class TestWrapper(unittest.TestCase):
def setUp(self):
self.correct_ob_graph = GraphWithReversals(
{
1: Block(7),
2: Block(4),
3: Block(7),
4: Block(4)
}, {
1: [2, 3],
2: [4],
3: [4]
})
self.correct_ob_graph.convert_to_numpy_backend()
self.correct_sequence_graph = SequenceGraph.create_empty_from_ob_graph(
self.correct_ob_graph)
self.correct_sequence_graph.set_sequences_using_vg_json_graph(
"tests/vg_test_graph.json")
remove_files = [
"tests/testgraph.obg", "tests/test_linear_map_starts.pickle",
"tests/test_linear_map_ends.pickle",
"tests/test_linear_map.length", "tests/sample.intervalcollection",
"tests/testintervals.intervalcollection",
"tests/testsequences.fasta",
"tests/node_range_test_data/vg_alignments_1.json"
"tests/node_range_test_data/vg_alignments_2.json"
"tests/node_range_test_data/vg_alignments_3.json"
"tests/node_range_test_data/vg_alignments_4.json"
"tests/node_range_test_data/vg_alignments_5.json"
]
for file in remove_files:
if os.path.isfile(file):
os.remove(file)
def test_create_from_nongraphpeakcollection(self):
graph = Graph({
1: Block(10),
2: Block(10),
3: Block(10)
}, {
1: [2],
2: [3]
})
graph.convert_to_numpy_backend()
linear_path = Interval(0, 10, [1, 2, 3], graph)
linear_path = linear_path.to_numpy_indexed_interval()
nongraph_peaks = NonGraphPeakCollection([
NonGraphPeak("chr1", 3, 10, 5),
NonGraphPeak("chr1", 13, 15, 7),
])
peaks = PeakCollection.create_from_nongraph_peak_collection(
graph, nongraph_peaks, linear_path, None)
self.assertEqual(peaks.intervals[0], Interval(3, 10, [1]))
self.assertEqual(peaks.intervals[1], Interval(3, 5, [2]))
peaks = PeakCollection.create_from_nongraph_peak_collection(
graph, nongraph_peaks, linear_path, LinearRegion("chr1", 3, 20))
self.assertEqual(peaks.intervals[0], Interval(0, 7, [1]))
self.assertEqual(peaks.intervals[1], Interval(0, 2, [2]))
class TestHaplotyper(unittest.TestCase):
def setUp(self):
self.complex_graph = Graph(
{i: Block(3) for i in range(1, 13)},
{
1: [2, 3],
2: [7, 8],
3: [4, 5],
4: [6],
5: [6],
6: [10],
7: [9],
8: [9],
9: [10],
10: [12]
})
self.complex_graph.convert_to_numpy_backend()
def test_simple(self):
graph = Graph(
{i: Block(3) for i in range(1, 5)},
{
1: [2, 3],
2: [4],
3: [4]
}
)
graph.convert_to_numpy_backend()
intervals = IntervalCollection([
Interval(0, 3, [1, 3])
])
haplotyper = HaploTyper(graph, intervals)
haplotyper.build()
max_interval = haplotyper.get_maximum_interval_through_graph()
self.assertEqual(
max_interval,
Interval(0, 3, [1, 3, 4])
)
def test_complex_graph(self):
intervals = IntervalCollection([
Interval(0, 3, [1, 3, 4, 6, 10]),
Interval(1, 2, [2]),
Interval(2, 3, [2]),
Interval(0, 3, [7, 9])
])
haplotyper = HaploTyper(self.complex_graph, intervals)
haplotyper.build()
max_interval = haplotyper.get_maximum_interval_through_graph()
self.assertEqual(
max_interval,
Interval(0, 3, [1, 2, 7, 9, 10, 12])
)
def test_simple(self):
graph = Graph(
{i: Block(3) for i in range(1, 5)},
{
1: [2, 3],
2: [4],
3: [4]
}
)
graph.convert_to_numpy_backend()
intervals = IntervalCollection([
Interval(0, 3, [1, 3])
])
haplotyper = HaploTyper(graph, intervals)
haplotyper.build()
max_interval = haplotyper.get_maximum_interval_through_graph()
self.assertEqual(
max_interval,
Interval(0, 3, [1, 3, 4])
)
def _create_data(self):
node_offset = 1
for chrom_number, chromosome in enumerate(self.chromosomes):
graph = Graph(
{i + node_offset: Block(10)
for i in range(0, 3)},
{i + node_offset: [i + 1 + node_offset]
for i in range(0, 2)})
linear_map = LinearMap.from_graph(graph)
linear_map_file_name = "linear_map_%s.npz" % chromosome
linear_map.to_file(linear_map_file_name)
self.linear_maps.append(linear_map_file_name)
self.sequence_retrievers.append(
SequenceRetriever(
{i + node_offset: "A" * 10
for i in range(0, 3)}))
self._create_reads(chrom_number, chromosome, graph)
node_offset += 3
graph.convert_to_numpy_backend()
SequenceGraph.create_empty_from_ob_graph(graph).to_file(
chromosome + ".nobg.sequences")
graph.to_file(chromosome + ".nobg")
def test_convert_to_approx_linear_peaks(self):
graph = Graph({i: Block(3)
for i in range(1, 10)}, {
1: [2],
2: [3],
3: [4],
4: [5],
5: [6],
6: [7, 8],
7: [9],
9: [9]
})
graph.convert_to_numpy_backend()
linear_interval = Interval(0, 3, [2, 4, 8, 9], graph)
linear_interval = linear_interval.to_numpy_indexed_interval()
peaks = PeakCollection([Peak(2, 2, [2, 3, 4]), Peak(1, 1, [3, 4, 5])])
linear_peaks = peaks.to_approx_linear_peaks(linear_interval, "chr4")
linear_peaks = linear_peaks.peaks
print(linear_peaks)
self.assertEqual(linear_peaks[0], NonGraphPeak("chr4", 2, 5))
self.assertEqual(linear_peaks[1], NonGraphPeak("chr4", 3, 3))