class TestHic(RegionMatrixContainerTestFactory):
def setup_method(self, method):
hic_file = os.path.join(test_dir, 'test_matrix', 'test_fanc.hic')
self.matrix = Hic(hic_file, mode='r')
def teardown_method(self, method):
self.matrix.close()
def test_masked_matrix(self):
hic = Hic()
m = np.zeros((12, 12))
row_regions = []
col_regions = []
# add some nodes (120 to be exact)
nodes = []
for i in range(1, 5000, 1000):
node = GenomicRegion(chromosome="chr1", start=i, end=i + 1000 - 1)
nodes.append(node)
row_regions.append(node)
col_regions.append(node)
for i in range(1, 3000, 1000):
node = GenomicRegion(chromosome="chr2", start=i, end=i + 1000 - 1)
nodes.append(node)
row_regions.append(node)
col_regions.append(node)
for i in range(1, 2000, 500):
node = GenomicRegion(chromosome="chr3", start=i, end=i + 1000 - 1)
nodes.append(node)
row_regions.append(node)
col_regions.append(node)
hic.add_regions(nodes)
# add some edges with increasing weight for testing
edges = []
weight = 1
for i in range(0, len(nodes)):
for j in range(i, len(nodes)):
if i != 1 and j != 1 and i != 5 and j != 5:
edges.append(Edge(source=i, sink=j, weight=weight))
m[i, j] = weight
m[j, i] = weight
weight += 1
hic.add_edges(edges)
m = hic.matrix()
hic.close()
# check masking
for i in range(m.shape[0]):
assert np.ma.is_masked(m[1, i])
assert np.ma.is_masked(m[i, 1])
assert np.ma.is_masked(m[5, i])
assert np.ma.is_masked(m[i, 5])
# check not masked
not_masked = {0, 2, 3, 4, 6, 7, 8, 9, 10, 11}
masked = {1, 5}
for j in not_masked:
for i in range(m.shape[0]):
if i not in masked:
assert not np.ma.is_masked(m[i, j])
assert not np.ma.is_masked(m[j, i])
else:
assert np.ma.is_masked(m[i, j])
assert np.ma.is_masked(m[j, i])
def test_merge(self):
hic = self.hic_class()
# add some nodes (120 to be exact)
nodes = []
for i in range(1, 5000, 1000):
nodes.append(
GenomicRegion(chromosome="chr1", start=i, end=i + 1000 - 1))
for i in range(1, 3000, 1000):
nodes.append(
GenomicRegion(chromosome="chr2", start=i, end=i + 1000 - 1))
for i in range(1, 2000, 500):
nodes.append(
GenomicRegion(chromosome="chr3", start=i, end=i + 1000 - 1))
hic.add_regions(nodes, preserve_attributes=False)
# add some edges with increasing weight for testing
edges = []
weight = 1
for i in range(0, len(nodes)):
for j in range(i, len(nodes)):
edges.append(Edge(source=i, sink=j, weight=weight))
weight += 1
hic.add_edges(edges)
# check length
merged_hic_2x = Hic.merge([self.hic, hic])
merged_hic_3x = Hic.merge([self.hic, hic, hic])
hic.close()
m = self.hic[:, :]
m_merged_2x = merged_hic_2x[:, :]
m_merged_3x = merged_hic_3x[:, :]
for i in range(m.shape[0]):
for j in range(m.shape[1]):
assert m[i, j] == 0 or m[i, j] == m_merged_2x[i, j] / 2
assert m[i, j] == 0 or m[i, j] == m_merged_3x[i, j] / 3
merged_hic_2x.close()
merged_hic_3x.close()
def setup_method(self, method):
self.dir = os.path.dirname(os.path.realpath(__file__))
hic = Hic()
# add some nodes (120 to be exact)
nodes = []
for i in range(1, 5000, 1000):
nodes.append(
GenomicRegion(chromosome="chr1", start=i, end=i + 1000 - 1))
for i in range(1, 3000, 1000):
nodes.append(
GenomicRegion(chromosome="chr2", start=i, end=i + 1000 - 1))
for i in range(1, 2000, 500):
nodes.append(
GenomicRegion(chromosome="chr3", start=i, end=i + 1000 - 1))
hic.add_regions(nodes)
# add some edges with increasing weight for testing
edges = []
weight = 1
for i in range(0, len(nodes)):
for j in range(i, len(nodes)):
edges.append(Edge(source=i, sink=j, weight=weight))
weight += 1
hic.add_edges(edges)
self.hic = hic
self.hic_cerevisiae = load(
self.dir + "/test_matrix/cerevisiae.chrI.HindIII_upgrade.hic",
mode='r')
self.hic_class = Hic
def sample_fa_hic(file_name=None, zero_indices=set(), tmpdir=None):
hic = Hic(file_name=file_name, tmpdir=tmpdir, mode='w')
# add some nodes (120 to be exact)
nodes = []
for i in range(1, 50000, 1000):
nodes.append(GenomicRegion(chromosome="chr1", start=i, end=i + 1000 - 1))
for i in range(1, 30000, 1000):
nodes.append(GenomicRegion(chromosome="chr2", start=i, end=i + 1000 - 1))
for i in range(1, 20000, 500):
nodes.append(GenomicRegion(chromosome="chr3", start=i, end=i + 1000 - 1))
hic.add_regions(nodes)
# add some edges with increasing weight for testing
edges = []
weight = 1
for i in range(0, len(nodes)):
for j in range(i, len(nodes)):
if i not in zero_indices and j not in zero_indices:
edges.append(Edge(source=i, sink=j, weight=weight))
weight += 1
hic.add_edges(edges)
return hic
def sample_hic_matrix1(file_name=None, tmpdir=None):
# 0 1 2 3 4 5 6 7 8 9
# #####################
# 0 # 0 1 0 2 0 3 0 4 0 5
# 1 # 6 0 7 0 8 0 9 0 1
# 2 # 2 3 4 0 5 0 0 6
# 3 # 7 0 8 9 0 1 0
# 4 # 0 2 3 0 0 4
# 5 # 5 6 7 8 9
# 6 # 1 0 0 0
# 7 # 2 3 0
# 8 # 0 4
# 9 # 5
nodes = [
GenomicRegion('chr1', 1, 1000),
GenomicRegion('chr1', 1001, 2000),
GenomicRegion('chr1', 2001, 3000),
GenomicRegion('chr1', 3001, 4000),
GenomicRegion('chr1', 4001, 5000),
GenomicRegion('chr1', 5001, 6000),
GenomicRegion('chr1', 6001, 7000),
GenomicRegion('chr1', 7001, 8000),
GenomicRegion('chr1', 8001, 9000),
GenomicRegion('chr1', 9001, 10000)
]
edges = [
Edge(source=0, sink=1, weight=1), Edge(source=3, sink=5, weight=8),
Edge(source=0, sink=3, weight=2), Edge(source=3, sink=6, weight=9),
Edge(source=0, sink=5, weight=3), Edge(source=3, sink=8, weight=1),
Edge(source=0, sink=7, weight=4), Edge(source=4, sink=5, weight=2),
Edge(source=0, sink=9, weight=5), Edge(source=4, sink=6, weight=3),
Edge(source=1, sink=1, weight=6), Edge(source=4, sink=9, weight=4),
Edge(source=1, sink=3, weight=7), Edge(source=5, sink=5, weight=5),
Edge(source=1, sink=5, weight=8), Edge(source=5, sink=6, weight=6),
Edge(source=1, sink=7, weight=9), Edge(source=5, sink=7, weight=7),
Edge(source=1, sink=9, weight=1), Edge(source=5, sink=8, weight=8),
Edge(source=2, sink=2, weight=2), Edge(source=5, sink=9, weight=9),
Edge(source=2, sink=3, weight=3), Edge(source=6, sink=6, weight=1),
Edge(source=2, sink=4, weight=4), Edge(source=7, sink=7, weight=2),
Edge(source=2, sink=6, weight=5), Edge(source=7, sink=8, weight=3),
Edge(source=2, sink=9, weight=6), Edge(source=8, sink=9, weight=4),
Edge(source=3, sink=3, weight=7), Edge(source=9, sink=9, weight=5)
]
hic = Hic(file_name=file_name, tmpdir=tmpdir)
hic.add_regions(nodes)
hic.add_edges(edges)
return hic
def sample_hic_matrix2(file_name=None, tmpdir=None):
# 0 1 2 3 4 5 6 7 8 9
# #####################
# 0 # 0 1 0 2 - 3 0 - 0 5
# 1 # 6 0 7 - 8 0 - 0 1
# 2 # 2 3 - 0 5 - 0 6
# 3 # 7 - 8 9 - 1 0
# 4 # - - - - - -
# 5 # 5 6 - 8 9
# 6 # 1 - 0 0
# 7 # - - -
# 8 # 0 4
# 9 # 5
nodes = [
GenomicRegion('chr1', 1, 1000),
GenomicRegion('chr1', 1001, 2000),
GenomicRegion('chr1', 2001, 3000),
GenomicRegion('chr1', 3001, 4000),
GenomicRegion('chr1', 4001, 5000),
GenomicRegion('chr1', 5001, 6000),
GenomicRegion('chr1', 6001, 7000),
GenomicRegion('chr1', 7001, 8000),
GenomicRegion('chr1', 8001, 9000),
GenomicRegion('chr1', 9001, 10000)
]
edges = [
Edge(source=0, sink=1, weight=1), Edge(source=3, sink=5, weight=8),
Edge(source=0, sink=3, weight=2), Edge(source=3, sink=6, weight=9),
Edge(source=0, sink=5, weight=3), Edge(source=3, sink=8, weight=1),
Edge(source=0, sink=9, weight=5),
Edge(source=1, sink=1, weight=6),
Edge(source=1, sink=3, weight=7), Edge(source=5, sink=5, weight=5),
Edge(source=1, sink=5, weight=8), Edge(source=5, sink=6, weight=6),
Edge(source=1, sink=9, weight=1), Edge(source=5, sink=8, weight=8),
Edge(source=2, sink=2, weight=2), Edge(source=5, sink=9, weight=9),
Edge(source=2, sink=3, weight=3), Edge(source=6, sink=6, weight=1),
Edge(source=2, sink=6, weight=5),
Edge(source=2, sink=9, weight=6), Edge(source=8, sink=9, weight=4),
Edge(source=3, sink=3, weight=7), Edge(source=9, sink=9, weight=5)
]
hic = Hic(file_name=file_name, tmpdir=tmpdir)
hic.add_regions(nodes)
hic.add_edges(edges)
return hic
def setup_method(self, method):
hic = Hic()
m = np.zeros((12, 12))
row_regions = []
col_regions = []
# add some nodes (120 to be exact)
nodes = []
for i in range(1, 5000, 1000):
node = GenomicRegion(chromosome="chr1", start=i, end=i + 1000 - 1)
nodes.append(node)
row_regions.append(node)
col_regions.append(node)
for i in range(1, 3000, 1000):
node = GenomicRegion(chromosome="chr2", start=i, end=i + 1000 - 1)
nodes.append(node)
row_regions.append(node)
col_regions.append(node)
for i in range(1, 2000, 500):
node = GenomicRegion(chromosome="chr3", start=i, end=i + 1000 - 1)
nodes.append(node)
row_regions.append(node)
col_regions.append(node)
hic.add_regions(nodes)
# add some edges with increasing weight for testing
edges = []
weight = 1
for i in range(0, len(nodes)):
for j in range(i, len(nodes)):
edges.append(Edge(source=i, sink=j, weight=weight))
m[i, j] = weight
m[j, i] = weight
weight += 1
hic.add_edges(edges)
self.hic = hic
self.m = RegionMatrix(m,
row_regions=row_regions,
col_regions=col_regions)
def sample_hic_big(file_name=None, tmpdir=None):
hic = Hic(file_name=file_name, tmpdir=tmpdir, mode='w')
# add some nodes (120 to be exact)
nodes = []
for i in range(1, 50000, 1000):
nodes.append(GenomicRegion(chromosome="chr1", start=i, end=i + 1000 - 1))
for i in range(1, 30000, 1000):
nodes.append(GenomicRegion(chromosome="chr2", start=i, end=i + 1000 - 1))
for i in range(1, 20000, 500):
nodes.append(GenomicRegion(chromosome="chr3", start=i, end=i + 1000 - 1))
hic.add_regions(nodes)
# add some edges with increasing weight for testing
edges = []
for i in range(0, len(nodes)):
for j in range(i, len(nodes)):
weight = i + len(nodes) - (j - i)
edges.append(Edge(source=i, sink=j, weight=weight))
hic.add_edges(edges)
return hic
def setup_method(self, method):
hic_file = os.path.join(test_dir, 'test_matrix', 'test_fanc.hic')
self.matrix = Hic(hic_file, mode='r')