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_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']]]))