def test_01(self):
args = CLI(['-m','egm','--no-strand-specific-matching','-s','','-o','test_ComparisonTriangle.test_02.output.txt'])
experiment_a = ReadChimeraScanAbsoluteBEDPE("tests/data/test_Functional.test_01.Example_01.bedpe","test1")
experiment_b = ReadChimeraScanAbsoluteBEDPE("tests/data/test_Functional.test_01.Example_02.bedpe","test2")
self.assertEqual(len(experiment_a), 2)
self.assertEqual(len(experiment_b), 2)
genes = ParseBED("tests/data/refseq_hg19.bed","hg19",200000)
#F[a] + F[b] = MF(A, B)
mf_a = MergedFusion()
for fusion in experiment_a:
mf_a.add_fusion(fusion)
self.assertEqual(len(mf_a), 2)
#F[c] + F[d] = MF(C, D)
mf_b = MergedFusion()
for fusion in experiment_b:
mf_b.add_fusion(fusion)
self.assertEqual(len(mf_b), 2)
self.assertEqual(str(mf_b), "--- MergedFusion container of size 2 ---\nFusion 't2_431' (from dataset 'test2'): chr22:15465000(-)<-chr22:41929200(+)\n\nFusion 't2_223' (from dataset 'test2'): chr11:524500(-)<-chr11:62910000(+)\n----------------------------------------\n")
def test_01(self):
args = CLI([
'-m', 'egm', '--no-strand-specific-matching', '-s', '', '-o',
'test_ComparisonTriangle.test_02.output.txt'
])
experiment_a = ReadChimeraScanAbsoluteBEDPE(
"tests/data/test_Functional.test_01.Example_01.bedpe", "test1")
experiment_b = ReadChimeraScanAbsoluteBEDPE(
"tests/data/test_Functional.test_01.Example_02.bedpe", "test2")
self.assertEqual(len(experiment_a), 2)
self.assertEqual(len(experiment_b), 2)
genes = ParseBED("tests/data/refseq_hg19.bed", "hg19", 200000)
#F[a] + F[b] = MF(A, B)
mf_a = MergedFusion()
for fusion in experiment_a:
mf_a.add_fusion(fusion)
self.assertEqual(len(mf_a), 2)
#F[c] + F[d] = MF(C, D)
mf_b = MergedFusion()
for fusion in experiment_b:
mf_b.add_fusion(fusion)
self.assertEqual(len(mf_b), 2)
def test_03(self):
"""
Check the duplication removal - simple test; 2 identical fusions
"""
args = CLI(['-m','subset','--no-strand-specific-matching','-s',''])
fusion_1 = Fusion("chr1","chr2",15000,20000,"+","+","Experiment","",True)
fusion_2 = Fusion("chr1","chr2",15000,20000,"+","+","Experiment","",True)
experiment = FusionDetectionExperiment("Experiment_1")
experiment.add_fusion(fusion_1)
experiment.add_fusion(fusion_2)
self.assertEqual(len(experiment), 2)
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_03.bed","hg18", 200000)
experiment.annotate_genes(genes)
experiment.remove_duplicates(args)
self.assertEqual(len(experiment), 1)
for fusion in experiment:
self.assertEqual(len(fusion.annotated_genes_left), 1)
self.assertEqual(len(fusion.annotated_genes_right), 1)
def test_01(self):
args = CLI(['-m', 'subset', '--no-strand-specific-matching', '-s', ''])
experiment_a = ReadChimeraScanAbsoluteBEDPE(
"tests/data/test_CompareFusionsBySpanningGenes.TestCompareFusionsBySpanningGenes.test_01.bedpe",
"TestExperimentA")
experiment_b = ReadChimeraScanAbsoluteBEDPE(
"tests/data/test_CompareFusionsBySpanningGenes.TestCompareFusionsBySpanningGenes.test_01.bedpe",
"TestExperimentB")
self.assertEqual(len(experiment_a), 690)
self.assertEqual(len(experiment_b), 690)
genes = ParseBED(
"tests/data/test_CompareFusionsBySpanningGenes.TestCompareFusionsBySpanningGenes.test_01.bed",
"hg18", 200000)
self.assertEqual(len(genes), 47790)
experiment_a.annotate_genes(genes)
experiment_b.annotate_genes(genes)
experiment_a.remove_duplicates(args)
experiment_b.remove_duplicates(args)
overlap = CompareFusionsBySpanningGenes(experiment_a, experiment_b,
args)
overlapping_fusions = overlap.find_overlap()
self.assertLessEqual(len(overlapping_fusions[0]), 538)
self.assertLessEqual(538, len(experiment_a))
def test_08(self):
"""
Check the duplication removal - 2 fusions where one is missing
annotations -> one should be lost because it isn't gene
spanning
"""
args = CLI(['-m','subset','--no-strand-specific-matching','-s',''])
fusion_1 = Fusion("chr1","chr2",15000,20000,"+","+","Experiment","",True)
fusion_2 = Fusion("chr1","chr2",15000,30000,"+","+","Experiment","",True)
experiment = FusionDetectionExperiment("Experiment_1")
experiment.add_fusion(fusion_1)
experiment.add_fusion(fusion_2)
self.assertEqual(len(experiment), 2)
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_08.bed","hg18", 200000)
experiment.annotate_genes(genes)
experiment.remove_duplicates(args)
self.assertEqual(len(experiment), 1)
for fusion in experiment:
self.assertEqual(len(fusion.annotated_genes_left), 1)
self.assertEqual(len(fusion.annotated_genes_right), 1)
def test_05(self):
"""
Check the duplication removal - 2 fusions; one has two
overlapping left genes, the other only one, and for right is
vice versa. The 1 gene is always a subset of the others 2 genes
and must therefore be treated as identical annotations
"""
args = CLI(['-m','subset','--no-strand-specific-matching','-s',''])
fusion_1 = Fusion("chr1","chr2",15000,20050,"+","+","Experiment","",True)#(1A):(2A,2B)
fusion_2 = Fusion("chr1","chr2",15050,20000,"+","+","Experiment","",True)#(1A,1B):(2A)
experiment = FusionDetectionExperiment("Experiment_1")
experiment.add_fusion(fusion_1)
experiment.add_fusion(fusion_2)
self.assertEqual(len(experiment), 2)
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_05.bed","hg18", 200000)
experiment.annotate_genes(genes)
experiment.remove_duplicates(args)
self.assertEqual(len(experiment), 1)
for fusion in experiment:
self.assertEqual(len(fusion.annotated_genes_left), 1) # subset = (1A)
self.assertEqual(len(fusion.annotated_genes_right), 1) # subset = (2A)
def test_12(self):
"""
{a} = 1 , 4 , 5
{b} = 2 , 6
{c} = 7 , 8
{d} = 3
| f1 | f2 | f3 | f4 | f5 | f6 | f7 | f8 |
---+----+----+----+----+----+----+----+----+
f1 | * | | | a | a | | | |
---+----+----+----+----+----+----+----+----+
f2 | | * | | | | b | | |
---+----+----+----+----+----+----+----+----+
f3 | | | d | | | | | |
---+----+----+----+----+----+----+----+----+
f4 | a | | | * | a | | | |
---+----+----+----+----+----+----+----+----+
f5 | a | | | a | * | | | |
---+----+----+----+----+----+----+----+----+
f6 | | b | | | | * | | |
---+----+----+----+----+----+----+----+----+
f7 | | | | | | | * | d |
---+----+----+----+----+----+----+----+----+
f8 | | | | | | | c | * |
---+----+----+----+----+----+----+----+----+
"""
args = CLI(['-m','subset','--no-strand-specific-matching','-s',''])
fusion_1 = Fusion("chr1","chr1",15010,80040,"+","+","Experiment","",True)
fusion_2 = Fusion("chr2","chr2",15030,80030,"+","+","Experiment","",True)
fusion_3 = Fusion("chr4","chr4",15050,80070,"+","+","Experiment","",True)
fusion_4 = Fusion("chr1","chr1",15060,80010,"+","+","Experiment","",True)
fusion_5 = Fusion("chr1","chr1",15020,80050,"+","+","Experiment","",True)
fusion_6 = Fusion("chr2","chr2",15080,80080,"+","+","Experiment","",True)
fusion_7 = Fusion("chr3","chr3",15040,80020,"+","+","Experiment","",True)
fusion_8 = Fusion("chr3","chr3",15070,80060,"+","+","Experiment","",True)
experiment = FusionDetectionExperiment("Experiment_1")
experiment.add_fusion(fusion_2)
experiment.add_fusion(fusion_8)
experiment.add_fusion(fusion_6)
experiment.add_fusion(fusion_4)
experiment.add_fusion(fusion_5)
experiment.add_fusion(fusion_3)
experiment.add_fusion(fusion_1)
experiment.add_fusion(fusion_7)
self.assertEqual(len(experiment), 8)
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_12.bed","hg18", 200000)
experiment.annotate_genes(genes)
experiment.remove_duplicates(args)
self.assertEqual(len(experiment), 4)
def test_01(self):
args = CLI(['-m','subset','--no-strand-specific-matching','-s',''])
experiment = ReadChimeraScanAbsoluteBEDPE("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_01.bedpe","TestExperiment")
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_01.bed","hg18", 200000)
length_before_duplication_removal = len(experiment)
experiment.annotate_genes(genes)
experiment.remove_duplicates(args)
length_after_duplication_removal = len(experiment)
self.assertTrue(length_before_duplication_removal > length_after_duplication_removal)
def test_11(self):
"""
May not have overlap:
{1} = a,b,c
{2} = a,b,d
Break {1}:
|
Break {2}: :
| :
a: [ : : ]
b: [ | : ]
c: : [ : ]
d: [ : ]
"""
args = CLI(['-m','subset','--no-strand-specific-matching','-s',''])
fusion_1 = Fusion("chr1","chr2",15000,27500,"+","+","Experiment","",True)
fusion_2 = Fusion("chr1","chr2",15000,25500,"+","+","Experiment","",True)
experiment = FusionDetectionExperiment("Experiment_1")
experiment.add_fusion(fusion_1)
experiment.add_fusion(fusion_2)
self.assertEqual(len(experiment), 2)
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_11.bed","hg18", 200000)
experiment.annotate_genes(genes)
for fusion in experiment:
if(fusion.get_right_break_position() == 30000):
self.assertEqual(len(fusion.annotated_genes_right), 3)
self.assertTrue("NM_00002A" in [str(gene_name) for gene_name in fusion.annotated_genes_right])
self.assertTrue("NM_00002B" in [str(gene_name) for gene_name in fusion.annotated_genes_right])
self.assertTrue("NM_00002C" in [str(gene_name) for gene_name in fusion.annotated_genes_right])
if(fusion.get_right_break_position() == 25000):
self.assertEqual(len(fusion.annotated_genes_right), 3)
self.assertTrue("NM_00002A" in [str(gene_name) for gene_name in fusion.annotated_genes_right])
self.assertTrue("NM_00002B" in [str(gene_name) for gene_name in fusion.annotated_genes_right])
self.assertTrue("NM_00002D" in [str(gene_name) for gene_name in fusion.annotated_genes_right])
experiment.remove_duplicates(args)
self.assertEqual(len(experiment), 2)
def test_02(self):
experiment = ReadChimeraScanAbsoluteBEDPE("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_02.bedpe","TestExperiment")
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_02.bed","hg18", 200000)
self.assertEqual(len(experiment), 1)
experiment.annotate_genes(genes)
for fusion in experiment:
left_genes = fusion.get_annotated_genes_left(False)
right_genes = fusion.get_annotated_genes_right(False)
self.assertEqual(len(left_genes), 8)
self.assertEqual(len(right_genes), 8)
# Ensure all annotated gene names do NOT contain substring '_invalid'
self.assertEqual(min([str(gene_name).find("_invalid") for gene_name in left_genes]) , -1)
self.assertEqual(min([str(gene_name).find("_invalid") for gene_name in right_genes]) , -1)
def test_02(self):
args_a = CLI(['-m', 'subset', '-s', ''])
args_b = CLI(['-m', 'subset', '--strand-specific-matching', '-s', ''])
## First test the matches if strand-specific-matching is disabled (all 4 fusions should be identical)
experiment_a = ReadChimeraScanAbsoluteBEDPE(
"tests/data/test_CompareFusionsBySpanningGenes.TestCompareFusionsBySpanningGenes.test_02_a.bedpe",
"TestExperimentA")
experiment_b = ReadChimeraScanAbsoluteBEDPE(
"tests/data/test_CompareFusionsBySpanningGenes.TestCompareFusionsBySpanningGenes.test_02_b.bedpe",
"TestExperimentB")
self.assertEqual(len(experiment_a), 4)
self.assertEqual(len(experiment_b), 4)
genes = ParseBED(
"tests/data/test_CompareFusionsBySpanningGenes.TestCompareFusionsBySpanningGenes.test_02.bed",
"hg18", 200000)
self.assertEqual(len(genes), 8)
experiment_a.annotate_genes(genes)
experiment_b.annotate_genes(genes)
## @todo -> remove duplicates should be done separately
experiment_a.remove_duplicates(args_a)
experiment_b.remove_duplicates(args_a)
overlap = CompareFusionsBySpanningGenes(
experiment_a, experiment_b,
args_a) # No EGM, no strand-specific-matching
overlapping_fusions = overlap.find_overlap()
self.assertLessEqual(len(overlapping_fusions[0]), 4)
## Second, test the matches if strand-specific-matching is disabled (only the first fusion should be identical)
overlap = CompareFusionsBySpanningGenes(
experiment_a, experiment_b,
args_b) # No EGM, but strand-specific-matching
overlapping_fusions = overlap.find_overlap()
self.assertLessEqual(len(overlapping_fusions[0]), 1)
def test_01(self):
filename = "tests/data/test_ParseBED.TestParseBED.test_01.bed"
gene_annotation = ParseBED(filename, "test", 200000)
self.assertEqual(gene_annotation.n, 88)
def test_13(self):
"""
Crazy stuff:
{a} = 1 , 2
{b} = 1 , 3
| f1 | f2 | f3 |
---+----+----+----+
f1 | * | 1 | 2 |
---+----+----+----+
f2 | 1 | * | |
---+----+----+----+
f3 | 2 | | * |
---+----+----+----+
1 = [a,b,c]
2 = [a,c]
3 = [a,b]
>> Try in all possible orders
Break {1}:
|
Break {2}: :
| :
Break {3}: : :
: : |
a: [ : : : ]
b: : [ : : ]
c: [ : : ] :
"""
args = CLI(['-m','subset','--no-strand-specific-matching','-s',''])
fusion_1_exp_1 = Fusion("chr1","chr2",15000,70000,"+","+","Experiment_1","",True)
fusion_2_exp_1 = Fusion("chr1","chr2",15000,80000,"+","+","Experiment_1","",True)
fusion_3_exp_1 = Fusion("chr1","chr2",15000,60000,"+","+","Experiment_1","",True)
fusion_1_exp_2 = Fusion("chr1","chr2",15000,70000,"+","+","Experiment_2","",True)
fusion_2_exp_2 = Fusion("chr1","chr2",15000,80000,"+","+","Experiment_2","",True)
fusion_3_exp_2 = Fusion("chr1","chr2",15000,60000,"+","+","Experiment_2","",True)
fusion_1_exp_3 = Fusion("chr1","chr2",15000,70000,"+","+","Experiment_3","",True)
fusion_2_exp_3 = Fusion("chr1","chr2",15000,80000,"+","+","Experiment_3","",True)
fusion_3_exp_3 = Fusion("chr1","chr2",15000,60000,"+","+","Experiment_3","",True)
fusion_1_exp_4 = Fusion("chr1","chr2",15000,70000,"+","+","Experiment_4","",True)
fusion_2_exp_4 = Fusion("chr1","chr2",15000,80000,"+","+","Experiment_4","",True)
fusion_3_exp_4 = Fusion("chr1","chr2",15000,60000,"+","+","Experiment_4","",True)
fusion_1_exp_5 = Fusion("chr1","chr2",15000,70000,"+","+","Experiment_5","",True)
fusion_2_exp_5 = Fusion("chr1","chr2",15000,80000,"+","+","Experiment_5","",True)
fusion_3_exp_5 = Fusion("chr1","chr2",15000,60000,"+","+","Experiment_5","",True)
fusion_1_exp_6 = Fusion("chr1","chr2",15000,70000,"+","+","Experiment_6","",True)
fusion_2_exp_6 = Fusion("chr1","chr2",15000,80000,"+","+","Experiment_6","",True)
fusion_3_exp_6 = Fusion("chr1","chr2",15000,60000,"+","+","Experiment_6","",True)
experiment_1 = FusionDetectionExperiment("Experiment_1")
experiment_2 = FusionDetectionExperiment("Experiment_2")
experiment_3 = FusionDetectionExperiment("Experiment_3")
experiment_4 = FusionDetectionExperiment("Experiment_4")
experiment_5 = FusionDetectionExperiment("Experiment_5")
experiment_6 = FusionDetectionExperiment("Experiment_6")
experiment_1.add_fusion(fusion_1_exp_1)
experiment_1.add_fusion(fusion_2_exp_1)
experiment_1.add_fusion(fusion_3_exp_1)
experiment_2.add_fusion(fusion_1_exp_2)
experiment_2.add_fusion(fusion_3_exp_2)
experiment_2.add_fusion(fusion_2_exp_2)
experiment_3.add_fusion(fusion_2_exp_3)
experiment_3.add_fusion(fusion_1_exp_3)
experiment_3.add_fusion(fusion_3_exp_3)
experiment_4.add_fusion(fusion_2_exp_4)
experiment_4.add_fusion(fusion_3_exp_4)
experiment_4.add_fusion(fusion_1_exp_4)
experiment_5.add_fusion(fusion_3_exp_5)
experiment_5.add_fusion(fusion_1_exp_5)
experiment_5.add_fusion(fusion_2_exp_5)
experiment_6.add_fusion(fusion_3_exp_6)
experiment_6.add_fusion(fusion_2_exp_6)
experiment_6.add_fusion(fusion_1_exp_6)
self.assertEqual(len(experiment_1), 3)
self.assertEqual(len(experiment_2), 3)
self.assertEqual(len(experiment_3), 3)
self.assertEqual(len(experiment_4), 3)
self.assertEqual(len(experiment_5), 3)
self.assertEqual(len(experiment_6), 3)
genes = ParseBED("tests/data/test_FusionDetectionExperiment.TestFusionDetectionExperiment.test_13.bed","hg18", 200000)
experiment_1.annotate_genes(genes)
experiment_2.annotate_genes(genes)
experiment_3.annotate_genes(genes)
experiment_4.annotate_genes(genes)
experiment_5.annotate_genes(genes)
experiment_6.annotate_genes(genes)
experiment_1.remove_duplicates(args)
experiment_2.remove_duplicates(args)
experiment_3.remove_duplicates(args)
experiment_4.remove_duplicates(args)
experiment_5.remove_duplicates(args)
experiment_6.remove_duplicates(args)
# Removing duplicates:
#
# order 1:
# -> [A,B] <-> [A,C] = [A,B] & [A,C] & [A,B,C]
# -> [A,B] <-> [A,B,C] = [A,B*] & [A,C]
# -> [A,B*] <-> [A,C] = [A,B*] & [A,C]
# order 2:
# -> [A,B] <-> [A,B,C] = [A,B*] & [A,C]
# -> [A,B*] <-> [A,C] = [A,B*] & [A,C]
# order 3:
# -> [A,C] <-> [A,B] = [A,C] & [A,B] & [A,B,C]
# -> [A,C] <-> [A,B,C] = [A,C*] & [A,B]
# -> [A,C*] <-> [A,B] = [A,C*] & [A,B]
# order 4:
# -> [A,C] <-> [A,B,C] = [A,C*] & [A,B]
# -> [A,C*] <-> [A,B] = [A,C*] & [A,B]
# order 5:
# -> [A,B,C] <-> [A,B] = [A,B*] & [A,C]
# -> [A,B*] <-> [A,C] = [A,B*] & [A,C]
# order 6:
# -> [A,B,C] <-> [A,C] = [A,C*] & [A,B]
# -> [A,C*] <-> [A,B] = [A,C*] & [A,B]
self.assertEqual(len(experiment_1), 2)
self.assertEqual(len(experiment_2), 2)
self.assertEqual(len(experiment_3), 2)
self.assertEqual(len(experiment_4), 2)
self.assertEqual(len(experiment_5), 2)
self.assertEqual(len(experiment_6), 2)