def test_overlapping_deletion(self):
graph = graphContainer.GraphContainer()
n = graph.add_altNode("chr", 10, 17, "ATCGATCG")
varDel = {"start": 2, "end": 4, "alt": ""}
varSNV = {"start": 4, "end": 4, "alt": "C"}
variants.add_variants(graph, {n['name']: [varDel, varSNV]})
graphUtils.remove_empty_nodes(graph)
nodeNames = [n['name'] for n in graph.nodes.values()]
self.assertCountEqual(nodeNames, [
"chr:10-11:AT", "chr:12-13:CG", "chr:14-14:A", "chr:14-14:C",
"chr:15-17:TCG"
])
left = graph.nodes["chr:10-11:AT"]
right = graph.nodes["chr:15-17:TCG"]
alt = graph.nodes["chr:14-14:C"]
ref = graph.nodes["chr:14-14:A"]
ins = graph.nodes["chr:12-13:CG"]
self.assertTrue(graph.has_edge(left, right))
self.assertTrue(graph.has_edge(left, ins))
self.assertTrue(graph.has_edge(ins, ref))
self.assertTrue(graph.has_edge(ref, right))
self.assertTrue(graph.has_edge(ins, alt))
self.assertTrue(graph.has_edge(alt, right))
self.assertFalse(graph.has_edge(left, ref))
self.assertFalse(graph.has_edge(left, alt))
self.assertFalse(graph.has_edge(ref, ins))
def test_alt_snv(self):
graph = graphContainer.GraphContainer()
n = graph.add_altNode("chr", 10, 20, "ATCGATCG")
var = {"start": 2, "end": 2, "alt": "T"}
variants.add_variants(graph, {n['name']: [var]})
nodeNames = [n['name'] for n in graph.nodes.values()]
self.assertCountEqual(nodeNames, ["chr:10-11:AT", "chr:12-12:C", "chr:13-20:GATCG", "chr:12-12:T"])
self.assertEqual(graph.nodes["chr:10-11:AT"]["sequence"], "AT")
self.assertEqual(graph.nodes["chr:13-20:GATCG"]["sequence"], "GATCG")
self.assertEqual(graph.nodes["chr:12-12:C"]["sequence"], "C")
self.assertEqual(graph.nodes["chr:12-12:T"]["sequence"], "T")
def test_insertion(self):
graph = graphContainer.GraphContainer()
n = graph.add_altNode("chr", 10, 17, "ATCGATCG")
var = {"start": 3, "end": 2, "alt": "TTT"}
variants.add_variants(graph, {n['name']: [var]})
nodeNames = [n['name'] for n in graph.nodes.values()]
self.assertCountEqual(nodeNames, ["chr:10-12:ATC", "chr:13-17:GATCG", "chr:13-12:TTT"])
left = graph.nodes["chr:10-12:ATC"]
right = graph.nodes["chr:13-17:GATCG"]
ins = graph.nodes["chr:13-12:TTT"]
self.assertTrue(graph.has_edge(left, right))
self.assertTrue(graph.has_edge(left, ins))
self.assertTrue(graph.has_edge(ins, right))
self.assertFalse(graph.has_edge(ins, left))
self.assertFalse(graph.has_edge(right, ins))
def test_ref_snv(self):
graph = graphContainer.GraphContainer()
n = graph.add_refNode("chr", 10, 20)
var = {"start": 2, "end": 2, "alt": "C"}
variants.add_variants(graph, {n['name']: [var]})
nodeNames = [n['name'] for n in graph.nodes.values()]
self.assertCountEqual(nodeNames, ["ref-chr:10-11", "ref-chr:12-12", "ref-chr:13-20", "chr:12-12:C"])
self.assertEqual(graph.nodes["chr:12-12:C"]["sequence"], "C")
left = graph.nodes["ref-chr:10-11"]
right = graph.nodes["ref-chr:13-20"]
alt = graph.nodes["chr:12-12:C"]
ref = graph.nodes["ref-chr:12-12"]
self.assertTrue(graph.has_edge(left, ref))
self.assertTrue(graph.has_edge(left, alt))
self.assertTrue(graph.has_edge(ref, right))
self.assertTrue(graph.has_edge(alt, right))
self.assertFalse(graph.has_edge(left, right))
self.assertFalse(graph.has_edge(ref, alt))
def test_ins_end(self):
graph = graphContainer.GraphContainer()
r = graph.add_refNode("chr", 18, 20)
n = graph.add_altNode("chr", 10, 17, "ATCGATCG")
graph.add_edge(n, r, ["foo"])
var = {"start": 8, "end": 7, "alt": "CCC"}
variants.add_variants(graph, {n['name']: [var]})
graphUtils.remove_empty_nodes(graph)
left = graph.nodes["chr:10-17:ATCGATCG"]
right = graph.nodes["ref-chr:18-20"]
alt = graph.nodes["chr:18-17:CCC"]
self.assertEqual(len(graph.nodes), 3)
self.assertTrue(graph.has_edge(left, right))
self.assertTrue(graph.has_edge(left, alt))
self.assertTrue(graph.has_edge(alt, right))
self.assertFalse(graph.has_edge(alt, left))
self.assertFalse(graph.has_edge(right, alt))
self.assertCountEqual(graph.get_edge(left['name'], right['name'])['sequences'], ["foo"])
self.assertCountEqual(graph.get_edge(alt['name'], right['name'])['sequences'], ["foo"])
def run(args):
levels = [logging.WARNING, logging.INFO, logging.DEBUG]
logging.basicConfig(format='%(message)s',
level=levels[min(args.verbose,
len(levels) - 1)])
graphDict = load_json(args.graph)
graph = graphUtils.load_json(graphDict)
if args.variants:
varJson = load_json(args.variants)
if "variants" not in varJson:
raise Exception("No variants in variant JSON")
varDict = varJson["variants"]
else:
varDict = graphDict.get("variants", {})
if not varDict:
logging.warning("No variants in graph")
print(varDict.keys())
variants.add_variants(graph, varDict)
graphUtils.remove_empty_nodes(graph)
json.dump(graph.json_dict(), args.output, sort_keys=True)
def test_var_begin(self):
graph = graphContainer.GraphContainer()
r = graph.add_refNode("chr", 1, 9)
n = graph.add_altNode("chr", 10, 17, "ATCGATCG")
graph.add_edge(r, n, ["foo"])
var = {"start": 0, "end": 0, "alt": "G"}
variants.add_variants(graph, {n['name']: [var]})
graphUtils.remove_empty_nodes(graph)
left = graph.nodes["ref-chr:1-9"]
right = graph.nodes["chr:11-17:TCGATCG"]
ref = graph.nodes["chr:10-10:A"]
alt = graph.nodes["chr:10-10:G"]
self.assertEqual(len(graph.nodes), 4)
self.assertTrue(graph.has_edge(left, ref))
self.assertTrue(graph.has_edge(left, alt))
self.assertTrue(graph.has_edge(ref, right))
self.assertTrue(graph.has_edge(alt, right))
self.assertFalse(graph.has_edge(left, right))
self.assertFalse(graph.has_edge(ref, alt))
self.assertCountEqual(graph.get_edge(left['name'], ref['name'])['sequences'], ["foo"])
self.assertCountEqual(graph.get_edge(left['name'], alt['name'])['sequences'], ["foo"])