def setUp(self): # pylint; disable-msg=C0103
"""test fixture"""
self.search_distances = {'upstream': (-20, 150)}
self.scan_distances = {'upstream': (-30, 250)}
matrix_factory = dm.DataMatrixFactory(
[dm.nochange_filter, dm.center_scale_filter])
infile = util.read_dfile('example_data/hal/halo_ratios5.tsv',
has_header=True,
quote='\"')
self.ratio_matrix = matrix_factory.create_from(infile)
self.organism = make_halo(self.ratio_matrix, self.search_distances,
self.scan_distances)
self.config_params = {
'memb.min_cluster_rows_allowed': 3,
'memb.max_cluster_rows_allowed': 70,
'multiprocessing': False,
'memb.clusters_per_row': 2,
'memb.clusters_per_col': int(round(43 * 2.0 / 3.0)),
'num_clusters': 43,
'output_dir': 'out',
'remap_network_nodes': False,
'num_iterations': 2000,
'debug': False
}
self.membership = self.__read_members() # relies on config_params
self.iteration_result = {'iteration': 51}
def test_motif_scoring(self):
"""tests the motif scoring in integration"""
search_distances = {'upstream': (-20, 150)}
scan_distances = {'upstream': (-30, 250)}
matrix_factory = dm.DataMatrixFactory([dm.nochange_filter, dm.center_scale_filter])
infile = util.read_dfile('example_data/hal/halo_ratios5.tsv',
has_header=True, quote='\"')
ratio_matrix = matrix_factory.create_from(infile)
meme_suite = meme.MemeSuite430(remove_tempfiles=True)
sequence_filters = [
motif.unique_filter,
motif.get_remove_low_complexity_filter(meme_suite),
motif.get_remove_atgs_filter(search_distances['upstream'])]
organism = make_halo(ratio_matrix, search_distances, scan_distances)
membership = FakeMembership()
config_params = {'memb.min_cluster_rows_allowed': 3,
'memb.max_cluster_rows_allowed': 70,
'multiprocessing': False,
'num_clusters': 1,
'output_dir': 'out',
'debug': False,
'num_iterations': 2000}
func = motif.MemeScoringFunction(organism, membership, ratio_matrix,
meme_suite,
sequence_filters=sequence_filters,
scaling_func=lambda iter: 1.0,
num_motif_func=lambda iter: 1,
update_in_iteration=lambda x: True,
motif_in_iteration=lambda x: True,
config_params=config_params)
iteration_result = { 'iteration': 100 }
matrix = func.compute(iteration_result)
"""
def test_simple_filter(self):
"""test a factory using a single filter"""
factory = dm.DataMatrixFactory([times2])
matrix = factory.create_from(self.dfile)
self.assertEquals(2, matrix.num_rows)
self.assertEquals(2, matrix.num_columns)
self.assertEquals(matrix.column_names, ["H2", "H3"])
self.assertEquals(matrix.row_names, ["R1", "R2"])
self.assertTrue((matrix.values[0] == [2, 4]).all())
self.assertTrue((matrix.values[1] == [6, 8]).all())
def test_no_filters(self):
"""test a factory without filters"""
factory = dm.DataMatrixFactory([])
matrix = factory.create_from(self.dfile)
self.assertEquals(2, matrix.num_rows)
self.assertEquals(2, matrix.num_columns)
self.assertEquals(matrix.column_names, ["H2", "H3"])
self.assertEquals(matrix.row_names, ["R1", "R2"])
self.assertTrue((matrix.values[0] == [1, 2]).all())
self.assertTrue((matrix.values[1] == [3, 4]).all())
def test_with_na_values(self):
"""test a factory with a DelimitedFile containing NA values"""
factory = dm.DataMatrixFactory([])
matrix = factory.create_from(self.dfile_with_na)
self.assertEquals(2, matrix.num_rows)
self.assertEquals(2, matrix.num_columns)
self.assertEquals(matrix.column_names, ["H2", "H3"])
self.assertEquals(matrix.row_names, ["R1", "R2"])
self.assertTrue(np.isnan(matrix.values[0][0]))
self.assertEquals(2.0, matrix.values[0][1])
self.assertTrue(np.isnan(matrix.values[1][0]))
self.assertEquals(4.0, matrix.values[1][1])
def test_motif_scoring(self):
"""tests the motif scoring in integration"""
search_distances = {'upstream': (-20, 150)}
scan_distances = {'upstream': (-30, 250)}
matrix_factory = dm.DataMatrixFactory(
[dm.nochange_filter, dm.center_scale_filter])
infile = util.read_dfile('example_data/hal/halo_ratios5.tsv',
has_header=True,
quote='\"')
ratio_matrix = matrix_factory.create_from(infile)
organism = testutil.make_halo(search_distances, scan_distances,
ratio_matrix)
membership = FakeMembership()
config_params = {
'memb.min_cluster_rows_allowed': 3,
'memb.max_cluster_rows_allowed': 70,
'multiprocessing': False,
'num_clusters': 1,
'output_dir': 'out',
'debug': {},
'search_distances': {
'upstream': (-20, 150)
},
'num_iterations': 2000,
'MEME': {
'schedule': lambda i: True,
'version': '4.3.0',
'global_background': False,
'arg_mod': 'zoops',
'nmotifs_rvec':
'c(rep(1, num_iterations/3), rep(2, num_iterations/3))',
'use_revcomp': 'True',
'max_width': 24,
'background_order': 3
},
'Motifs': {
'schedule': lambda i: True,
'scaling': ('scaling_const', 1.0)
}
}
func = motif.MemeScoringFunction(organism,
membership,
ratio_matrix,
config_params=config_params)
iteration_result = {'iteration': 100}
matrix = func.compute(iteration_result)
def __read_colscores_refresult(self):
dfile = util.read_dfile('testdata/column_scores_refresult.tsv',
has_header=True, quote='"')
return dm.DataMatrixFactory([]).create_from(dfile, case_sensitive=True)
def __read_ratios(self):
dfile = util.read_dfile('testdata/row_scores_testratios.tsv',
has_header=True)
return dm.DataMatrixFactory([]).create_from(dfile, case_sensitive=True)
def setUp(self): # pylint; disable-msg=C0103
"""test fixture"""
self.search_distances = {'upstream': (-20, 150)}
self.scan_distances = {'upstream': (-30, 250)}
matrix_factory = dm.DataMatrixFactory(
[dm.nochange_filter, dm.center_scale_filter])
infile = util.read_dfile('example_data/hal/halo_ratios5.tsv',
has_header=True,
quote='\"')
self.ratio_matrix = matrix_factory.create_from(infile)
self.organism = testutil.make_halo(self.search_distances,
self.scan_distances,
self.ratio_matrix)
self.config_params = {
'memb.min_cluster_rows_allowed': 3,
'memb.max_cluster_rows_allowed': 70,
'multiprocessing': False,
'num_cores': None,
'memb.clusters_per_row': 2,
'memb.clusters_per_col': int(round(43 * 2.0 / 3.0)),
'num_clusters': 43,
'output_dir': 'out',
'remap_network_nodes': False,
'use_BSCM': False,
'num_iterations': 2000,
'debug': {},
'search_distances': {
'upstream': (-20, 150)
},
'Columns': {
'schedule': lambda i: True
},
'Rows': {
'schedule': lambda i: True,
'scaling': ('scaling_const', 6.0)
},
'Motifs': {
'schedule': lambda i: True,
'scaling':
('scaling_rvec', 'seq(0, 1, length=num_iterations*3/4)')
},
'MEME': {
'version': '4.3.0',
'global_background': False,
'schedule': lambda i: True,
'nmotifs_rvec':
'c(rep(1, num_iterations/3), rep(2, num_iterations/3))',
'max_width': 24,
'arg_mod': 'zoops',
'background_order': 3,
'use_revcomp': 'True'
},
'Networks': {
'schedule':
lambda i: True,
'scaling':
('scaling_rvec', 'seq(1e-5, 0.5, length=num_iterations*3/4)')
}
}
self.membership = self.__read_members() # relies on config_params
self.iteration_result = {'iteration': 51, 'score_means': {}}
def read_matrix(filename):
"""reads a matrix file"""
infile = util.read_dfile(filename, has_header=True, quote='\"')
return dm.DataMatrixFactory([]).create_from(
infile, case_sensitive=True).sorted_by_row_name()
def __read_rowscores_refresult(self):
dfile = util.read_dfile('testdata/row_scores_refresult.tsv',
has_header=True,
quote='"')
return dm.DataMatrixFactory([]).create_from(dfile)