def get_step_parents(self, pipeline):
"""Return the possible step-parents associated with the parent"""
step_parents = set()
parent_name = self.x_name.value
for module in pipeline.modules():
if module.module_num == self.module_num:
return list(step_parents)
# Objects that are the parent of the parents
grandparents = module.get_measurements(
pipeline, parent_name, cellprofiler_core.measurement.C_PARENT)
step_parents.update(grandparents)
# Objects that are the children of the parents
siblings = module.get_measurements(
pipeline, parent_name,
cellprofiler_core.measurement.C_CHILDREN)
for sibling in siblings:
match = re.match("^([^_]+)_Count", sibling)
if match is not None:
sibling_name = match.groups()[0]
if parent_name in module.get_measurements(
pipeline, sibling_name,
cellprofiler_core.measurement.C_PARENT):
step_parents.add(sibling_name)
return list(step_parents)
def test_get_measurements():
workspace, module = make_workspace(dict(image=numpy.zeros((20, 10), bool)),
dict(image=numpy.zeros((20, 10), bool)))
for wants_emd, features in (
(
True,
list(cellprofiler.modules.measureimageoverlap.FTR_ALL) + [
cellprofiler.modules.measureimageoverlap.
FTR_EARTH_MOVERS_DISTANCE
],
),
(False, cellprofiler.modules.measureimageoverlap.FTR_ALL),
):
module.wants_emd.value = wants_emd
mnames = module.get_measurements(
workspace.pipeline,
"Image",
cellprofiler.modules.measureimageoverlap.C_IMAGE_OVERLAP,
)
assert len(mnames) == len(features)
assert all(n in features for n in mnames)
assert all(f in mnames for f in features)
mnames = module.get_measurements(
workspace.pipeline,
"Foo",
cellprofiler.modules.measureimageoverlap.C_IMAGE_OVERLAP,
)
assert len(mnames) == 0
mnames = module.get_measurements(workspace.pipeline, "Image", "Foo")
assert len(mnames) == 0
def validate_module(self, pipeline):
"""Validate the module's settings
Relate will complain if the children and parents are related
by a prior module or if a step-parent is named twice"""
for module in pipeline.modules():
if module == self:
break
parent_features = module.get_measurements(pipeline,
self.y_name.value,
"Parent")
if self.x_name.value in parent_features:
raise cellprofiler_core.setting.ValidationError(
"{} and {} were related by the {} module".format(
self.y_name.value, self.x_name.value,
module.module_name),
self.x_name,
)
if self.has_step_parents and self.wants_step_parent_distances:
step_parents = set()
for group in self.step_parent_names:
if group.step_parent_name.value in step_parents:
raise cellprofiler_core.setting.ValidationError(
"{} has already been chosen".format(
group.step_parent_name.value),
group.step_parent_name,
)
step_parents.add(group.step_parent_name.value)
def test_metadata_row_and_column_and_well():
csv_text = """"Metadata_Row","Metadata_Column","Metadata_Well"
"C","03","B14"
"""
pipeline, module, filename = make_pipeline(csv_text)
columns = module.get_measurement_columns(pipeline)
assert any([
c[0] == cellprofiler_core.measurement.IMAGE and c[1] == "Metadata_Row"
and c[2] == "varchar(1)" for c in columns
])
assert any([
c[0] == cellprofiler_core.measurement.IMAGE
and c[1] == "Metadata_Column" and c[2] == "varchar(2)" for c in columns
])
assert any([
c[0] == cellprofiler_core.measurement.IMAGE and c[1] == "Metadata_Well"
and c[2] == "varchar(3)" for c in columns
])
m = pipeline.run()
features = module.get_measurements(
pipeline,
cellprofiler_core.measurement.IMAGE,
cellprofiler_core.measurement.C_METADATA,
)
for feature, expected in (("Row", "C"), ("Column", "03"), ("Well", "B14")):
assert feature in features
value = m.get_current_image_measurement("_".join(
(cellprofiler_core.measurement.C_METADATA, feature)))
assert value == expected
def test_means_of_distances(self):
#
# Regression test of issue #1409
#
# Make sure means of minimum and mean distances of children
# are recorded properly
#
i, j = numpy.mgrid[0:14, 0:30]
#
# Make the objects different sizes to exercise more code
#
parent_labels = (i >= 7) * 1 + (j >= 15) * 2 + 1
child_labels = numpy.zeros(i.shape)
numpy.random.seed(0)
# Take 12 random points and label them
child_centers = numpy.random.permutation(numpy.prod(i.shape))[:12]
child_centers = numpy.vstack(
(i.flatten()[child_centers], j.flatten()[child_centers]))
child_labels[child_centers[0], child_centers[1]] = numpy.arange(1, 13)
parent_centers = numpy.array([[3, 7], [10, 7], [3, 22], [10, 22]],
float)
parent_indexes = parent_labels[child_centers[0], child_centers[1]] - 1
expected = numpy.sqrt(
numpy.sum((parent_centers[parent_indexes, :] -
child_centers.transpose())**2, 1))
workspace, module = self.make_workspace(parent_labels, child_labels)
assert isinstance(module, cellprofiler.modules.relateobjects.Relate)
module.find_parent_child_distances.value = (
cellprofiler.modules.relateobjects.D_CENTROID)
module.wants_per_parent_means.value = True
mnames = module.get_measurements(
workspace.pipeline,
PARENT_OBJECTS,
"_".join(
(cellprofiler.modules.relateobjects.C_MEAN, CHILD_OBJECTS)),
)
assert cellprofiler.modules.relateobjects.FF_CENTROID % PARENT_OBJECTS in mnames
feat_mean = cellprofiler.modules.relateobjects.FF_MEAN % (
CHILD_OBJECTS,
cellprofiler.modules.relateobjects.FF_CENTROID % PARENT_OBJECTS,
)
mcolumns = module.get_measurement_columns(workspace.pipeline)
assert any(
[c[0] == PARENT_OBJECTS and c[1] == feat_mean for c in mcolumns])
m = workspace.measurements
m[CHILD_OBJECTS, M_LOCATION_CENTER_X, 1] = child_centers[1]
m[CHILD_OBJECTS, M_LOCATION_CENTER_Y, 1] = child_centers[0]
module.run(workspace)
v = m[PARENT_OBJECTS, feat_mean, 1]
plabel = m[CHILD_OBJECTS, "_".join((C_PARENT, PARENT_OBJECTS)), 1, ]
assert len(v) == 4
for idx in range(4):
if numpy.any(plabel == idx + 1):
assert (round(
abs(v[idx] - numpy.mean(expected[plabel == idx + 1])),
4) == 0)
def test_get_measurements_other_other(self):
module = (cellprofiler_core.module.image_segmentation.
_object_processing.ObjectProcessing())
module.x_name.value = "Objects"
actual = module.get_measurements(None, "foo", "bar")
expected = []
assert actual == expected
def test_get_measurements_other_other(self):
module = (cellprofiler_core.module.image_segmentation.
_image_segmentation.ImageSegmentation())
module.x_name.value = "Image"
actual = module.get_measurements(None, "foo", "bar")
expected = []
assert actual == expected
def test_get_measurements_output_object_parent(self):
module = (cellprofiler_core.module.image_segmentation.
_object_processing.ObjectProcessing())
module.x_name.value = "Objects"
actual = module.get_measurements(
None, "ObjectProcessing",
cellprofiler_core.constants.measurement.C_PARENT)
expected = ["Objects"]
assert actual == expected
def test_get_measurements_output_object_number(self):
module = (cellprofiler_core.module.image_segmentation.
_object_processing.ObjectProcessing())
module.x_name.value = "Objects"
actual = module.get_measurements(
None, "ObjectProcessing",
cellprofiler_core.constants.measurement.C_NUMBER)
expected = [cellprofiler_core.constants.measurement.FTR_OBJECT_NUMBER]
assert actual == expected
def test_get_measurements_input_object_children(self):
module = (cellprofiler_core.module.image_segmentation.
_object_processing.ObjectProcessing())
module.x_name.value = "Objects"
actual = module.get_measurements(
None, "Objects",
cellprofiler_core.constants.measurement.C_CHILDREN)
expected = ["%s_Count" % "ObjectProcessing"]
assert actual == expected
def test_get_measurements_image_count(self):
module = (cellprofiler_core.module.image_segmentation.
_object_processing.ObjectProcessing())
module.x_name.value = "Objects"
actual = module.get_measurements(
None,
cellprofiler_core.constants.measurement.IMAGE,
cellprofiler_core.constants.measurement.C_COUNT,
)
expected = ["ObjectProcessing"]
assert actual == expected
def test_get_measurements_image_count(self):
module = (cellprofiler_core.module.image_segmentation.
_image_segmentation.ImageSegmentation())
module.x_name.value = "Image"
actual = module.get_measurements(
None,
cellprofiler_core.constants.measurement.IMAGE,
cellprofiler_core.constants.measurement.C_COUNT,
)
expected = ["ImageSegmentation"]
assert actual == expected
def test_get_measurements_output_object_location(self):
module = (cellprofiler_core.module.image_segmentation.
_object_processing.ObjectProcessing())
module.x_name.value = "Objects"
actual = module.get_measurements(
None, "ObjectProcessing",
cellprofiler_core.constants.measurement.C_LOCATION)
expected = [
cellprofiler_core.constants.measurement.FTR_CENTER_X,
cellprofiler_core.constants.measurement.FTR_CENTER_Y,
cellprofiler_core.constants.measurement.FTR_CENTER_Z,
]
assert actual == expected
def test_02_get_zernike_measurements():
module = (cellprofiler.modules.measureobjectintensitydistribution.
MeasureObjectIntensityDistribution())
for wants_zernikes, ftrs in (
(
cellprofiler.modules.measureobjectintensitydistribution.
Z_MAGNITUDES,
(cellprofiler.modules.measureobjectintensitydistribution.
FF_ZERNIKE_MAGNITUDE, ),
),
(
cellprofiler.modules.measureobjectintensitydistribution.
Z_MAGNITUDES_AND_PHASE,
(
cellprofiler.modules.measureobjectintensitydistribution.
FF_ZERNIKE_MAGNITUDE,
cellprofiler.modules.measureobjectintensitydistribution.
FF_ZERNIKE_PHASE,
),
),
):
module.wants_zernikes.value = wants_zernikes
module.zernike_degree.value = 2
module.images_list.value = "DNA, Cytoplasm, Actin"
for i, object_name, center_name in (
(0, "Nucleii", None),
(1, "Cells", "Nucleii"),
(2, "Cytoplasm", "Nucleii"),
):
if i:
module.add_object()
module.objects[i].object_name.value = object_name
if center_name is None:
module.objects[i].center_choice.value = (
cellprofiler.modules.measureobjectintensitydistribution.
C_SELF)
else:
module.objects[i].center_choice.value = (
cellprofiler.modules.measureobjectintensitydistribution.
C_CENTERS_OF_OTHER)
module.objects[i].center_object_name.value = center_name
for object_name in "Nucleii", "Cells", "Cytoplasm":
result = module.get_measurements(
None,
object_name,
cellprofiler.modules.measureobjectintensitydistribution.
M_CATEGORY,
)
for ftr in ftrs:
assert ftr in result
iresult = module.get_measurement_images(
None,
object_name,
cellprofiler.modules.measureobjectintensitydistribution.
M_CATEGORY,
ftr,
)
for image in "DNA", "Cytoplasm", "Actin":
assert image in iresult
sresult = module.get_measurement_scales(
None,
object_name,
cellprofiler.modules.
measureobjectintensitydistribution.M_CATEGORY,
ftr,
image,
)
for n, m in ((0, 0), (1, 1), (2, 0), (2, 2)):
assert "%d_%d" % (n, m) in sresult
def test_01_get_measurements():
module = (cellprofiler.modules.measureobjectintensitydistribution.
MeasureObjectIntensityDistribution())
module.images_list.value = "DNA, Cytoplasm, Actin"
for i, object_name, center_name in (
(0, "Nucleii", None),
(1, "Cells", "Nucleii"),
(2, "Cytoplasm", "Nucleii"),
):
if i:
module.add_object()
module.objects[i].object_name.value = object_name
if center_name is None:
module.objects[i].center_choice.value = (
cellprofiler.modules.measureobjectintensitydistribution.C_SELF)
else:
module.objects[i].center_choice.value = (
cellprofiler.modules.measureobjectintensitydistribution.
C_CENTERS_OF_OTHER)
module.objects[i].center_object_name.value = center_name
for i, bin_count in ((0, 4), (0, 5), (0, 6)):
if i:
module.add_bin_count()
module.bin_counts[i].bin_count.value = bin_count
for object_name in [x.object_name.value for x in module.objects]:
assert tuple(module.get_categories(None, object_name)) == (
cellprofiler.modules.measureobjectintensitydistribution.M_CATEGORY,
)
for feature in cellprofiler.modules.measureobjectintensitydistribution.F_ALL:
assert feature in module.get_measurements(
None,
object_name,
cellprofiler.modules.measureobjectintensitydistribution.
M_CATEGORY,
)
for image_name in module.images_list.value:
for (
feature
) in cellprofiler.modules.measureobjectintensitydistribution.F_ALL:
assert image_name in module.get_measurement_images(
None,
object_name,
cellprofiler.modules.measureobjectintensitydistribution.
M_CATEGORY,
feature,
)
for bin_count in [x.bin_count.value for x in module.bin_counts]:
for bin in range(1, bin_count + 1):
for (
feature
) in cellprofiler.modules.measureobjectintensitydistribution.F_ALL:
assert "%dof%d" % (
bin,
bin_count,
) in module.get_measurement_scales(
None,
object_name,
cellprofiler.modules.
measureobjectintensitydistribution.M_CATEGORY,
feature,
image_name,
)
