12941796"
m2.end = "12941796"
m2.alt_allele = "G"
m2.ref_allele = "T"
muts = [m, m2]
muts = annotator.annotate_mutations(muts)
ctr = 0
for m in muts:
ctr += 1
return ctr
def testMulticoreAnnotate(self):
"""Test a (too) simple annotating exercise from GAF on 2 cores"""
gafDatasource = TestUtils.createGafDatasourceProxy(self.config)
# Test pickling
dump(gafDatasource, file('out/testGAFPickle.pkl','w'))
m1 = MutationData()
m1.chr = '3'
m1.start = '178866811'
m1.end = '178866811'
m1.ref_allele = "A"
m1.alt_allele = "C"
m1.build = "hg19"
m2 = MutationData()
m2.chr = '3'
m2.start = '178866812'
m2.end = '178866812'
m2.ref_allele = "A"
m2.alt_allele = "C"
m2.build = "hg19"
p = LoggingPool(processes=2)
result = p.map(annotate_mutation_global, [(gafDatasource, m1), (gafDatasource, m2)])
p.close()
p.join()
for r in result:
self.assertTrue("transcript_id" in r.keys())
self.assertTrue("gene" in r.keys())
self.assertTrue(r["gene"] == "PIK3CA")
self.assertTrue(result[0].start != result[1].start)
def testAnnotateListOfMutations(self):
"""Test that we can initialize an Annotator, without an input or output and then feed mutations,
one at a time... using a runspec"""
# Locate the datasource directory and create a runspec
dbDir = self.config.get("DEFAULT", "dbDir")
ds = DatasourceFactory.createDatasources(dbDir)
runSpec = RunSpecification()
runSpec.initialize(None, None, datasources=ds)
# Initialize the annotator with the runspec
annotator = Annotator()
annotator.initialize(runSpec)
m = MutationData()
m.chr = "1"
m.start = "12941796"
m.end = "12941796"
m.alt_allele = "G"
m.ref_allele = "T"
muts = [m]
muts = annotator.annotate_mutations(muts)
m2 = muts.next()
self.assertTrue(m2.get("gene", None) is not None)
def testFlank(self):
"""Test that we can see a Flank mutation."""
#chr1:28,233,780-28,233,805 Junction is at chr1:28,233,793 & 94
#
refs = "TGGGCTCGGGCTCTCTGAAAAGAAAA"
alts = "TGGGCTCAGGCTCTCTGAAAAGAAAA"
vcs = []
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
numSpliceSites = 0
numSilent = 0
startWindow = 11042200
for s in range(startWindow, startWindow+len(refs)):
m = MutationData()
m.start = str(s)
m.end = str(s)
m.chr="1"
m.ref_allele = refs[s-startWindow]
m.alt_allele = alts[s-startWindow]
m = gafDatasource.annotate_mutation(m)
vc = m['variant_classification']
vcs.append(vc)
print vc + " " + m.start
pass
def testSpliceSiteWithinNBases(self):
"""Test that a silent mutation is changed to splice site w/in 10 bases of a splice site """
# chr21:10,998,326-10,998,346
# 10,998,336 is a splice site. (Junction between 10998335 and 336)
# AGTTCTCCTT C TGGAAAAAAG
refs = 'AGTTCTCCTTCTGGAAAAAAG'
alts = 'TCAGACTGAAAATACCCCCCT'
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
vcs = []
for s in range(10998326, 10998347):
m = MutationData()
m.start = str(s)
m.end = str(s)
m.chr = "21"
m.ref_allele = refs[s - 10998326]
m.alt_allele = alts[s - 10998326]
m = gafDatasource.annotate_mutation(m)
distanceFromSpliceSite = abs(10998336 - int(m.start))
vc = m['variant_classification']
self.assertTrue(vc != 'Silent', 'Silent mutation found when it should be a splice site.')
vcs.append(vc)
print vc + " " + m.start
self.assertTrue(all([tmp == "Splice_Site" for tmp in vcs[8:12]]), "Not all vcs within 2 bases were splice site: " + str(vcs[8:12]))
self.assertTrue(all([tmp != "Splice_Site" for tmp in vcs[0:8]]), "No splice sites should be seen: " + str(vcs[0:8]))
self.assertTrue(all([tmp != "Splice_Site" for tmp in vcs[12:20]]), "No splice sites should be seen: " + str(vcs[12:20]))
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 = MutationData()
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 test_validation_correction_valid(self):
""" Test that the validation allele fields are determined automatically when not specified by the user for a valid mutation.
"""
m = MutationData()
m.chr = "3"
m.start = "178948145"
m.end = "178948145"
m.alt_allele = "A"
m.ref_allele = "G"
m['validation_status'] = "Valid"
m['Match_Norm_Validation_Allele1'] = ""
m['Match_Norm_Validation_Allele2'] = ""
m['Tumor_Validation_Allele1'] = ""
m['Tumor_Validation_Allele2'] = ""
m['Mutation_Status'] = "Somatic"
output_filename = os.path.join("out", "test_validation_correction2.maf.tsv")
outputRenderer = TcgaMafOutputRenderer(output_filename,
configFile=os.path.join("configs", "tcgaMAF2.4_output.config"))
outputRenderer.renderMutations([m].__iter__())
tsv_reader = GenericTsvReader(output_filename)
for line_dict in tsv_reader:
self.assertTrue(line_dict['Match_Norm_Validation_Allele1'] == line_dict['Match_Norm_Validation_Allele2'], "Matched norm alleles did not match.")
self.assertTrue(line_dict['Tumor_Validation_Allele1'] == line_dict['Reference_Allele'], "Tumor validation allele 1 did not match reference for a valid validation result.")
self.assertTrue(line_dict['Tumor_Validation_Allele2'] == line_dict['Tumor_Seq_Allele2'], "Tumor validation allele 2 did not match Tumor_Seq_Allele2 for a valid validation result.")
self.assertTrue(line_dict['Match_Norm_Validation_Allele1'] == line_dict['Tumor_Validation_Allele1'], "Tumor allele 1 did not match normal alleles for a valid validation result.")
self.assertTrue(line_dict['Match_Norm_Validation_Allele1'] == line_dict['Reference_Allele'], "Norm validation alleles did not match reference (norm, reference): (%s, %s)" %(line_dict['Match_Norm_Validation_Allele1'] ,line_dict['Reference_Allele']) )
self.assertTrue("G" == line_dict['Reference_Allele'], "Reference allele should have been G, but was " + line_dict['Reference_Allele'])
self.assertTrue("A" == line_dict['Tumor_Seq_Allele2'], "Alt allele should have been A, but was " + line_dict['Tumor_Seq_Allele2'])
def testdbNSFPNoRefAltAnnotationWithExactMatch(self):
"""
"""
self.logger.info("Initializing dbNSFP")
tabixIndexedTsvDirName = os.path.join(*["testdata", "dbNSFP_chr1_chr3_100vars_exact_no_ref_alt_ds", "hg19"])
tabixIndexedTsvDatasource = DatasourceFactory.createDatasource(
os.path.join(tabixIndexedTsvDirName, "dbNSFP_chr1_chr3_100vars_exact_no_ref_alt_ds.config"),
tabixIndexedTsvDirName)
m1 = MutationData()
m1.chr = "1"
m1.start = "35140"
m1.end = "35140"
m1_annotated = tabixIndexedTsvDatasource.annotate_mutation(m1)
m1_annotation = m1_annotated.getAnnotation("dbNSFP_codonpos")
cur_annotation = Annotation(value="1|1|1", datasourceName="dbNSFP", dataType="String",
description="", tags=[TagConstants.INFO, TagConstants.NOT_SPLIT], number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation), "Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("dbNSFP_refcodon")
cur_annotation = Annotation(value="TAA|TAA|TAA", datasourceName="dbNSFP", dataType="String",
description="", tags=[TagConstants.INFO, TagConstants.NOT_SPLIT], number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation), "Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("dbNSFP_cds_strand")
cur_annotation = Annotation(value="-|-|-", datasourceName="dbNSFP", dataType="String",
description="", tags=[TagConstants.INFO, TagConstants.NOT_SPLIT], number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation), "Annotations do not match.")
def testSilentMutationGoingToSpliceSite(self):
"""Test that a silent mutation within 10 bp of a splice junction should become a splice site"""
#chr1:28,233,780-28,233,805 Junction is at chr1:28,233,793 & 94
#
refs = "TGGGCTCGGGCTCTCTGAAAAGAAAA"
alts = "TGGGCTCAGGCTCGCTGAAAAGAAAA"
vcs = []
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
numSpliceSites = 0
numSilent = 0
startWindow = 28233780
for s in range(startWindow, 28233806):
m = MutationData()
m.start = str(s)
m.end = str(s)
m.chr = "1"
m.ref_allele = refs[s - startWindow]
m.alt_allele = alts[s - startWindow]
m = gafDatasource.annotate_mutation(m)
distanceFromSpliceSite = abs(28233793 - int(m.start))
vc = m['variant_classification']
vcs.append(vc)
# self.assertTrue(vc <> 'Silent', 'Silent mutation found when it should be a splice site.')
if vc.lower() == "splice_site":
numSpliceSites += 1
if vc.lower() == "silent":
numSilent += 1
print vc + " " + m.start + " " + str(distanceFromSpliceSite)
self.assertTrue(numSpliceSites == 4, "Should have seen 4 splice site mutations, but saw: " + str(numSpliceSites))
self.assertTrue(numSilent == 11, "Should have seen 11 Silent mutations, but saw: " + str(numSilent))
def testFlank(self):
"""Test that we can see a Flank mutation."""
#chr1:28,233,780-28,233,805 Junction is at chr1:28,233,793 & 94
#
refs = "TGGGCTCGGGCTCTCTGAAAAGAAAA"
alts = "TGGGCTCAGGCTCTCTGAAAAGAAAA"
vcs = []
gafDatasource = TestUtils.createTranscriptProviderDatasource(
self.config)
numSpliceSites = 0
numSilent = 0
startWindow = 11042200
for s in range(startWindow, startWindow + len(refs)):
m = MutationData()
m.start = str(s)
m.end = str(s)
m.chr = "1"
m.ref_allele = refs[s - startWindow]
m.alt_allele = alts[s - startWindow]
m = gafDatasource.annotate_mutation(m)
vc = m['variant_classification']
vcs.append(vc)
print vc + " " + m.start
pass
def testAnnotateListOfMutations(self):
"""Test that we can initialize an Annotator, without an input or output and then feed mutations,
one at a time... using a runspec"""
# Locate the datasource directory and create a runspec
dbDir = self.config.get("DEFAULT", "dbDir")
ds = DatasourceFactory.createDatasources(dbDir)
runSpec = RunSpecification()
runSpec.initialize(None, None, datasources=ds)
# Initialize the annotator with the runspec
annotator = Annotator()
annotator.initialize(runSpec)
m = MutationData()
m.chr = "1"
m.start = "12941796"
m.end = "12941796"
m.alt_allele = "G"
m.ref_allele = "T"
muts = [m]
muts = annotator.annotate_mutations(muts)
m2 = muts.next()
self.assertTrue(m2.get("gene", None) is not None)
def testESPCoverageAnnotationWithMissingAnnotationValuesIndelAvgMatch(
self):
"""
"""
self.logger.info("Initializing ESP6500SI-V2 Coverage")
tabixIndexedTsvDirName = os.path.join(
*["testdata", "small_esp_coverage_avg_ds", "hg19"])
tabixIndexedTsvDatasource = DatasourceFactory.createDatasource(
os.path.join(tabixIndexedTsvDirName,
"small_esp_coverage_avg_ds.config"),
tabixIndexedTsvDirName)
m1 = MutationData()
m1.chr = "X"
m1.start = "100075350"
m1.end = "100075356"
m1_annotated = tabixIndexedTsvDatasource.annotate_mutation(m1)
m1_annotation = m1_annotated.getAnnotation("ESP_AvgSampleReadDepth")
cur_annotation = Annotation(
value="91.25",
datasourceName="ESP",
dataType="Float",
description="",
tags=[TagConstants.INFO, TagConstants.NOT_SPLIT],
number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation),
"Annotations do not match.")
def testESPCoverageAnnotationWithSNPAvgMatch(self):
"""
"""
self.logger.info("Initializing ESP6500SI-V2 Coverage")
tabixIndexedTsvDirName = os.path.join(*["testdata", "small_esp_coverage_avg_ds", "hg19"])
tabixIndexedTsvDatasource = DatasourceFactory.createDatasource(
os.path.join(tabixIndexedTsvDirName, "small_esp_coverage_avg_ds.config"), tabixIndexedTsvDirName)
m1 = MutationData()
m1.chr = "X"
m1.start = "100075334"
m1.end = "100075334"
m1_annotated = tabixIndexedTsvDatasource.annotate_mutation(m1)
m1_annotation = m1_annotated.getAnnotation("ESP_AvgAAsampleReadDepth")
cur_annotation = Annotation(value="75.0", datasourceName="ESP", dataType="Float",
description="", tags=[TagConstants.INFO, TagConstants.NOT_SPLIT], number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation), "Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("ESP_TotalAAsamplesCovered")
cur_annotation = Annotation(value="692.0", datasourceName="ESP", dataType="Float",
description="", tags=[TagConstants.INFO, TagConstants.NOT_SPLIT], number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation), "Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("ESP_Chromosome")
cur_annotation = Annotation(value="X", datasourceName="ESP", dataType="String",
description="", tags=[TagConstants.INFO, TagConstants.NOT_SPLIT], number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation), "Annotations do not match.")
def testPickleable(self):
"""Test that a near-empty MutationData can be pickled"""
m = MutationData()
m.chr = "2"
m.createAnnotation("fake1", "1")
m.addTagToAnnotation("fake1", "fakeTag")
import cPickle
cPickle.dump(m, open("out/testMDPickle.pkl", 'w'))
def testPickleable(self):
"""Test that a near-empty MutationData can be pickled"""
m = MutationData()
m.chr = "2"
m.createAnnotation("fake1", "1")
m.addTagToAnnotation("fake1", "fakeTag")
import cPickle
cPickle.dump(m, open("out/testMDPickle.pkl", 'w'))
def test_validation_correction(self):
""" Test that the validation allele fields are determined automatically when not specified by the user for invalid mutation.
"""
m = MutationData()
m.chr = "3"
m.start = "178948145"
m.end = "178948145"
m.alt_allele = "A"
m.ref_allele = "G"
m['validation_status'] = "Invalid"
m['Match_Norm_Validation_Allele1'] = ""
m['Match_Norm_Validation_Allele2'] = ""
m['Tumor_Validation_Allele1'] = ""
m['Tumor_Validation_Allele2'] = ""
m['Mutation_Status'] = "Somatic"
output_filename = os.path.join("out",
"test_validation_correction1.maf.tsv")
outputRenderer = TcgaMafOutputRenderer(output_filename,
configFile=os.path.join(
"configs",
"tcgaMAF2.4_output.config"))
outputRenderer.renderMutations([m].__iter__())
tsv_reader = GenericTsvReader(output_filename)
for line_dict in tsv_reader:
self.assertTrue(
line_dict['Match_Norm_Validation_Allele1'] ==
line_dict['Match_Norm_Validation_Allele2'],
"Matched norm alleles did not match.")
self.assertTrue(
line_dict['Tumor_Validation_Allele1'] ==
line_dict['Tumor_Validation_Allele2'],
"Tumor alleles did not match for an invalid validation result."
)
self.assertTrue(
line_dict['Match_Norm_Validation_Allele1'] ==
line_dict['Tumor_Validation_Allele2'],
"Tumor alleles did not match normal alleles for an invalid validation result."
)
self.assertTrue(
line_dict['Match_Norm_Validation_Allele1'] ==
line_dict['Reference_Allele'],
"Norm validation alleles did not match reference (norm, reference): (%s, %s)"
% (line_dict['Match_Norm_Validation_Allele1'],
line_dict['Reference_Allele']))
self.assertTrue(
"G" == line_dict['Reference_Allele'],
"Reference allele should have been G, but was " +
line_dict['Reference_Allele'])
self.assertTrue(
"None" == line_dict['Mutation_Status'],
"Mutation Status must be None when Validation Status is Invalid: "
+ line_dict['Mutation_Status'])
def testAKT1(self):
""" Test that this version of the GAF produces the up to date gene for a position given from a website user.
"""
m = MutationData()
m.chr = '14'
m.start = '105246407'
m.end = '105246407'
m.ref_allele = 'G'
m.alt_allele = 'A'
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['gene'] == "AKT1", "Incorrect gene found: " + m['gene'] + " If updating GAF, this may not be an error, but should be confirmed manually.")
def test_start_codon(self):
"""Test a start codon hit in a GAF datasource"""
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
m = MutationData()
m.start = str(22221729)
m.end = str(22221729)
m.chr="22"
m.ref_allele = 'A'
m.alt_allele = 'T'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['variant_classification'] == VariantClassification.MISSENSE)
def testdbNSFPAnnotationWithMissingExactMatch(self): # SNPs only
"""
"""
self.logger.info("Initializing dbNSFP")
tabixIndexedTsvDirName = os.path.join(
*["testdata", "dbNSFP_chr1_6vars_exact_ds", "hg19"])
tabixIndexedTsvDatasource = DatasourceFactory.createDatasource(
os.path.join(tabixIndexedTsvDirName,
"dbNSFP_chr1_6vars_exact_ds.config"),
tabixIndexedTsvDirName)
m1 = MutationData()
m1.chr = "1"
m1.start = "35138"
m1.end = "35138"
m1.ref_allele = "T"
m1.alt_allele = "C"
m1_annotated = tabixIndexedTsvDatasource.annotate_mutation(m1)
m1_annotation = m1_annotated.getAnnotation("dbNSFP_codonpos")
cur_annotation = Annotation(
value="",
datasourceName="dbNSFP",
dataType="Integer",
description="",
tags=[TagConstants.INFO, TagConstants.NOT_SPLIT],
number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation),
"Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("dbNSFP_refcodon")
cur_annotation = Annotation(
value="",
datasourceName="dbNSFP",
dataType="String",
description="",
tags=[TagConstants.INFO, TagConstants.NOT_SPLIT],
number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation),
"Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("dbNSFP_cds_strand")
cur_annotation = Annotation(
value="",
datasourceName="dbNSFP",
dataType="Float",
description="",
tags=[TagConstants.INFO, TagConstants.NOT_SPLIT],
number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation),
"Annotations do not match.")
def test_simple_annotate(self):
ds = self._create_test_ds("testdata/small_tsv_leveldb/dbNSFP2.4_variant.chr1_cut5000.tsv", os.path.abspath("out/test_simple_annotate_snp_only_leveldb"), ["chr", "pos(1-coor)", "pos(1-coor)", "ref", "alt"])
m = MutationData()
# 1 35138 T A
m.chr = "1"
m.start = "35138"
m.end = "35138"
m.ref_allele = "T"
m.alt_allele = "A"
m = ds.annotate_mutation(m)
self.assertTrue(m['phyloP100way_vertebrate_rankscore'] == "0.19875")
def testBasicAnnotation(self):
ds = GenericGenomicMutationDatasource('testdata/small_cosmic_2/cosmic_v65_chr18.tsv')
m = MutationData()
m.chr = '18'
m.start = '48604683'
m.end = '48604683'
m.ref_allele = 'G'
m.alt_allele = 'A'
m.createAnnotation('strand', '+')
guess = ds.annotate_mutation(m)
self.assertTrue(guess['_cosmic_muts_disease_counts'], 'Unable to annotate mutation correctly')
def testMC1R(self):
"""Test that this version of the GAF produces a MC1R, instead of TUBB gene"""
m = MutationData()
m.chr = '16'
m.start = '89985913'
m.end = '89985913'
m.ref_allele = 'G'
m.alt_allele = 'A'
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
m = gafDatasource.annotate_mutation(m)
# At some point, we would expect this to be MC1R, not TUBB3
self.assertTrue(m['gene'] == "TUBB3", "Incorrect gene found: " + m['gene'] + " If updating GAF, this may not be an error, but should be confirmed manually.")
def testESPCoverageAnnotationWithMissingIndelOverlapMatch(self):
"""
"""
self.logger.info("Initializing ESP6500SI-V2 Coverage")
tabixIndexedTsvDirName = os.path.join(
*["testdata", "small_esp_coverage_overlap_ds", "hg19"])
tabixIndexedTsvDatasource = DatasourceFactory.createDatasource(
os.path.join(tabixIndexedTsvDirName,
"small_esp_coverage_overlap_ds.config"),
tabixIndexedTsvDirName)
m1 = MutationData()
m1.chr = "X"
m1.start = "100075300"
m1.end = "100075336"
m1_annotated = tabixIndexedTsvDatasource.annotate_mutation(m1)
m1_annotation = m1_annotated.getAnnotation("ESP_AvgAAsampleReadDepth")
cur_annotation = Annotation(
value="75.0|81.0|81.0",
datasourceName="ESP",
dataType="String",
description="",
tags=[TagConstants.INFO, TagConstants.NOT_SPLIT],
number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation),
"Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("ESP_TotalAAsamplesCovered")
cur_annotation = Annotation(
value="692|692|692",
datasourceName="ESP",
dataType="String",
description="",
tags=[TagConstants.INFO, TagConstants.NOT_SPLIT],
number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation),
"Annotations do not match.")
m1_annotation = m1_annotated.getAnnotation("ESP_Chromosome")
cur_annotation = Annotation(
value="X|X|X",
datasourceName="ESP",
dataType="String",
description="",
tags=[TagConstants.INFO, TagConstants.NOT_SPLIT],
number=None)
self.assertTrue(m1_annotation.isEqual(cur_annotation),
"Annotations do not match.")
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 = MutationData()
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 = MutationData()
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 testEmptyAnswer(self):
''' The Reference Datasource should return a blank result if the chromosome is not found.
Note: A log entry should also be written, but this is not tested. '''
self.logger.info("Please ignore the next logging warning: testdata/reference_ds/chrTHIS_DOES_NOT_EXIST.txt not found. Please add it.")
ds = ReferenceDatasource('testdata/reference_ds')
m = MutationData()
m.chr = "THIS_DOES_NOT_EXIST"
m.start = "11"
m.end = "11"
groundTruth = ""
# remember that the annotate_mutation returns a generator, so we use an iterator
guess = ds.annotate_mutation(m)
self.assertTrue(guess['ref_context'] == groundTruth, "ref_context was not populated properly -- should be blank: " + str(guess['ref_context']))
def testBasicCosmicInit(self):
""" Very simple test that will create a datasource from a sample datasource directory.
The directory conforms to the standard datasource structure, including placement of the config file.
"""
ds = DatasourceFactory.createDatasource('testdata/small_cosmic/small_cosmic.config', "testdata/small_cosmic")
m = MutationData()
m.chr = 19
m.start = 58858921
m.end = 58858921
m = ds.annotate_mutation(m)
self.assertTrue(m['COSMIC_overlapping_mutation_AAs'] == 'p.P426P(1)', "Did not properly annotate mutation: " + m['COSMIC_overlapping_mutation_AAs'])
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_ds("out/small_positive_strand_")
# Now for a negative strand
m = MutationData()
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 = MutationData()
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_start_codon(self):
"""Test a start codon hit in a GAF datasource"""
gafDatasource = TestUtils.createTranscriptProviderDatasource(
self.config)
m = MutationData()
m.start = str(22221729)
m.end = str(22221729)
m.chr = "22"
m.ref_allele = 'A'
m.alt_allele = 'T'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(
m['variant_classification'] == VariantClassification.MISSENSE)
def test_simple_annotate_with_nonhuman(self):
"""Test a very simple annotation with a nonhuman genome (saccer)"""
ensembl_ds = self._create_ensembl_ds_from_saccer()
m = MutationData()
m.chr = "I"
m.start = "500"
m.end = "500"
m.ref_allele = "C"
m.alt_allele = "A"
m2 = ensembl_ds.annotate_mutation(m)
self.assertTrue(m2['annotation_transcript'] == "YAL069W")
self.assertTrue(m2['gene'] == "YAL069W")
def testMicroRNA(self):
"""Test proper annotation of miRNA
"""
#uc021qwk.1 chr12:31379258-31379277:- hsa-miR-3194-3p|? chr12:31379258-31379277:- Confidence=100
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
m = MutationData()
m.start = 31379268
m.end = 31379268
m.chr= "12"
m.alt_allele = 'G'
# This is accurate
m.ref_allele = 'A'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['gene'].lower() == "hsa-mir-3194-3p", "Wrong gene (GT: hsa-mir-3194-3p): " + m['gene'] + " -- if updating GAF, this test may fail as this result may not be appropriate.")
def test_simple_annotate_with_nonhuman(self):
"""Test a very simple annotation with a nonhuman genome (saccer)"""
ensembl_ds = self._create_ensembl_ds_from_saccer()
m = MutationData()
m.chr = "I"
m.start = "500"
m.end = "500"
m.ref_allele = "C"
m.alt_allele = "A"
m2 = ensembl_ds.annotate_mutation(m)
self.assertTrue(m2['annotation_transcript'] == "YAL069W")
self.assertTrue(m2['gene'] == "YAL069W")
def testFlank2(self):
"""Test a second real-world flank scenario"""
gafDatasource = TestUtils.createTranscriptProviderDatasource(self.config)
# 1 228646357 nearest Gene=HIST3H2A C>T
m = MutationData()
m.start = str(228646357)
m.end = str(228646357)
m.chr="1"
m.ref_allele = 'C'
m.alt_allele = 'T'
m = gafDatasource.annotate_mutation(m)
self.assertTrue(m['gene'] == "HIST3H2A", "Wrong gene (GT: HIST3H2A): " + m['gene'] + " -- if updating GAF, this test may fail as this gene may not be appropriate.")
self.assertTrue(m['variant_classification'] == "5'Flank", "Should be 5'Flank, but was " + m['variant_classification'] + " -- if updating GAF, this test may fail as this test is data specific. Also, this may fail if padding parameters are changed.")
def testBasicAnnotation(self):
ds = GenericGenomicMutationDatasource(
'testdata/small_cosmic_2/cosmic_v65_chr18.tsv')
m = MutationData()
m.chr = '18'
m.start = '48604683'
m.end = '48604683'
m.ref_allele = 'G'
m.alt_allele = 'A'
m.createAnnotation('strand', '+')
guess = ds.annotate_mutation(m)
self.assertTrue(guess['_cosmic_muts_disease_counts'],
'Unable to annotate mutation correctly')
def testSimpleRendering(self):
m = MutationData()
m.chr = '1'
m.start = 1000000
m.end = 1000000
outputFilename = "out/simpleBEDTest.bed"
outputRenderer = SimpleBedOutputRenderer(outputFilename)
outputRenderer.renderMutations([m], Metadata())
fp = file(outputFilename,'r')
mOut = fp.readline().strip().split(' ')
self.assertTrue(mOut[0] == "chr1")
self.assertTrue(mOut[1] == "999999")
self.assertTrue(mOut[2] == "1000000")
fp.close()
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 = MutationData()
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 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 testSimpleGLAnnotate(self):
''' Test a simple annotation case. Make sure that the ref_context and gc_content annotations are correct. '''
ds = ReferenceDatasource('testdata/reference_ds', windowSizeGCContent=5)
m = MutationData()
m.chr = "GL000211.1"
m.start = "11"
m.end = "11"
groundTruth = "gaattctttttcaagtaagtc"
guess = ds.annotate_mutation(m)
self.assertTrue(guess['ref_context'] == groundTruth, "ref_context was not populated properly: " + str(m['ref_context']))
# gc_content is rounded to 3 decimal places
self.assertTrue(fabs(float(guess['gc_content']) - (float(3)/float(11))) < .001, "gc_content was not populated properly: " + str(m['gc_content']))
def testAKT1(self):
""" Test that this version of the GAF produces the up to date gene for a position given from a website user.
"""
m = MutationData()
m.chr = '14'
m.start = '105246407'
m.end = '105246407'
m.ref_allele = 'G'
m.alt_allele = 'A'
gafDatasource = TestUtils.createTranscriptProviderDatasource(
self.config)
m = gafDatasource.annotate_mutation(m)
self.assertTrue(
m['gene'] == "AKT1", "Incorrect gene found: " + m['gene'] +
" If updating GAF, this may not be an error, but should be confirmed manually."
)
def testSimpleRendering(self):
m = MutationData()
m.chr = '1'
m.start = 1000000
m.end = 1000000
outputFilename = "out/simpleBEDTest.bed"
outputRenderer = SimpleBedOutputRenderer(outputFilename)
outputRenderer.renderMutations([m], Metadata())
fp = file(outputFilename, 'r')
mOut = fp.readline().strip().split(' ')
self.assertTrue(mOut[0] == "chr1")
self.assertTrue(mOut[1] == "999999")
self.assertTrue(mOut[2] == "1000000")
fp.close()
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_ds("out/test_hgvs_annotations_")
# 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 testMC1R(self):
"""Test that this version of the GAF produces a MC1R, instead of TUBB gene"""
m = MutationData()
m.chr = '16'
m.start = '89985913'
m.end = '89985913'
m.ref_allele = 'G'
m.alt_allele = 'A'
gafDatasource = TestUtils.createTranscriptProviderDatasource(
self.config)
m = gafDatasource.annotate_mutation(m)
# At some point, we would expect this to be MC1R, not TUBB3
self.assertTrue(
m['gene'] == "TUBB3", "Incorrect gene found: " + m['gene'] +
" If updating GAF, this may not be an error, but should be confirmed manually."
)
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_ds("out/test_hgvs_annotations_no_mapping_", 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 = 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 testBasicRefInit(self):
""" Very simple test that will create a reference datasource from a sample datasource directory.
The directory conforms to the standard datasource structure, including placement of the config file.
"""
ds = DatasourceFactory.createDatasource('testdata/reference_ds/reference_ds.config', "testdata/reference_ds")
m = MutationData()
m.chr = "22"
m.start = "11"
m.end = "11"
groundTruth = "CCCAAGCTAAACCCAGGCCAC"
# remember that the annotate_mutation returns a generator, so we use an iterator
m = ds.annotate_mutation(m)
self.assertTrue(m['ref_context'] == groundTruth, "ref_context was not populated properly: " + str(m['ref_context']))
def testSimpleAnnotate(self):
''' Perform a simple test of one of the aligned chromosomes (chr22) and make sure that we get a reasonable answer.
'''
ds = ReferenceDatasource('testdata/reference_ds', windowSizeGCContent=5)
m = MutationData()
m.chr = "22"
m.start = "11"
m.end = "11"
groundTruth = "CCCAAGCTAAACCCAGGCCAC"
guess = ds.annotate_mutation(m)
self.assertTrue(guess['ref_context'] == groundTruth, "ref_context was not populated properly: " + str(guess['ref_context']))
# gc_content is rounded to 3 decimal places
self.assertTrue(fabs(float(guess['gc_content'])- (float(6)/float(11))) < .001, "gc_content was not populated properly: " + str(guess['gc_content']))
def testExtentOutOfRangeError(self):
''' If a window is specified that extends beyond the beginning or end of a file, truncate the ref_context.
Use what is left for gc_content as well.'''
ds = ReferenceDatasource('testdata/reference_ds', windowSizeRef=6, windowSizeGCContent=5)
m = MutationData()
m.chr = "22"
m.start = "4"
m.end = "4"
# "CCCAAGCTAAACCCAGGCCAC"
groundTruth = "CCCAAGCTAA"
guess = ds.annotate_mutation(m)
self.assertTrue(guess['ref_context'] == groundTruth, "ref_context was not populated properly: " + str(guess['ref_context']))
# gc_content is rounded to 3 decimal places
self.assertTrue(fabs(float(guess['gc_content']) - (float(5)/float(9))) < .001, "gc_content was not populated properly: " + str(guess['gc_content']))
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_protein_position_off_by_one(self, chrom, start, end, ref, alt, gt_prot_change):
config = TestUtils.createUnitTestConfig()
transcript_ds = TestUtils.createTranscriptProviderDatasource(config)
cc_txs_fp = file("testdata/tx_exact_uniprot_matches.txt", 'r')
cc_txs = [tx.rsplit(".", 1)[0] for tx in cc_txs_fp]
cc_txs.append("ENST00000338368") # Add a transcript that is not exactly the same, but close
cc_txs_fp.close()
transcript_ds.set_custom_canonical_txs(cc_txs)
m = MutationData()
m.chr = chrom
m.start = start
m.end = end
m.ref_allele = ref
m.alt_allele = alt
m2 = transcript_ds.annotate_mutation(m)
self.assertEqual(m2['protein_change'], gt_prot_change)
def test_simple_annotate(self):
""" Annotate a simple example.
"""
base_config_location = "testdata/ensembl/saccer/"
config_parser = ConfigUtils.createConfigParser(base_config_location + "ensembl.config")
title = config_parser.get("general", "title")
version = config_parser.get("general", "version")
src_file = config_parser.get("general", "src_file")
ensembl_ds = EnsemblTranscriptDatasource(title=title, version=version, src_file=src_file)
m = MutationData()
m.chr = "22"
m.start = "22161963"
m.end = "22161963"
m.ref_allele = "C"
m.alt_allele = "A"
m2 = ensembl_ds.annotate_mutation(m)
def test_simple_annotate(self):
""" Annotate a simple example.
"""
base_config_location = "testdata/ensembl/saccer/"
config_parser = ConfigUtils.createConfigParser(base_config_location + "ensembl.config")
title = config_parser.get("general", "title")
version = config_parser.get("general", "version")
src_file = config_parser.get("general", "src_file")
ensembl_ds = EnsemblTranscriptDatasource(title=title, version=version, src_file=src_file)
m = MutationData()
m.chr = "22"
m.start = "22161963"
m.end = "22161963"
m.ref_allele = "C"
m.alt_allele = "A"
m2 = ensembl_ds.annotate_mutation(m)
