def test_contigs_to_genes_four_hits(self):
"""Tests output for multiple blast hits"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa', 'madansi/tests/data/four_blast_hits' )
self.assertCountEqual(list(gene_detector.contigs_to_genes().keys()), ['Contig1', 'Contig2', 'Contig3'])
self.assertCountEqual(list(gene_detector.contigs_to_genes()['Contig1'].gene_objects.keys()), ['gene1', 'gene2', 'gene3'])
self.assertCountEqual(list(gene_detector.contigs_to_genes()['Contig2'].gene_objects.keys()), ['geneA'])
self.assertCountEqual(list(gene_detector.contigs_to_genes()['Contig3'].gene_objects.keys()), [])
def test_contigs_to_genes_no_hits(self):
"""Tests that the correct keys are given in the contigs object and that there is no value for each contig in this object"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/empty_file')
self.assertCountEqual(list(gene_detector.contigs_to_genes().keys()),
['Contig1', 'Contig2', 'Contig3'])
self.assertEqual(
gene_detector.contigs_to_genes()['Contig1'].gene_objects, {})
def test_finding_contig_ends_gene_degree_one(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene9', 'gene8'), ('gene8', 'gene7'),
('gene7', 'gene6'), ('gene7', 'gene4'),
('gene4', 'gene5')])
neighbouring_contigs = [[('Contig3', 'Contig2'), 1, ['gene7',
'gene4']]]
gene_detector = GeneDetector(
'madansi/tests/data/assembly_4_sequences.fa',
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
refine_contig_neighbours_object = RefineContigNeighbours(
neighbouring_contigs, filtered_graph, filtered_graph,
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file',
gene_detector, {
'Contig2': ['', '', 800],
'Contig3': ['', '', 2000]
})
self.assertDictEqual(refine_contig_neighbours_object.ends_of_contigs(),
{
'Contig3': {
'Contig2': [250, 700]
},
'Contig2': {
'Contig3': [2, 301]
}
})
def test_one_gene_on_contig(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene4', 'gene5'), ('gene5', 'gene10'),
('gene10', 'geneA')])
neighbouring_contigs = [[('Contig2', 'Contig4'), 1,
['gene5', 'gene10']]]
gene_detector = GeneDetector(
'madansi/tests/data/assembly_4_sequences.fa',
'madansi/tests/data/refine_contig_neighbours_10_blast_hits_file')
refine_contig_neighbours_object = RefineContigNeighbours(
neighbouring_contigs, filtered_graph, filtered_graph,
'madansi/tests/data/refine_contig_neighbours_10_blast_hits_file',
gene_detector, {
'Contig2': ['', '', 800],
'Contig4': ['', '', 50]
})
self.assertDictEqual(refine_contig_neighbours_object.ends_of_contigs(),
{
'Contig4': {
'Contig2': [5, 50]
},
'Contig2': {
'Contig4': [301, 2]
}
})
def test_most_occurent_contig_not_in_neighbours(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene1', 'gene2'), ('gene2', 'gene4'), ('gene4', 'gene5'), ('gene2', 'gene6'), ('gene6', 'gene7'), ('gene7', 'gene8'), ('gene4', 'gene8')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1,['gene2','gene4']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
refine_contig_neighbours = RefineContigNeighbours(neighbouring_contigs, filtered_graph,filtered_graph, 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file', gene_detector, {})
self.assertEqual(sorted(refine_contig_neighbours.refine_contig_neighbours()), [])
def test_finds_orientation_two_contigs(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'geneA'),\
('geneA', 'geneB'), ('geneB', 'gene4'), ('gene4', 'gene5')])
neighbouring_contigs = [[('Contig1', 'Contig2'), 2, ['geneA',
'geneB']]]
gene_detector = GeneDetector(
'madansi/tests/data/assembly_4_sequences.fa',
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
refine_contig_neighbours_object = RefineContigNeighbours(
neighbouring_contigs, filtered_graph, filtered_graph,
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file',
gene_detector, {
'Contig1': ['', '', 1000],
'Contig2': ['', '', 800]
})
refine_contig_neighbours_object.contigs = { 'Contig1':{'gene1':None, 'gene2':None, 'gene3':None},\
'Contig2':{'gene4':None, 'gene5':None}}
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig1',\
'gene4':'Contig2', 'gene5':'Contig2'}
expected_dict = {
'Contig1': {
'Contig2': [980, 490]
},
'Contig2': {
'Contig1': [2, 301]
}
}
refine_contig_neighbours_object.ends_of_contigs()
self.assertDictEqual(refine_contig_neighbours_object.contig_ends,
expected_dict)
def test_keep_three_cycle_two_equal_weights(self):
"""Tests that when there are three cycles with all connections having equal weight, these are preserved under refinement"""
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'gene4'),\
('gene4', 'gene5'), ('gene6', 'gene7'), ('gene7', 'gene8'),\
('gene8', 'gene1')])
neighbouring_contigs = [[('Contig1', 'Contig2'), 1, ['gene3', 'gene4']],\
[('Contig2', 'Contig3'), 1, ['gene5', 'gene6']],\
[('Contig1', 'Contig3'), 1, ['gene8', 'gene1']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
refine_contig_neighbours_object = RefineContigNeighbours(neighbouring_contigs, filtered_graph,filtered_graph, 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file', gene_detector, {})
refine_contig_neighbours_object.contigs = { 'Contig1':{'gene1':None, 'gene2':None, 'gene3':None},\
'Contig2':{'gene4':None, 'gene5':None} ,\
'Contig3':{'gene6':None, 'gene7':None, 'gene8':None}}
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig1',\
'gene4':'Contig2', 'gene5':'Contig2', 'gene6':'Contig3',\
'gene7':'Contig3', 'gene8':'Contig3'}
self.assertEqual(sorted(refine_contig_neighbours_object.refine_contig_neighbours()), sorted(neighbouring_contigs))
def test_contig_appearances(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'gene4'), ('gene4', 'gene5')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1,['gene3','gene4']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
refine_contig_neighbours = RefineContigNeighbours(neighbouring_contigs, filtered_graph,filtered_graph, 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file', gene_detector, {})
self.assertDictEqual(refine_contig_neighbours.find_contig_appearances(neighbouring_contigs[0]),{'Contig1':[3, {0:['gene3'], 1:['gene2'], 2:['gene1']}],\
'Contig2':[2, {0:['gene4'], 1:['gene5']}]})
def test_keep_all_connections_empty_list(self):
"""Tests an empty graph and initial list of neighbouring contigs"""
filtered_graph = nx.Graph()
neighbouring_contigs = []
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/empty_file' )
refine_contig_neighbours_object = RefineContigNeighbours(neighbouring_contigs,filtered_graph,filtered_graph,'madansi/tests/data/empty_file',gene_detector, {})
refine_contig_neighbours_object.genes = {}
refine_contig_neighbours_object.refine_contig_neighbours()
self.assertEqual(sorted(refine_contig_neighbours_object.refined_neighbouring_contigs), sorted(neighbouring_contigs))
def test_no_difference(self):
"""Tests the case when all the contigs in the assembly file are also given in the filtered blast hits file"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa', 'madansi/tests/data/test_blast_hits')
output_file = 'output.fa'
unused_contigs = UnusedContigs(gene_detector, output_file, 'madansi/tests/data/assembly.fa')
unused_contigs.contigs_not_in_filtered_file()
unused_contigs.add_unused_contigs_to_end()
self.assertTrue(filecmp.cmp(output_file, 'madansi/tests/data/empty_file'))
os.unlink(output_file)
def test_missing_sequence(self):
"""Tests when there is a difference between the contigs present in the assembly file and those where a gene is present"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa', 'madansi/tests/data/test_blast_hits_2')
output_file = 'output'
unused_contigs = UnusedContigs(gene_detector, output_file, 'madansi/tests/data/assembly.fa')
unused_contigs.contigs_not_in_filtered_file()
unused_contigs.add_unused_contigs_to_end()
self.assertTrue(filecmp.cmp(output_file, 'madansi/tests/data/contig3'))
os.unlink(output_file)
def test_set_expansion_one_contig_not_isolated(self):
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/test_blast_hits')
filtered_graph = nx.Graph(
nx.drawing.nx_pydot.read_dot('madansi/tests/data/test_graph.dot'))
contig_searching = ContigSearching(gene_detector, filtered_graph)
self.assertEqual(
contig_searching.set_expansion(['gene1', 'gene2', 'gene3'],
'Contig1').finished_contigs, set())
def test_loop_of_genes_between_contigs(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'geneA'), ('geneA', 'geneB'), ('geneB', 'gene4'), ('gene4', 'gene5'),\
('gene3', 'gene6'), ('gene6', 'gene7'), ('gene7', 'gene8'), ('gene8', 'gene9'), ('gene9', 'geneB')])
neighbouring_contigs = [[('Contig1', 'Contig2'),2, ['geneA', 'geneB']], [('Contig2', 'Contig3'), 1, ['geneB']], [('Contig1', 'Contig3'),1,['geneA']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
refine_contig_neighbours = RefineContigNeighbours(neighbouring_contigs, filtered_graph, filtered_graph,'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file', gene_detector, {})
expected_neighbours = [[('Contig2', 'Contig3'), 1, ['geneB']], [('Contig1', 'Contig3'),1,['geneA']]]
self.assertEqual(sorted(refine_contig_neighbours.refine_contig_neighbours()), sorted(expected_neighbours))
def test_one_contig_dummy_genes(self):
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/one_blast_hit')
filtered_graph = nx.Graph(
nx.drawing.nx_pydot.read_dot(
'madansi/tests/data/one_contig_dummy_genes.dot'))
contig_searching = ContigSearching(gene_detector, filtered_graph)
contig_searching.expand_all_contigs()
contig_neighbourhoods = contig_searching.neighbouring_contigs
self.assertEqual(contig_neighbourhoods, [])
def test_initialisation(self):
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/test_blast_hits')
filtered_graph = nx.Graph(
nx.drawing.nx_pydot.read_dot(
'madansi/tests/data/graph_3_nodes.dot'))
contig_searching = ContigSearching(gene_detector, filtered_graph)
self.assertDictEqual(contig_searching.genes_in_contig_radius, {})
self.assertCountEqual(contig_searching.neighbouring_contigs, [])
self.assertEqual(contig_searching.finished_contigs, set())
def test_find_contig_orientation(self):
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/four_blast_hits')
contig_orientation_object = ContigOrientation(nx.Graph(),
gene_detector)
self.assertEqual(
contig_orientation_object.find_contig_orientation('Contig1'), -1)
self.assertEqual(
contig_orientation_object.find_contig_orientation('Contig2'), 1)
def test_expand_all_contigs_three_contigs(self):
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/test_blast_hits')
filtered_graph = nx.Graph(
nx.drawing.nx_pydot.read_dot('madansi/tests/data/test_graph.dot'))
contig_searching = ContigSearching(gene_detector, filtered_graph)
contig_searching.expand_all_contigs()
contig_neighbourhoods = contig_searching.neighbouring_contigs
self.assertTrue(sorted(contig_neighbourhoods) == sorted([ [sorted(('Contig1', 'Contig2')),1,sorted(['gene1', 'gene4', 'gene3', 'gene5'])] ,\
[sorted(('Contig1', 'Contig3')),1,sorted(['gene2', 'gene6'])] ]))
def test_compare_graphs(self):
gene_detector = GeneDetector('madansi/tests/data/assembly_7_sequences.fa', 'madansi/tests/data/test_blast_hits_2')
output_file = 'output'
unused_contigs = UnusedContigs(gene_detector, output_file, 'madansi/tests/data/assembly_7_sequences.fa')
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('Contig1', 'Contig2'), ('Contig2', 'Contig3'), ('Contig3', 'Contig4')])
filtered_graph.add_node('Contig7')
unused_contigs.contigs_not_in_filtered_graph(filtered_graph)
unused_contigs.add_unused_contigs_to_end()
self.assertTrue(filecmp.cmp(output_file, 'madansi/tests/data/test_unused_contigs_in_graph'))
os.unlink(output_file)
def test_single_gene_from_a_contig(self):
"""Tests when there are two contigs present, one with a single gene on it when other genes from the same contig are elsewhere in the graph"""
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene1', 'gene2'), ('gene2', 'gene3'), ('gene4', 'gene5')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1,['gene2', 'gene3']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/refine_contig_neighbours_5_blast_hits_file' )
refine_contig_neighbours_object = RefineContigNeighbours(neighbouring_contigs, filtered_graph,filtered_graph, 'madansi/tests/data/refine_contig_neighbours_5_blast_hits_file',gene_detector, {})
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig2',\
'gene4':'Contig2', 'gene5':'Contig2'}
refine_contig_neighbours_object.refine_contig_neighbours()
self.assertEqual(sorted(refine_contig_neighbours_object.refined_neighbouring_contigs), [])
def test_keep_one_connection(self):
"""Tests that one connection in neighbouring contigs is kept under refinement"""
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene5'), ('gene5', 'gene6'),\
('gene6', 'gene3'), ('gene3', 'gene4')])
neighbouring_contigs = [[('Contig1', 'Contig2'), 2, ['gene5', 'gene6']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/empty_file' )
refine_contig_neighbours_object = RefineContigNeighbours(neighbouring_contigs,filtered_graph,filtered_graph,'madansi/tests/data/empty_file',gene_detector, {})
refine_contig_neighbours_object.genes = {'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig2', 'gene4':'Contig2'}
refine_contig_neighbours_object.refine_contig_neighbours()
self.assertEqual(sorted(refine_contig_neighbours_object.refined_neighbouring_contigs), sorted(neighbouring_contigs))
def test_expand_all_contigs_two_neighbouring_contigs(self):
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/four_blast_hits')
filtered_graph = nx.Graph(
nx.drawing.nx_pydot.read_dot(
'madansi/tests/data/graph_4_nodes.dot'))
contig_searching = ContigSearching(gene_detector, filtered_graph)
contig_searching.expand_all_contigs()
contig_neighbourhoods = contig_searching.neighbouring_contigs
self.assertTrue(
contig_neighbourhoods ==
[[sorted(('Contig1', 'Contig2')), 1,
sorted(['geneA', 'gene1'])]])
def test_expand_all_contigs_three_contigs_multiple_iterations(self):
gene_detector = GeneDetector(
'madansi/tests/data/assembly.fa',
'madansi/tests/data/test_blast_hits_three_contigs')
filtered_graph = nx.Graph(
nx.drawing.nx_pydot.read_dot(
'madansi/tests/data/three_contigs_separated.dot'))
contig_searching = ContigSearching(gene_detector, filtered_graph)
contig_searching.expand_all_contigs()
contig_neighbourhoods = contig_searching.neighbouring_contigs
self.assertTrue(sorted(contig_neighbourhoods) == sorted([[ sorted(('Contig1', 'Contig2')),2, sorted(['geneA', 'geneD'])],\
[ sorted(('Contig2', 'Contig3')),2, sorted(['geneB', 'geneE'])],\
[ sorted(('Contig3', 'Contig1')),2, sorted(['geneC', 'geneF'])]]))
def test_contig_joins_in_middle(self):
"""Tests when an intersection is found in the case where one end of the contig is closest to the middle of a second"""
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'gene4'),\
('gene4', 'gene5'), ('gene5', 'gene6') ,('gene3', 'gene7'),\
('gene7', 'gene8')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1,['gene3', 'gene7']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/empty_file' )
refine_contig_neighbours_object = RefineContigNeighbours(neighbouring_contigs, filtered_graph,filtered_graph, 'madansi/tests/data/empty_file',gene_detector, {})
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig1', 'gene4':'Contig1',\
'gene5':'Contig1', 'gene6':'Contig1', 'gene7':'Contig2', 'gene8':'Contig2'}
self.assertEqual(sorted(refine_contig_neighbours_object.refine_contig_neighbours()), [])
def test_three_contigs_together(self):
"""Tests when there are three contigs within a small distance of the intersection points"""
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene3', 'gene4'), ('gene5', 'gene6'),\
('gene7','gene2'), ('gene7','gene3'), ('gene7', 'gene5')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1,['gene7']], [('Contig2', 'Contig3'),1,['gene7']],\
[('Contig1', 'Contig3'),1,['gene7']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_4_sequences.fa', 'madansi/tests/data/empty_file' )
refine_contig_neighbours_object = RefineContigNeighbours(neighbouring_contigs, filtered_graph, filtered_graph,'madansi/tests/data/empty_file',gene_detector, {})
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig2',\
'gene4':'Contig2', 'gene5':'Contig3', 'gene6':'Contig3'}
refine_contig_neighbours_object.refine_contig_neighbours()
self.assertEqual(sorted(refine_contig_neighbours_object.refined_neighbouring_contigs), [])
def test_two_separated_sections(self):
gene_detector = GeneDetector(
'madansi/tests/data/assembly_4_sequences.fa',
'madansi/tests/data/seven_blast_hits')
filtered_graph = nx.Graph(
nx.drawing.nx_pydot.read_dot(
'madansi/tests/data/two_separated_sections.dot'))
contig_searching = ContigSearching(gene_detector, filtered_graph)
contig_searching.expand_all_contigs()
contig_neighbourhoods = contig_searching.neighbouring_contigs
possible_expected_lists = \
[sorted([('Contig1', 'Contig2', 1), ('Contig3', 'Contig4', 2)]),\
sorted([('Contig2', 'Contig1', 1), ('Contig3', 'Contig4', 2)]),\
sorted([('Contig1', 'Contig2', 1), ('Contig4', 'Contig3', 2)]),\
sorted([('Contig2', 'Contig1', 1), ('Contig4', 'Contig3', 2)])]
self.assertTrue(sorted(contig_neighbourhoods) == sorted([[ sorted(('Contig1', 'Contig2')), 1, sorted(['gene2', 'gene3'])],\
[ sorted(('Contig3', 'Contig4')), 2, sorted(['geneA', 'geneB'])]]))
def test_contigs_to_genes_one_hit(self):
"""Tests that given one blast hit, will give the correct keys and values"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa', 'madansi/tests/data/one_blast_hit')
self.assertCountEqual(list(gene_detector.contigs_to_genes().keys()), ['Contig1', 'Contig2', 'Contig3'])
self.assertEqual(gene_detector.contigs_to_genes()['Contig2'].gene_objects, {})
self.assertEqual(gene_detector.contigs_to_genes()['Contig3'].gene_objects, {})
self.assertCountEqual(list(gene_detector.contigs_to_genes()['Contig1'].gene_objects.keys()), ['gene1'])
my_gene = Gene(-1,402,1,None, 'Contig1',3)
self.assertEqual(gene_detector.contigs_to_genes()['Contig1'].gene_objects['gene1'].orientation, my_gene.orientation)
self.assertEqual(gene_detector.contigs_to_genes()['Contig1'].gene_objects['gene1'].start, my_gene.start)
self.assertEqual(gene_detector.contigs_to_genes()['Contig1'].gene_objects['gene1'].end, my_gene.end)
self.assertEqual(gene_detector.contigs_to_genes()['Contig1'].gene_objects['gene1'].node, my_gene.node)
self.assertEqual(gene_detector.contigs_to_genes()['Contig1'].gene_objects['gene1'].contig, my_gene.contig)
self.assertEqual(gene_detector.contigs_to_genes()['Contig1'].gene_objects['gene1'].qry_start, my_gene.qry_start)
def test_keep_two_connections(self):
"""Tests that two connections are preserved under refinement"""
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'gene4'),\
('gene4', 'gene5'), ('gene5', 'gene6'), ('gene6', 'gene7'),\
('gene7', 'gene8'), ('gene8', 'gene9'), ('gene9', 'gene10'),\
('gene10', 'gene11'), ('gene11', 'gene12'), ('gene12', 'gene13')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1, ['gene4']], [('Contig2', 'Contig3'),2,['gene8','gene9']]]
gene_detector = GeneDetector('madansi/tests/data/assembly_7_sequences.fa', 'madansi/tests/data/blast_hits_13' )
refine_contig_neighbours_object = RefineContigNeighbours(neighbouring_contigs,filtered_graph,filtered_graph,'madansi/tests/data/blast_hits_13', gene_detector, {})
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig1',\
'gene5':'Contig2', 'gene6':'Contig2', 'gene7':'Contig2', \
'gene10':'Contig3', 'gene11':'Contig3', 'gene13':'Contig3',\
'gene12':'Contig3'}
refine_contig_neighbours_object.refine_contig_neighbours()
self.assertEqual(sorted(refine_contig_neighbours_object.refined_neighbouring_contigs), sorted(neighbouring_contigs))
def test_finding_contig_ends_multiple_genes_same_side_of_closer_gene(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene1', 'gene2'), ('gene3', 'gene2'),
('gene3', 'gene4'), ('gene4', 'gene5')])
neighbouring_contigs = [[('Contig1', 'Contig2'), 1, ['gene3',
'gene4']]]
gene_detector = GeneDetector(
'madansi/tests/data/assembly_4_sequences.fa',
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
refine_contig_neighbours_object = RefineContigNeighbours(
neighbouring_contigs, filtered_graph, filtered_graph,
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file',
gene_detector, {
'Contig1': ['', '', 1000],
'Contig2': ['', '', 800]
})
refine_contig_neighbours_object.ends_of_contigs()
self.assertTrue(refine_contig_neighbours_object.contig_ends == {'Contig1':{'Contig2':[980,3]}, 'Contig2':{'Contig1':[2,301]}} or \
refine_contig_neighbours_object.contig_ends == {'Contig1':{'Contig2':[980,490]}, 'Contig2':{'Contig1':[2,301]}})
def test_add_to_contig_appearances(self):
filtered_graph = nx.Graph()
gene_detector = GeneDetector(
'madansi/tests/data/assembly_4_sequences.fa',
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
neighbouring_contigs = []
refine_contig_neighbours_object = RefineContigNeighbours(
neighbouring_contigs, filtered_graph, filtered_graph,
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file',
gene_detector, {
'Contig1': ['', '', 1000],
'Contig2': ['', '', 800],
'Contig3': ['', '', 2000]
})
self.assertDictEqual(
refine_contig_neighbours_object.add_to_contig_appearance(
'gene1', {}, 0), {'Contig1': [1, {
0: ['gene1']
}]})
self.assertDictEqual(
refine_contig_neighbours_object.add_to_contig_appearance(
'gene1', {'Contig1': [0, {}]}, 1),
{'Contig1': [1, {
1: ['gene1']
}]})
self.assertDictEqual(
refine_contig_neighbours_object.add_to_contig_appearance(
'gene1', {'Contig1': [1, {
0: ['gene2']
}]}, 1), {'Contig1': [2, {
0: ['gene2'],
1: ['gene1']
}]})
self.assertDictEqual(
refine_contig_neighbours_object.add_to_contig_appearance(
'gene1', {'Contig1': [2, {
1: ['gene2'],
2: ['gene3']
}]}, 2),
{'Contig1': [3, {
1: ['gene2'],
2: ['gene1', 'gene3']
}]})
def test_contigs_to_genes_four_hits(self):
"""Tests output for multiple blast hits"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa',
'madansi/tests/data/four_blast_hits')
self.assertCountEqual(list(gene_detector.contigs_to_genes().keys()),
['Contig1', 'Contig2', 'Contig3'])
self.assertCountEqual(
list(gene_detector.contigs_to_genes()
['Contig1'].gene_objects.keys()), ['gene1', 'gene2', 'gene3'])
self.assertCountEqual(
list(gene_detector.contigs_to_genes()
['Contig2'].gene_objects.keys()), ['geneA'])
self.assertCountEqual(
list(gene_detector.contigs_to_genes()
['Contig3'].gene_objects.keys()), [])
def test_two_contigs_unconnected(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([('gene1', 'gene2'), ('gene2', 'gene3'),
('gene4', 'gene5')])
gene_detector = GeneDetector(
'madansi/tests/data/assembly.fa',
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file')
filtered_blast_hits_file = 'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file'
expected_ordered_contig_graph = nx.Graph()
produced_ordered_contig_graph = ProduceOrderedContigGraph(
gene_detector, filtered_graph, filtered_graph,
filtered_blast_hits_file, {
'Contig1': ['', '', 1000],
'Contig2': ['', '', 800],
'Contig3': ['', '', 2000]
})
self.assertTrue(
nx.is_isomorphic(
expected_ordered_contig_graph,
produced_ordered_contig_graph.produce_ordered_contig_graph()))
def test_orientate_edges_one_edge(self):
gene_detector = GeneDetector(
'madansi/tests/data/assembly_4_sequences.fa',
'madansi/tests/data/seven_blast_hits')
contig_graph_1 = nx.Graph()
contig_graph_2 = nx.Graph()
contig_graph_3 = nx.Graph()
contig_graph_4 = nx.Graph()
contig_graph_1.add_edge('Contig2', 'Contig4', weight=0.5)
contig_graph_2.add_edge('Contig2', 'Contig3', weight=0.5)
contig_graph_3.add_edge('Contig1', 'Contig2', weight=0.5)
contig_graph_4.add_edge('Contig1', 'Contig3', weight=0.5)
contig_orientation_1_object = ContigOrientation(
contig_graph_1, gene_detector)
contig_orientation_1_object.repeat_all_connected_components()
contig_orientation_2_object = ContigOrientation(
contig_graph_2, gene_detector)
contig_orientation_2_object.repeat_all_connected_components()
contig_orientation_3_object = ContigOrientation(
contig_graph_3, gene_detector)
contig_orientation_3_object.repeat_all_connected_components()
contig_orientation_4_object = ContigOrientation(
contig_graph_4, gene_detector)
contig_orientation_4_object.repeat_all_connected_components()
self.assertEqual(
contig_orientation_1_object.contig_graph.edge['Contig2']['Contig4']
['weight'], 1)
self.assertEqual(
contig_orientation_2_object.contig_graph.edge['Contig2']['Contig3']
['weight'], 2)
self.assertEqual(
contig_orientation_3_object.contig_graph.edge['Contig1']['Contig2']
['weight'], 3)
self.assertEqual(
contig_orientation_4_object.contig_graph.edge['Contig1']['Contig3']
['weight'], 4)
def test_contigs_to_genes_no_hits(self):
"""Tests that the correct keys are given in the contigs object and that there is no value for each contig in this object"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa', 'madansi/tests/data/empty_file')
self.assertCountEqual(list(gene_detector.contigs_to_genes().keys()), ['Contig1', 'Contig2', 'Contig3'])
self.assertEqual(gene_detector.contigs_to_genes()['Contig1'].gene_objects, {})
def test_parse_blast_fits_empty_file(self):
"""Parses empty file correctly"""
gene_detector= GeneDetector('madansi/tests/data/empty_file.fa', 'madansi/tests/data/empty_file')
self.assertEqual(gene_detector.parse_blast_hits(), [])
def test_parse_blast_hits_one(self):
"""Tests that the correct type of object is obtained from parsing one blast hit"""
gene_detector = GeneDetector('madansi/tests/data/assembly.fa', 'madansi/tests/data/one_blast_hit')
self.assertTrue(isinstance(gene_detector.parse_blast_hits()[0], BlastHit))