def saveTable(table: DecodedIntensityTable, savename: str):
"""
Reformats and saves a DecodedIntensityTable.
"""
intensities = IntensityTable(
table.where(table[Features.PASSES_THRESHOLDS], drop=True))
traces = intensities.stack(traces=(Axes.ROUND.value, Axes.CH.value))
# traces = table.stack(traces=(Axes.ROUND.value, Axes.CH.value))
traces = traces.to_features_dataframe()
traces.to_csv(savename)
def test_from_spot_attributes_must_have_aligned_dimensions_spot_attributes_and_data(
):
"""
Number of features must match number of SpotAttributes. Pass two attributes and 3 features and
verify a ValueError is raised.
"""
spot_attributes = spot_attribute_factory(2)
data = np.zeros(30).reshape(3, 5, 2)
with pytest.raises(ValueError):
IntensityTable.from_spot_data(data, spot_attributes)
def test_from_spot_attributes_throws_type_error_when_passed_a_dataframe():
"""SpotAttributes should be passed instead."""
# input has two spots
not_spot_attributes = pd.DataFrame(
data=np.array([[1, 1, 1, 1], [2, 2, 2, 1]]),
columns=[Indices.Z, Indices.Y, Indices.X, Features.SPOT_RADIUS])
with pytest.raises(TypeError):
IntensityTable.empty_intensity_table(not_spot_attributes,
n_ch=1,
n_round=3)
def decoded_intensity_table_factory() -> Tuple[IntensityTable, np.ndarray]:
"""
Create an IntensityTable that has gene labels, including null labels. The data doesn't matter,
so will use np.zeros
"""
data = np.zeros((1, 1, 2, 3, 3), dtype=np.float32)
labels = np.array(
[[[0, 1, 1],
[0, 2, 2],
[1, 1, 1]],
[[0, 1, 1],
[1, 1, 1],
[0, 1, 2]]],
dtype='<U3'
)
labels_with_nan = labels.copy()
labels_with_nan[labels == '0'] = 'nan'
# create an intensity table and add the labels
image_stack = ImageStack.from_numpy(data)
intensities = IntensityTable.from_image_stack(image_stack)
intensities[Features.TARGET] = (Features.AXIS, np.ravel(labels_with_nan))
# label the third column of this data as failing filters
passes_filters = np.ones(data.shape, dtype=bool)
passes_filters[:, :, :, :, -1] = 0
intensities[Features.PASSES_THRESHOLDS] = (Features.AXIS, np.ravel(passes_filters))
return intensities, labels_with_nan
def create_intensity_table_with_coords(area: Area,
n_spots: int = 10) -> IntensityTable:
"""
Creates a 50X50 intensity table with physical coordinates within
the given Area.
Parameters
----------
area: Area
The area of physical space the IntensityTable should be defined over
n_spots:
Number of spots to add to the IntensityTable
"""
codebook = factories.codebook_array_factory()
it = IntensityTable.synthetic_intensities(codebook,
num_z=1,
height=50,
width=50,
n_spots=n_spots)
# intensity table 1 has 10 spots, xmin = 0, ymin = 0, xmax = 2, ymax = 1
it[Coordinates.X.value] = xr.DataArray(np.linspace(area.min_x, area.max_x,
n_spots),
dims=Features.AXIS)
it[Coordinates.Y.value] = xr.DataArray(np.linspace(area.min_y, area.max_y,
n_spots),
dims=Features.AXIS)
return it
def compute_magnitudes(stack, norm_order=2):
pixel_intensities = IntensityTable.from_image_stack(stack)
feature_traces = pixel_intensities.stack(traces=(Axes.CH.value, Axes.ROUND.value))
norm = np.linalg.norm(feature_traces.values, ord=norm_order, axis=1)
return norm
def labeled_intensities_factory(
) -> Tuple[IntensityTable, np.ndarray, np.ndarray]:
"""
Create a decoded IntensityTable with distance scores, and a corresponding label_image and
decoded_image.
"""
data = np.array(
[
[
[[0., 0.], [.1, .1]], # ch 1
[[.5, .5], [.2, .3]]
],
[
[[.1, .1], [0, 0]], # ch 2, x & y are reversed
[[.2, .3], [.5, .5]]
]
],
dtype=np.float32)
image_stack = ImageStack.from_numpy(data.reshape(1, 2, 2, 2, 2))
intensity_table = IntensityTable.from_image_stack(image_stack)
intensity_table[Features.DISTANCE] = (Features.AXIS,
np.zeros(intensity_table.shape[0]))
label_image = np.array([[[1, 2], [3, 4]], [[1, 2], [3, 4]]])
decoded_image = np.array([[[5, 4], [3, 2]], [[5, 4], [3, 2]]])
# verify that the listed label image is what would be created by the function we use in the
# code
assert np.array_equal(label(decoded_image), label_image)
return intensity_table, label_image, decoded_image
def intensities(self, codebook=None) -> IntensityTable:
if codebook is None:
codebook = self.codebook()
intensities = IntensityTable.synthetic_intensities(
codebook, self.n_z, self.height, self.width, self.n_spots,
self.mean_fluor_per_spot, self.mean_photons_per_fluor)
assert intensities.dtype == np.float32 and intensities.max() <= 1
return intensities
def verify_results(self, tempdir: Path):
intensities = IntensityTable.open_netcdf(
os.fspath(tempdir / "decoded_spots.nc"))
genes, counts = np.unique(intensities.coords[Features.TARGET],
return_counts=True)
gene_counts = pd.Series(counts, genes)
assert gene_counts['ACTB'] == 9
assert gene_counts['GAPDH'] == 9
def processing_pipeline(
experiment: starfish.Experiment,
fov_name: str,
n_processes: Optional[int] = None
) -> Tuple[starfish.ImageStack, starfish.IntensityTable]:
"""Process a single field of view of an experiment
Parameters
----------
experiment : starfish.Experiment
starfish experiment containing fields of view to analyze
fov_name : str
name of the field of view to process
n_processes : int
Returns
-------
starfish.IntensityTable :
decoded IntensityTable containing spots matched to the genes they are hybridized against
"""
print("Loading images...")
primary_image = experiment[fov_name].get_image(FieldOfView.PRIMARY_IMAGES)
all_intensities = list()
codebook = experiment.codebook
images = enumerate(experiment[fov_name].iterate_image_type(
FieldOfView.PRIMARY_IMAGES))
for image_number, primary_image in images:
print(f"Filtering image {image_number}...")
filter_kwargs = dict(in_place=True,
verbose=True,
n_processes=n_processes)
print("Applying Clip...")
clip1.run(primary_image, **filter_kwargs)
print("Applying Bandpass...")
bandpass.run(primary_image, **filter_kwargs)
print("Applying Gaussian Low Pass...")
glp.run(primary_image, **filter_kwargs)
print("Applying Clip...")
clip2.run(primary_image, **filter_kwargs)
print("Calling spots...")
spot_attributes = tlmpf.run(primary_image)
all_intensities.append(spot_attributes)
spot_attributes = IntensityTable.concatenate_intensity_tables(
all_intensities)
print("Decoding spots...")
decoded = codebook.decode_per_round_max(spot_attributes)
decoded = decoded[decoded["total_intensity"] > .025]
print("Processing complete.")
return primary_image, decoded
def test_intensity_table_can_be_constructed_from_a_numpy_array_and_spot_attributes(
):
"""
Verify that the IntensityTable can be created and that the resulting data matches the array
it was constructed from.
"""
spot_attributes = spot_attribute_factory(3)
data = np.zeros(30).reshape(3, 5, 2)
intensities = IntensityTable.from_spot_data(data, spot_attributes)
assert intensities.shape == data.shape
assert np.array_equal(intensities.values, data)
def intensity_table_factory() -> IntensityTable:
"""IntensityTable with a single feature that was measured over 2 channels and 2 rounds."""
intensities = np.array([[[0, 3], [4, 0]]], dtype=float)
spot_attribute_data = pd.DataFrame(
data=[0, 0, 0, 1],
index=[Axes.ZPLANE, Axes.Y, Axes.X, Features.SPOT_RADIUS]).T
spot_attributes = SpotAttributes(spot_attribute_data)
intensity_table = IntensityTable.from_spot_data(intensities,
spot_attributes)
return intensity_table
def intensity_table_factory(data: np.ndarray=np.array([[[0, 3], [4, 0]]])) -> IntensityTable:
"""IntensityTable with a single feature that was measured over 2 channels and 2 rounds."""
# generates spot attributes equal in size to the number of passed features.
# each attribute has coordinates (z, y, x) equal to the feature index, and radius 1.
spot_attributes_data = pd.DataFrame(
data=np.array([[i, i, i, 1] for i in np.arange(data.shape[0])]),
columns=[Axes.ZPLANE, Axes.Y, Axes.X, Features.SPOT_RADIUS]
)
spot_attributes = SpotAttributes(spot_attributes_data)
intensity_table = IntensityTable.from_spot_data(data, spot_attributes)
return intensity_table
def synthetic_decoded_intensity_table(
codebook,
num_z: int = 12,
height: int = 50,
width: int = 40,
n_spots: int = 10,
mean_fluor_per_spot: int = 200,
mean_photons_per_fluor: int = 50,
) -> DecodedIntensityTable:
"""
Creates an IntensityTable with synthetic spots, that correspond to valid
codes in a provided codebook.
Parameters
----------
codebook : Codebook
Starfish codebook object.
num_z : int
Number of z-planes to use when localizing spots.
height : int
y dimension of each synthetic plane.
width : int
x dimension of each synthetic plane.
n_spots : int
Number of spots to generate.
mean_fluor_per_spot : int
Mean number of fluorophores per spot.
mean_photons_per_fluor : int
Mean number of photons per fluorophore.
Returns
-------
DecodedIntensityTable
"""
intensities = IntensityTable.synthetic_intensities(
codebook,
num_z=num_z,
height=height,
width=width,
n_spots=n_spots,
mean_fluor_per_spot=mean_fluor_per_spot,
mean_photons_per_fluor=mean_photons_per_fluor)
targets = np.random.choice(codebook.coords[Features.TARGET],
size=n_spots,
replace=True)
return DecodedIntensityTable.from_intensity_table(intensities,
targets=(Features.AXIS,
targets))
def verify_results(self, intensities):
assert intensities[Features.PASSES_THRESHOLDS].sum()
spots_df = IntensityTable(
intensities.where(intensities[Features.PASSES_THRESHOLDS],
drop=True)).to_features_dataframe()
spots_df['area'] = np.pi * spots_df['radius']**2
# verify number of spots detected
spots_passing_filters = intensities[Features.PASSES_THRESHOLDS].sum()
assert spots_passing_filters == 53 # TODO note, had to change this by 1
# compare to benchmark data -- note that this particular part of the dataset
# appears completely uncorrelated
cnts_benchmark = pd.read_csv(
'https://d2nhj9g34unfro.cloudfront.net/20181005/DARTFISH/fov_001/counts.csv'
)
min_dist = 0.6
cnts_starfish = spots_df[spots_df.distance <= min_dist].groupby(
'target').count()['area']
cnts_starfish = cnts_starfish.reset_index(level=0)
cnts_starfish.rename(columns={
'target': 'gene',
'area': 'cnt_starfish'
},
inplace=True)
# get top 5 genes and verify they are correct
high_expression_genes = cnts_starfish.sort_values(
'cnt_starfish', ascending=False).head(5)
assert np.array_equal(high_expression_genes['cnt_starfish'].values,
[7, 3, 2, 2, 2])
assert np.array_equal(high_expression_genes['gene'].values,
['MBP', 'MOBP', 'ADCY8', 'TRIM66', 'SYT6'])
# verify correlation is accurate for this subset of the image
benchmark_comparison = pd.merge(cnts_benchmark,
cnts_starfish,
on='gene',
how='left')
benchmark_comparison.head(20)
x = benchmark_comparison.dropna().cnt.values
y = benchmark_comparison.dropna().cnt_starfish.values
corrcoef = np.corrcoef(x, y)
corrcoef = corrcoef[0, 1]
assert np.round(corrcoef, 5) == 0.03028
def test_intensity_table_concatenation():
"""create two IntensityTables and assert that they are being concatenated properly."""
r, c, z, y, x = 3, 3, 2, 2, 5
data = np.zeros(180, dtype=np.float32).reshape(r, c, z, y, x)
image_stack = ImageStack.from_numpy_array(data)
intensities = IntensityTable.from_image_stack(image_stack)
intensities2 = intensities.copy()
original_shape = intensities.shape
expected_shape = list(original_shape)
expected_shape[0] *= 2 # only features is concatenated
assert np.array_equal(
concatenate([intensities, intensities2]).shape, expected_shape)
# slice out a single channel and round from both experiments, such that the data no longer match
# across all dimensions but the concatenation dimension. The resulting structure should be
# 2 (r) * 2 (c) * 5 (z), 2, 2 = 40, 2, 2
i1 = intensities.where(np.logical_and(intensities.r == 0,
intensities.c == 0),
drop=True)
i2 = intensities.where(np.logical_and(intensities.r == 1,
intensities.c == 1),
drop=True)
expected_shape = (i1.shape[0] + i2.shape[0], 2, 2)
result = concatenate([i1, i2])
assert expected_shape == result.shape
# slice a larger r value for second array, however, there are still only two values, so
# shape should be 40, 2, 2
i3 = intensities.where(np.logical_and(intensities.r == 2,
intensities.c == 1),
drop=True)
expected_shape = (i1.shape[0] + i3.shape[0], 2, 2)
result = concatenate([i1, i3])
assert expected_shape == result.shape
# slice out z in addition to reduce the total feature number by 1/2
i4 = intensities.where(np.logical_and(intensities.r == 0,
intensities.z == 1),
drop=True)
expected_shape = (i1.shape[0] + i4.shape[0], 3, 1)
result = concatenate([i1, i4])
assert expected_shape == result.shape
def test_tranfering_physical_coords_to_intensity_table():
stack_shape = OrderedDict([(Axes.ROUND, 3), (Axes.CH, 2), (Axes.ZPLANE, 1),
(Axes.Y, 50), (Axes.X, 40)])
physical_coords = OrderedDict([(PhysicalCoordinateTypes.X_MIN, 1),
(PhysicalCoordinateTypes.X_MAX, 2),
(PhysicalCoordinateTypes.Y_MIN, 4),
(PhysicalCoordinateTypes.Y_MAX, 6),
(PhysicalCoordinateTypes.Z_MIN, 1),
(PhysicalCoordinateTypes.Z_MAX, 3)])
stack = factories.imagestack_with_coords_factory(stack_shape,
physical_coords)
codebook = factories.codebook_array_factory()
intensities = IntensityTable.synthetic_intensities(
codebook,
num_z=stack_shape[Axes.ZPLANE],
height=stack_shape[Axes.Y],
width=stack_shape[Axes.X],
n_spots=NUMBER_SPOTS)
intensities = intensity_table_coordinates.\
transfer_physical_coords_from_imagestack_to_intensity_table(stack, intensities)
# Assert that new cords were added
xc = intensities.coords[Coordinates.X]
yc = intensities.coords[Coordinates.Y]
zc = intensities.coords[Coordinates.Z]
assert xc.size == NUMBER_SPOTS
assert yc.size == NUMBER_SPOTS
assert zc.size == NUMBER_SPOTS
# Assert that the physical coords align with their corresponding pixel coords
for spot in xc.features:
pixel_x = spot[Axes.X.value].data
physical_x = stack.xarray[Coordinates.X.value][pixel_x]
assert np.isclose(spot[Coordinates.X.value], physical_x)
for spot in yc.features:
pixel_y = spot[Axes.Y.value].data
physical_y = stack.xarray[Coordinates.Y.value][pixel_y]
assert np.isclose(spot[Coordinates.Y.value], physical_y)
# Assert that zc value is middle of z range
for spot in zc.features:
z_plane = spot[Axes.ZPLANE.value].data
physical_z = stack.xarray[Coordinates.Z.value][z_plane]
assert np.isclose(spot[Coordinates.Z.value], physical_z)
def test_tranfering_physical_coords_to_expression_matrix():
stack_shape = OrderedDict([(Axes.ROUND, 3), (Axes.CH, 2), (Axes.ZPLANE, 1),
(Axes.Y, 50), (Axes.X, 40)])
physical_coords = OrderedDict([(PhysicalCoordinateTypes.X_MIN, 1),
(PhysicalCoordinateTypes.X_MAX, 2),
(PhysicalCoordinateTypes.Y_MIN, 4),
(PhysicalCoordinateTypes.Y_MAX, 6),
(PhysicalCoordinateTypes.Z_MIN, 1),
(PhysicalCoordinateTypes.Z_MAX, 3)])
stack = factories.imagestack_with_coords_factory(stack_shape,
physical_coords)
codebook = factories.codebook_array_factory()
intensities = IntensityTable.synthetic_intensities(
codebook,
num_z=stack_shape[Axes.ZPLANE],
height=stack_shape[Axes.Y],
width=stack_shape[Axes.X],
n_spots=NUMBER_SPOTS)
intensities = intensity_table_coordinates. \
transfer_physical_coords_from_imagestack_to_intensity_table(stack, intensities)
# Check that error is thrown before target assignment
try:
intensities.to_expression_matrix()
except KeyError as e:
# Assert value error is thrown with right message
assert e.args[0] == "IntensityTable must have 'cell_id' assignments for each cell before " \
"this function can be called. See starfish.TargetAssignment.Label."
# mock out come cell_ids
cell_ids = random.sample(range(1, 20), NUMBER_SPOTS)
intensities[Features.CELL_ID] = (Features.AXIS, cell_ids)
expression_matrix = intensities.to_expression_matrix()
# Assert that coords were transferred
xc = expression_matrix.coords[Coordinates.X]
yc = expression_matrix.coords[Coordinates.Y]
zc = expression_matrix.coords[Coordinates.Z]
assert xc.size == len(set(cell_ids))
assert yc.size == len(set(cell_ids))
assert zc.size == len(set(cell_ids))
def test_save_expression_matrix():
codebook = codebook_array_factory()
intensities = IntensityTable.synthetic_intensities(
codebook,
num_z=3,
height=100,
width=100,
n_spots=10
)
# mock out come cell_ids
cell_ids = random.sample(range(1, 20), NUMBER_SPOTS)
intensities[Features.CELL_ID] = (Features.AXIS, cell_ids)
expression_matrix = intensities.to_expression_matrix()
# test all saving methods
expression_matrix.save("expression")
def test_intensity_table_serialization():
"""
Test that an IntensityTable can be saved to disk, and that when it is reloaded, the data is
unchanged
"""
# create an IntensityTable
data = np.zeros(100, dtype=np.float32).reshape(1, 5, 2, 2, 5)
image_stack = ImageStack.from_numpy_array(data)
intensities = IntensityTable.from_image_stack(image_stack)
# dump it to disk
tempdir = tempfile.mkdtemp()
filename = os.path.join(tempdir, 'test.nc')
intensities.save(filename)
# verify the data has not changed
loaded = intensities.load(filename)
assert intensities.equals(loaded)
def dummy_intensities() -> IntensityTable:
codebook = test_utils.codebook_array_factory()
intensities = IntensityTable.synthetic_intensities(
codebook,
num_z=10,
height=10,
width=10,
n_spots=5,
)
intensities[Coordinates.Z.value] = (Features.AXIS, [0, 1, 0, 1, 0])
intensities[Coordinates.Y.value] = (Features.AXIS, [10, 30, 50, 40, 20])
intensities[Coordinates.X.value] = (Features.AXIS, [50.2, 30.2, 60.2, 40.2, 70.2])
# remove target from dummy to test error messages
del intensities[Features.TARGET]
return intensities
def test_intensity_table_can_be_created_from_spot_attributes():
"""
This test creates an IntensityTable from spot attributes, and verifies that the size matches
what was requested and that the values are all zero.
"""
# input has two spots
spot_attributes = SpotAttributes(
pd.DataFrame(
data=np.array([[1, 1, 1, 1], [2, 2, 2, 1]]),
columns=[Indices.Z, Indices.Y, Indices.X, Features.SPOT_RADIUS]))
intensities = IntensityTable.empty_intensity_table(spot_attributes,
n_ch=1,
n_round=3)
assert intensities.sizes[Indices.CH] == 1
assert intensities.sizes[Indices.ROUND] == 3
assert intensities.sizes[Features.AXIS] == 2
assert np.all(intensities.values == 0)
def _load_data(fov_data, exp):
"""
Load a field of view dataset
Parameters
----------
fov_data : pd.DataFrame
Table of file locations for the spot results and mask label image
corresponding to the dataset to be viewed
exp : Experiment
Experiment file corresponding to the data analysis
Returns
-------
im_max_proj : np.ndarray
Image used for the spot detection
points : List[Dict]
List of dictionaries containing the points and target names for the
points to be rendered
mask_im : np.ndarray
Label image for the segmentation mask
"""
# Get the RNAscope images
im = exp[fov_data.fov_name].get_image("primary")
im_max_proj = np.squeeze(im.max_proj(Axes.ZPLANE).xarray.values)
# Get the spots
spots_file = fov_data.spot_file
it = IntensityTable.open_netcdf(spots_file)
points = _get_points(it, exp)
# Get the segmentation mask
mask_file = fov_data.mask_file
mask_im = io.imread(mask_file)
return im_max_proj, points, mask_im
def test_to_mermaid_dataframe():
"""
Creates a basic IntensityTable from an ImageStack and verifies that it can be dumped to disk
as a DataFrame which MERmaid can load. Does not explicitly load the DataFrame in MERmaid.
Verifies that the save function throws an error when target assignments are not present, which
are required by MERmaid.
"""
r, c, z, y, x = 1, 5, 2, 2, 5
data = np.zeros(100, dtype=np.float32).reshape(r, c, z, y, x)
image_stack = ImageStack.from_numpy_array(data)
intensities = IntensityTable.from_image_stack(image_stack)
# without a target assignment, should raise RuntimeError.
with pytest.raises(RuntimeError):
with TemporaryDirectory() as dir_:
intensities.save_mermaid(os.path.join(dir_, 'test.csv.gz'))
# assign targets
intensities[Features.TARGET] = (Features.AXIS, np.random.choice(list('ABCD'), size=20))
intensities[Features.DISTANCE] = (Features.AXIS, np.random.rand(20))
with TemporaryDirectory() as dir_:
intensities.save_mermaid(os.path.join(dir_, 'test.csv.gz'))
def test_synthetic_intensity_generation():
"""
Create a 2-spot IntensityTable of pixel size (z=3, y=4, x=5) from a codebook with 3 channels
and 2 rounds.
Verify that the constructed Synthetic IntensityTable conforms to those dimensions, and given
a known random seed, that the output spots decode to match a target in the input Codebook
"""
# set seed to check that codebook is matched. This seed generates 2 instances of GENE_B
np.random.seed(1)
codebook = test_utils.codebook_array_factory()
num_z, height, width = 3, 4, 5
intensities = IntensityTable.synthetic_intensities(codebook,
num_z=num_z,
height=height,
width=width,
n_spots=2)
# sizes should match codebook
assert intensities.sizes[Axes.ROUND] == 2
assert intensities.sizes[Axes.CH] == 3
assert intensities.sizes[Features.AXIS] == 2
# attributes should be bounded by the specified size
assert np.all(intensities[Axes.ZPLANE.value] <= num_z)
assert np.all(intensities[Axes.Y.value] <= height)
assert np.all(intensities[Axes.X.value] <= width)
# both codes should match GENE_B
assert np.array_equal(
np.where(intensities.values),
[
[0, 0, 1, 1], # two each in feature 0 & 1
[1, 2, 1, 2], # one each in channel 1 & 2
[1, 0, 1, 0]
], # channel 1 matches round 1, channel 2 matches round zero
)
def test_intensity_table_can_be_constructed_from_an_imagestack():
"""
ImageStack has enough information to create an IntensityTable without additional SpotAttributes.
Each feature is a pixel, and therefore the SpotAttributes can be extracted from the relative
locations.
"""
r, c, z, y, x = 1, 5, 2, 2, 5
data = np.zeros(100, dtype=np.float32).reshape(r, c, z, y, x)
image_stack = ImageStack.from_numpy_array(data)
intensities = IntensityTable.from_image_stack(image_stack)
# there should be 100 features
assert np.product(intensities.shape) == 100
# the max features should be equal to the array extent (2, 2, 5) minus one, since indices
# are being compared and python is zero based
# import pdb; pdb.set_trace()
assert np.max(intensities[Axes.ZPLANE.value].values) == z - 1
assert np.max(intensities[Axes.Y.value].values) == y - 1
assert np.max(intensities[Axes.X.value].values) == x - 1
# the number of channels and rounds should match the ImageStack
assert intensities.sizes[Axes.CH.value] == c
assert intensities.sizes[Axes.ROUND.value] == r
def test_take_max():
"""
Create two overlapping IntensityTables with differing number of spots and verify that
by concatenating them with the TAKE_MAX strategy we only include spots in the overlapping
section from the IntensityTable that had the most.
"""
it1 = create_intensity_table_with_coords(Area(min_x=0,
max_x=2,
min_y=0,
max_y=2),
n_spots=10)
it2 = create_intensity_table_with_coords(Area(min_x=1,
max_x=2,
min_y=1,
max_y=3),
n_spots=20)
concatenated = IntensityTable.concatanate_intensity_tables(
[it1, it2], overlap_strategy=OverlapStrategy.TAKE_MAX)
# The overlap section hits half of the spots from each intensity table, 5 from it1
# and 10 from i21. It2 wins and the resulting concatenated table should have all the
# spots from it2 (20) and 6 (one on the border) from it1 (6) for a total of 26 spots
assert concatenated.sizes[Features.AXIS] == 26
tmp = tempfile.gettempdir()
iss_nc = os.path.join(tmp, "iss.nc")
merfish_nc = os.path.join(tmp, "merfish.nc")
dartfish_nc = os.path.join(tmp, "dartfish.nc")
def curl(dest_path, link):
with open(dest_path, "wb") as fh:
fh.write(requests.get(link).content)
curl(iss_nc, iss_link)
curl(merfish_nc, merfish_link)
curl(dartfish_nc, dartfish_link)
iss_intensity_table = IntensityTable.load(iss_nc)
merfish_intensity_table = IntensityTable.load(merfish_nc)
dartfish_intensity_table = IntensityTable.load(dartfish_nc)
# EPY: END code
# EPY: START code
datasets = [
iss_intensity_table, merfish_intensity_table, dartfish_intensity_table
]
# EPY: END code
# EPY: START markdown
### Load Background Images
# EPY: END markdown
# EPY: START code
plt.title('Set minimum distance threshold')
from starfish import IntensityTable
distance_threshold = min_dist
psd = DetectPixels.PixelSpotDecoder(codebook=experiment.codebook,
metric='euclidean',
distance_threshold=distance_threshold,
magnitude_threshold=magnitude_threshold,
min_area=area_threshold[0],
max_area=area_threshold[1])
spot_intensities, results = psd.run(filtered_imgs)
spot_intensities = IntensityTable(
spot_intensities.where(spot_intensities[Features.PASSES_THRESHOLDS],
drop=True))
###################################################################################################
# Here, we:
#
# 1. Pick a rolony that was succesfully decoded to a gene.
# 2. Pull out the average pixel trace for that rolony.
# 3. Plot that pixel trace against the barcode of that gene.
#
# In order to assess, visually, how close decoded barcodes match their targets.
# reshape the spot intensity table into a RxC barcode vector
pixel_traces = spot_intensities.stack(traces=(Axes.ROUND.value, Axes.CH.value))
# extract dataframe from spot intensity table for indexing purposes
def test_tranfering_physical_coords_to_intensity_table():
stack_shape = OrderedDict([(Axes.ROUND, 3), (Axes.CH, 2),
(Axes.ZPLANE, 1), (Axes.Y, 50), (Axes.X, 40)])
physical_coords = OrderedDict([(PhysicalCoordinateTypes.X_MIN, 1),
(PhysicalCoordinateTypes.X_MAX, 2),
(PhysicalCoordinateTypes.Y_MIN, 4),
(PhysicalCoordinateTypes.Y_MAX, 6),
(PhysicalCoordinateTypes.Z_MIN, 1),
(PhysicalCoordinateTypes.Z_MAX, 3)])
stack = test_utils.imagestack_with_coords_factory(stack_shape, physical_coords)
codebook = test_utils.codebook_array_factory()
intensities = IntensityTable.synthetic_intensities(
codebook,
num_z=stack_shape[Axes.ZPLANE],
height=stack_shape[Axes.Y],
width=stack_shape[Axes.X],
n_spots=NUMBER_SPOTS
)
intensities = intensity_table_coordinates.\
transfer_physical_coords_from_imagestack_to_intensity_table(stack, intensities)
# Assert that new cords were added
xc = intensities.coords[Coordinates.X]
yc = intensities.coords[Coordinates.Y]
zc = intensities.coords[Coordinates.Z]
assert xc.size == NUMBER_SPOTS
assert yc.size == NUMBER_SPOTS
assert zc.size == NUMBER_SPOTS
physical_pixel_size_x = physical_coordinate_calculator._calculate_physical_pixel_size(
coord_min=physical_coords[PhysicalCoordinateTypes.X_MIN],
coord_max=physical_coords[PhysicalCoordinateTypes.X_MAX],
num_pixels=stack_shape[Axes.X])
physical_pixel_size_y = physical_coordinate_calculator._calculate_physical_pixel_size(
coord_min=physical_coords[PhysicalCoordinateTypes.Y_MIN],
coord_max=physical_coords[PhysicalCoordinateTypes.Y_MAX],
num_pixels=stack_shape[Axes.Y])
# Assert that the physical coords align with their corresponding pixel coords
for spot in xc.features:
pixel_x = spot[Axes.X.value].data
physical_x = spot[Coordinates.X.value].data
calculated_pixel = physical_cord_to_pixel_value(physical_x,
physical_pixel_size_x,
physical_coords[
PhysicalCoordinateTypes.X_MIN
])
assert np.isclose(pixel_x, calculated_pixel)
for spot in yc.features:
pixel_y = spot[Axes.Y.value].data
physical_y = spot[Coordinates.Y.value].data
calculated_pixel = physical_cord_to_pixel_value(physical_y,
physical_pixel_size_y,
physical_coords[
PhysicalCoordinateTypes.Y_MIN
])
assert np.isclose(pixel_y, calculated_pixel)
# Assert that zc value is middle of z range
for spot in zc.features:
physical_z = spot[Coordinates.Z.value].data
assert np.isclose(physical_coords[PhysicalCoordinateTypes.Z_MAX],
(physical_z * 2) - physical_coords[PhysicalCoordinateTypes.Z_MIN])