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_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 produce_ordered_contig_graph(self):
contig_searching = ContigSearching(self.gene_detector,
self.filtered_graph)
contig_searching.expand_all_contigs()
refine_neighbouring_contigs = RefineContigNeighbours(
contig_searching.neighbouring_contigs, self.filtered_graph,
self.unfiltered_graph, self.filtered_blast_hits_file,
self.gene_detector, self.sequences)
refine_neighbouring_contigs.refine_contig_neighbours()
self.contig_ends = refine_neighbouring_contigs.ends_of_contigs()
refined_neighbouring_contigs = refine_neighbouring_contigs.refined_neighbouring_contigs
contig_graph_refined = ContigGraph(refined_neighbouring_contigs)
contig_graph_unrefined = ContigGraph(
contig_searching.neighbouring_contigs)
graph_refined = contig_graph_refined.create_contig_subgraph()
graph_unrefined = contig_graph_unrefined.create_contig_subgraph()
iterate_joining_contig_components = IterateJoiningContigComponents(
graph_unrefined)
ordered_contig_graph = iterate_joining_contig_components.iterate_joining(
graph_refined)
return ordered_contig_graph
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_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_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_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_finding_contig_ends_3_cycle_both_different_half(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene4', 'gene5'), ('gene5', 'geneA'), ('geneA', 'gene8'),\
('geneA', 'gene6'), ('gene6', 'gene8'), ('gene6', 'gene7')])
neighbouring_contigs = [[('Contig2', '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_object = RefineContigNeighbours(neighbouring_contigs,filtered_graph, filtered_graph,'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file', gene_detector, {'Contig2':['','',800], 'Contig3':['','',1200]})
self.assertDictEqual(refine_contig_neighbours_object.ends_of_contigs(), {'Contig3':{'Contig2':[250,1200]}, 'Contig2':{'Contig3':[301,2]}})
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_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_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_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_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_orientation_of_contigs(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'gene4'),\
('gene4', 'gene5'), ('gene5', 'gene6'), ('gene6', 'gene7'),\
('gene7', 'gene8')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1, ['gene3','gene4']], [('Contig2', 'Contig3'),1,['gene5','gene6']]]
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], 'Contig3':['','',2000]})
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig1',\
'gene4':'Contig2', 'gene5':'Contig2', 'gene6':'Contig3', \
'gene7':'Contig3', 'gene8':'Contig3'}
refine_contig_neighbours_object.ends_of_contigs()
self.assertDictEqual(refine_contig_neighbours_object.contig_ends, {'Contig1': {'Contig2':[980,490]}, 'Contig2':{'Contig1':[2,301], 'Contig3':[301,2]}, 'Contig3':{'Contig2':[1,250]}})
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_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_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 produce_ordered_contig_graph(self):
contig_searching = ContigSearching(self.gene_detector, self.filtered_graph)
contig_searching.expand_all_contigs()
refine_neighbouring_contigs = RefineContigNeighbours(contig_searching.neighbouring_contigs, self.filtered_graph, self.unfiltered_graph, self.filtered_blast_hits_file, self.gene_detector, self.sequences)
refine_neighbouring_contigs.refine_contig_neighbours()
self.contig_ends = refine_neighbouring_contigs.ends_of_contigs()
refined_neighbouring_contigs= refine_neighbouring_contigs.refined_neighbouring_contigs
contig_graph_refined = ContigGraph(refined_neighbouring_contigs)
contig_graph_unrefined = ContigGraph(contig_searching.neighbouring_contigs)
graph_refined = contig_graph_refined.create_contig_subgraph()
graph_unrefined = contig_graph_unrefined.create_contig_subgraph()
iterate_joining_contig_components = IterateJoiningContigComponents(graph_unrefined)
ordered_contig_graph = iterate_joining_contig_components.iterate_joining(graph_refined)
return ordered_contig_graph
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_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_short_contig_in_middle(self):
"""Tests the case when there is a contig with only two genes between longer ones, we should find that the connection that needs more iterations is removed"""
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'gene4'),\
('gene4', 'gene5'), ('gene5', 'gene6'), ('gene6', 'gene7'),\
('gene7', 'gene8')])
neighbouring_contigs = [[('Contig1', 'Contig2'),1,['gene3','gene4']],\
[('Contig2', 'Contig3'),1,['gene5', 'gene6']],\
[('Contig1', 'Contig3'),2,['gene4', 'gene5']]]
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([[('Contig1', 'Contig2'),1,['gene3', 'gene4']], [('Contig2', 'Contig3'),1,['gene5', 'gene6']]]))
def test_orientation_further_separation(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene1', 'gene2'), ('gene2', 'gene3'), ('gene3', 'geneA'),\
('geneA', 'geneB'), ('geneB', 'gene4'), ('gene4', 'gene5'),\
('gene5', 'geneC'), ('geneC', 'gene6'), ('gene6', 'gene7'),\
('gene7', 'gene8')])
neighbouring_contigs = [[('Contig1', 'Contig2'), 1, ['geneA',
'geneB']],
[('Contig2', 'Contig3'), 1, ['geneC']]]
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],
'Contig3': ['', '', 2000]
})
refine_contig_neighbours_object.genes = { 'gene1':'Contig1', 'gene2':'Contig1', 'gene3':'Contig1',\
'gene4':'Contig2', 'gene5':'Contig2', 'gene6':'Contig3', \
'gene7':'Contig3', 'gene8':'Contig3', 'geneA':None, 'geneB':None, 'geneC':None}
refine_contig_neighbours_object.ends_of_contigs()
self.assertDictEqual(
refine_contig_neighbours_object.contig_ends, {
'Contig1': {
'Contig2': [980, 490]
},
'Contig2': {
'Contig1': [2, 301],
'Contig3': [301, 2]
},
'Contig3': {
'Contig2': [1, 250]
}
})
def test_finding_contig_ends_3_cycle_both_different_half(self):
filtered_graph = nx.Graph()
filtered_graph.add_edges_from([ ('gene4', 'gene5'), ('gene5', 'geneA'), ('geneA', 'gene8'),\
('geneA', 'gene6'), ('gene6', 'gene8'), ('gene6', 'gene7')])
neighbouring_contigs = [[('Contig2', '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_object = RefineContigNeighbours(
neighbouring_contigs, filtered_graph, filtered_graph,
'madansi/tests/data/refine_contig_neighbours_9_blast_hits_file',
gene_detector, {
'Contig2': ['', '', 800],
'Contig3': ['', '', 1200]
})
self.assertDictEqual(refine_contig_neighbours_object.ends_of_contigs(),
{
'Contig3': {
'Contig2': [250, 1200]
},
'Contig2': {
'Contig3': [301, 2]
}
})
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']
}]})
