bsb.connectivity.geometric package¶

Submodules¶

bsb.connectivity.geometric.geometric_shapes module¶

class bsb.connectivity.geometric.geometric_shapes.Cone(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: GeometricShape

A cone, described in cartesian coordinates.

apex¶: The coordinates of the apex of the cone.

check_inside(points: ndarray[float])[source]¶

Check if the points given in input are inside the geometric shape.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool array with same length as points, containing whether the -ith point is inside the geometric shape or not.
Return type:: numpy.ndarray

find_mbb()[source]¶: Compute the minimum bounding box surrounding the shape.

generate_point_cloud(npoints: int)[source]¶

Generate a point cloud made by npoints points.

Parameters:: npoints (int) – The number of points to generate.
Returns:: a (npoints x 3) numpy array.
Return type:: numpy.ndarray

get_node_name()¶

get_volume()[source]¶

Get the volume of the geometric shape.

Returns:: The volume of the geometric shape.
Return type:: float

origin¶: The coordinates of the center of the cone’s base.

radius¶: The radius of the base circle.

rotate(r_versor: ndarray[float], angle: float)[source]¶

Rotate all the shapes around r_versor, which is a versor passing through the origin, by the specified angle.

Parameters:

r_versor (numpy.ndarray[float]) – A versor specifying the rotation axis.
angle (float) – the rotation angle, in radians.

translate(t_vector)[source]¶

Translate the geometric shape by the vector t_vector.

Parameters:: t_vector (numpy.ndarray) – The displacement vector

wireframe_points(nb_points_1=30, nb_points_2=30)[source]¶

Generate a wireframe to plot the geometric shape. If a sampling of points is needed (e.g. for sphere), the wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

Coordinate components of the wireframe

Return type:

tuple[numpy.ndarray[numpy.ndarray[float]]

class bsb.connectivity.geometric.geometric_shapes.Cuboid(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: GeometricShape

A rectangular parallelepiped, described in cartesian coordinates.

check_inside(points: ndarray[float])[source]¶

Check if the points given in input are inside the geometric shape.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool array with same length as points, containing whether the -ith point is inside the geometric shape or not.
Return type:: numpy.ndarray

find_mbb()[source]¶: Compute the minimum bounding box surrounding the shape.

generate_point_cloud(npoints: int)[source]¶

Generate a point cloud made by npoints points.

Parameters:: npoints (int) – The number of points to generate.
Returns:: a (npoints x 3) numpy array.
Return type:: numpy.ndarray

get_node_name()¶

get_volume()[source]¶

Get the volume of the geometric shape.

Returns:: The volume of the geometric shape.
Return type:: float

origin¶: The coordinates of the center of the barycenter of the bottom rectangle.

rotate(r_versor: ndarray[float], angle: float)[source]¶

Rotate all the shapes around r_versor, which is a versor passing through the origin, by the specified angle.

Parameters:

r_versor (numpy.ndarray[float]) – A versor specifying the rotation axis.
angle (float) – the rotation angle, in radians.

side_length_1¶: Length of one side of the base rectangle.

side_length_2¶: Length of the other side of the base rectangle.

top_center¶: The coordinates of the center of the barycenter of the top rectangle.

translate(t_vector: ndarray[float])[source]¶

Translate the geometric shape by the vector t_vector.

Parameters:: t_vector (numpy.ndarray) – The displacement vector

wireframe_points(**kwargs)[source]¶

Generate a wireframe to plot the geometric shape. If a sampling of points is needed (e.g. for sphere), the wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

Coordinate components of the wireframe

Return type:

tuple[numpy.ndarray[numpy.ndarray[float]]

class bsb.connectivity.geometric.geometric_shapes.Cylinder(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: GeometricShape

A cylinder, described in cartesian coordinates.

check_inside(points: ndarray[float])[source]¶

Check if the points given in input are inside the geometric shape.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool array with same length as points, containing whether the -ith point is inside the geometric shape or not.
Return type:: numpy.ndarray

find_mbb()[source]¶: Compute the minimum bounding box surrounding the shape.

generate_point_cloud(npoints: int)[source]¶

Generate a point cloud made by npoints points.

Parameters:: npoints (int) – The number of points to generate.
Returns:: a (npoints x 3) numpy array.
Return type:: numpy.ndarray

get_node_name()¶

get_volume()[source]¶

Get the volume of the geometric shape.

Returns:: The volume of the geometric shape.
Return type:: float

origin¶: The coordinates of the center of the bottom circle of the cylinder.

radius¶: The radius of the base circle.

rotate(r_versor: ndarray[float], angle: float)[source]¶

Rotate all the shapes around r_versor, which is a versor passing through the origin, by the specified angle.

Parameters:

r_versor (numpy.ndarray[float]) – A versor specifying the rotation axis.
angle (float) – the rotation angle, in radians.

top_center¶: The coordinates of the center of the top circle of the cylinder.

translate(t_vector: ndarray[float])[source]¶

Translate the geometric shape by the vector t_vector.

Parameters:: t_vector (numpy.ndarray) – The displacement vector

wireframe_points(nb_points_1=30, nb_points_2=30)[source]¶

Generate a wireframe to plot the geometric shape. If a sampling of points is needed (e.g. for sphere), the wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

Coordinate components of the wireframe

Return type:

tuple[numpy.ndarray[numpy.ndarray[float]]

class bsb.connectivity.geometric.geometric_shapes.Ellipsoid(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: GeometricShape

An ellipsoid, described in cartesian coordinates.

check_inside(points: ndarray[float])[source]¶

Check if the points given in input are inside the geometric shape.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool array with same length as points, containing whether the -ith point is inside the geometric shape or not.
Return type:: numpy.ndarray

find_mbb()[source]¶: Compute the minimum bounding box surrounding the shape.

generate_point_cloud(npoints: int)[source]¶

Generate a point cloud made by npoints points.

Parameters:: npoints (int) – The number of points to generate.
Returns:: a (npoints x 3) numpy array.
Return type:: numpy.ndarray

get_node_name()¶

get_volume()[source]¶

Get the volume of the geometric shape.

Returns:: The volume of the geometric shape.
Return type:: float

lambdas¶: The length of the three semi-axes.

origin¶: The coordinates of the center of the ellipsoid.

rotate(r_versor: ndarray[float], angle: float)[source]¶

Rotate all the shapes around r_versor, which is a versor passing through the origin, by the specified angle.

Parameters:

r_versor (numpy.ndarray[float]) – A versor specifying the rotation axis.
angle (float) – the rotation angle, in radians.

surface_point(theta, phi)[source]¶

Convert polar coordinates into their 3D location on the ellipsoid surface.

Parameters:

theta (float|numpy.ndarray[float]) – first polar coordinate in [0; 2*np.pi]
phi (float|numpy.ndarray[float]) – second polar coordinate in [0; np.pi]

Returns:

surface coordinates

Return type:

float|numpy.ndarray[float]

translate(t_vector: ndarray)[source]¶

Translate the geometric shape by the vector t_vector.

Parameters:: t_vector (numpy.ndarray) – The displacement vector

v0¶

v1¶

v2¶

wireframe_points(nb_points_1=30, nb_points_2=30)[source]¶

Generate a wireframe to plot the geometric shape. If a sampling of points is needed (e.g. for sphere), the wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

Coordinate components of the wireframe

Return type:

tuple[numpy.ndarray[numpy.ndarray[float]]

class bsb.connectivity.geometric.geometric_shapes.GeometricShape(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: ABC

Base class for geometric shapes.

abstractmethod check_inside(points: ndarray[float]) → ndarray[bool][source]¶

Check if the points given in input are inside the geometric shape.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool array with same length as points, containing whether the -ith point is inside the geometric shape or not.
Return type:: numpy.ndarray

check_mbox(points: ndarray[float])[source]¶

Check if the points given in input are inside the minimal bounding box.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool np.ndarray specifying whether each point of the input array is inside the minimal bounding box or not.
Return type:: numpy.ndarray

epsilon¶: Tolerance value to compare coordinates.

abstractmethod find_mbb()[source]¶: Compute the minimum bounding box surrounding the shape.

abstractmethod generate_point_cloud(npoints: int)[source]¶

Generate a point cloud made by npoints points.

Parameters:: npoints (int) – The number of points to generate.
Returns:: a (npoints x 3) numpy array.
Return type:: numpy.ndarray

get_node_name()¶

abstractmethod get_volume()[source]¶

Get the volume of the geometric shape.

Returns:: The volume of the geometric shape.
Return type:: float

abstractmethod rotate(r_versor: ndarray[float], angle: float)[source]¶

Rotate all the shapes around r_versor, which is a versor passing through the origin, by the specified angle.

Parameters:

r_versor (numpy.ndarray[float]) – A versor specifying the rotation axis.
angle (float) – the rotation angle, in radians.

abstractmethod translate(t_vector: ndarray[float])[source]¶

Translate the geometric shape by the vector t_vector.

Parameters:: t_vector (numpy.ndarray) – The displacement vector

type¶

Base implementation of all the different configuration attributes.

Call the factory function attr() instead.

abstractmethod wireframe_points(nb_points_1=30, nb_points_2=30)[source]¶

Generate a wireframe to plot the geometric shape. If a sampling of points is needed (e.g. for sphere), the wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

Coordinate components of the wireframe

Return type:

tuple[numpy.ndarray[numpy.ndarray[float]]

class bsb.connectivity.geometric.geometric_shapes.Parallelepiped(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: GeometricShape

A generic parallelepiped, described by the vectors (following the right-hand orientation) of the sides in cartesian coordinates.

check_inside(points: ndarray[float])[source]¶

Check if the points given in input are inside the geometric shape.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool array with same length as points, containing whether the -ith point is inside the geometric shape or not.
Return type:: numpy.ndarray

find_mbb()[source]¶: Compute the minimum bounding box surrounding the shape.

generate_point_cloud(npoints: int)[source]¶

Generate a point cloud made by npoints points.

Parameters:: npoints (int) – The number of points to generate.
Returns:: a (npoints x 3) numpy array.
Return type:: numpy.ndarray

get_node_name()¶

get_volume()[source]¶

Get the volume of the geometric shape.

Returns:: The volume of the geometric shape.
Return type:: float

origin¶: The coordinates of the left-bottom edge.

rotate(r_versor: ndarray[float], angle: float)[source]¶

Rotate all the shapes around r_versor, which is a versor passing through the origin, by the specified angle.

Parameters:

r_versor (numpy.ndarray[float]) – A versor specifying the rotation axis.
angle (float) – the rotation angle, in radians.

side_vector_1¶: The first vector identifying the parallelepiped (using the right-hand orientation: the thumb).

side_vector_2¶: The second vector identifying the parallelepiped (using the right-hand orientation: the index).

side_vector_3¶: The third vector identifying the parallelepiped (using the right-hand orientation: the middle finger).

translate(t_vector: ndarray[float])[source]¶

Translate the geometric shape by the vector t_vector.

Parameters:: t_vector (numpy.ndarray) – The displacement vector

wireframe_points(**kwargs)[source]¶

Generate a wireframe to plot the geometric shape. If a sampling of points is needed (e.g. for sphere), the wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

Coordinate components of the wireframe

Return type:

tuple[numpy.ndarray[numpy.ndarray[float]]

class bsb.connectivity.geometric.geometric_shapes.ShapesComposition(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: object

A collection of geometric shapes, which can be labelled to distinguish different parts of a neuron.

add_shape(shape: GeometricShape, labels: list[str])[source]¶

Add a geometric shape to the collection.

Parameters:

shape (GeometricShape) – A GeometricShape to add to the collection.
labels (list[str]) – A list of labels for the geometric shape to add.

compute_n_points() → list[int][source]¶

Compute the number of points to generate in a point cloud, using the dimension of the voxel specified in self._voxel_size.

Returns:: The number of points to generate.
Return type:: numpy.ndarray[int]

filter_by_labels(labels: list[str]) → ShapesComposition[source]¶

Filter the collection of shapes, returning only the ones corresponding to the given labels.

Parameters:: labels (list[str]) – A list of labels.
Returns:: A new ShapesComposition object containing only the shapes labelled as specified.
Return type:: ShapesComposition

find_mbb() → tuple[ndarray[float], ndarray[float]][source]¶

Compute the minimal bounding box containing the collection of shapes.

Returns:: The two corners individuating the minimal bounding box of the shapes collection.
Return type:: tuple(numpy.ndarray[float], numpy.ndarray[float])

generate_point_cloud() → ndarray[float] | None[source]¶

Generate a point cloud. The number of points to generate is determined automatically using the voxel size.

Returns:: A numpy.ndarray containing the 3D points of the cloud. If there are no shapes in the collection, it returns None.
Return type:: numpy.ndarray[float] | None

generate_wireframe(nb_points_1=30, nb_points_2=30) → tuple[list, list, list] | None[source]¶

Generate the wireframes of a collection of shapes. If a sampling of points is needed for certain shapes (e.g. for sphere), their wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

The x,y,z coordinates of the wireframe of each shape.

Return type:

tuple[list[numpy.ndarray[numpy.ndarray[float]]]] | None

get_mbb_max()[source]¶

Returns the top corner of the minimum bounding box containing the collection of shapes.

Returns:: The top corner individuating the minimal bounding box of the shapes collection.
Return type:: numpy.ndarray[float]

get_mbb_min()[source]¶

Returns the bottom corner of the minimum bounding box containing the collection of shapes.

Returns:: The bottom corner individuating the minimal bounding box of the shapes collection.
Return type:: numpy.ndarray[float]

get_node_name()¶

get_volumes() → list[float][source]¶

Compute the volumes of all the shapes.

Return type:: list[float]

inside_mbox(points: ndarray[float]) → ndarray[bool][source]¶

Check if the points given in input are inside the minimal bounding box of the collection.

Parameters:: points (numpy.ndarray) – An array of 3D points.
Returns:: A bool np.ndarray specifying whether each point of the input array is inside the minimal bounding box of the collection.
Return type:: numpy.ndarray[bool]

inside_shapes(points: ndarray[float]) → ndarray[bool] | None[source]¶

Check if the points given in input are inside at least in one of the shapes of the collection.

Parameters:: points (numpy.ndarray) – An array of 3D points.
Returns:: A bool numpy.ndarray specifying whether each point of the input array is inside the collection of shapes or not.
Return type:: numpy.ndarray[bool]

labels¶: List of lists of labels associated to each geometric shape.

shapes¶: List of GeometricShape that make up the neuron.

translate(t_vec: ndarray[float])[source]¶

Translate all the shapes in the collection by the vector t_vec. It also automatically translate the minimal bounding box.

Parameters:: t_vec (numpy.ndarray) – The displacement vector.

voxel_size¶: Dimension of the side of a voxel, used to determine how many points must be generated to represent the geometric shape.

class bsb.connectivity.geometric.geometric_shapes.Sphere(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: GeometricShape

A sphere, described in cartesian coordinates.

check_inside(points: ndarray[float])[source]¶

Check if the points given in input are inside the geometric shape.

Parameters:: points (numpy.ndarray) – A cloud of points.
Returns:: A bool array with same length as points, containing whether the -ith point is inside the geometric shape or not.
Return type:: numpy.ndarray

find_mbb()[source]¶: Compute the minimum bounding box surrounding the shape.

generate_point_cloud(npoints: int)[source]¶

Generate a point cloud made by npoints points.

Parameters:: npoints (int) – The number of points to generate.
Returns:: a (npoints x 3) numpy array.
Return type:: numpy.ndarray

get_node_name()¶

get_volume()[source]¶

Get the volume of the geometric shape.

Returns:: The volume of the geometric shape.
Return type:: float

origin¶: The coordinates of the center of the sphere.

radius¶: The radius of the sphere.

rotate(r_versor: ndarray[float], angle: float)[source]¶

Rotate all the shapes around r_versor, which is a versor passing through the origin, by the specified angle.

Parameters:

r_versor (numpy.ndarray[float]) – A versor specifying the rotation axis.
angle (float) – the rotation angle, in radians.

surface_function(theta, phi)[source]¶

Convert polar coordinates into their 3D location on the sphere surface.

Parameters:

theta (float|numpy.ndarray[float]) – first polar coordinate in [0; 2*np.pi]
phi (float|numpy.ndarray[float]) – second polar coordinate in [0; np.pi]

Returns:

surface coordinates

Return type:

float|numpy.ndarray[float]

translate(t_vector: ndarray[float])[source]¶

Translate the geometric shape by the vector t_vector.

Parameters:: t_vector (numpy.ndarray) – The displacement vector

wireframe_points(nb_points_1=30, nb_points_2=30)[source]¶

Generate a wireframe to plot the geometric shape. If a sampling of points is needed (e.g. for sphere), the wireframe is based on a grid of shape (nb_points_1, nb_points_2).

Parameters:

nb_points_1 (int) – number of points sampled along the first dimension
nb_points_2 (int) – number of points sampled along the second dimension

Returns:

Coordinate components of the wireframe

Return type:

tuple[numpy.ndarray[numpy.ndarray[float]]

bsb.connectivity.geometric.geometric_shapes.inside_mbox(points: ndarray[float], mbb_min: ndarray[float], mbb_max: ndarray[float])[source]¶

Check if the points given in input are inside the minimal bounding box.

Parameters:

points (numpy.ndarray) – An array of 3D points.
mbb_min (numpy.ndarray) – 3D point representing the lowest coordinate of the minimal bounding box.
mbb_max (numpy.ndarray) – 3D point representing the highest coordinate of the minimal bounding box.

Returns:

A bool np.ndarray specifying whether each point of the input array is inside the minimal bounding box or not.

Return type:

numpy.ndarray[bool]

bsb.connectivity.geometric.morphology_shape_intersection¶

class bsb.connectivity.geometric.morphology_shape_intersection.MorphologyToShapeIntersection(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: ConnectionStrategy

affinity¶: Ratio of apositions to keep over the total number of contact points.

connect(pre, post)[source]¶

Central method of each connection strategy. Given a pair of HemitypeCollection (one for each connection side), should connect cell population using the scaffold’s (available as self.scaffold) bsb.core.Scaffold.connect_cells() method.

Parameters:

presyn_collection (bsb.connectivity.strategy.HemitypeCollection) – presynaptic filtered cell population.
postsyn_collection (bsb.connectivity.strategy.HemitypeCollection) – postsynaptic filtered cell population.

get_node_name()¶

get_region_of_interest(chunk)[source]¶

Returns the list of chunks containing the potential postsynaptic neurons, based on a chunk containing the presynaptic neurons.

Parameters:: chunk (bsb.storage._chunks.Chunk) – Presynaptic chunk
Returns:: List of postsynaptic chunks
Return type:: list[bsb.storage._chunks.Chunk]

postsynaptic: Hemitype¶: Postsynaptic (target) neuron population.

pruning_ratio¶: Ratio of conections to keep over the total number of apositions.

bsb.connectivity.geometric.shape_morphology_intersection¶

class bsb.connectivity.geometric.shape_morphology_intersection.ShapeToMorphologyIntersection(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: ConnectionStrategy

affinity¶: Ratio of apositions to keep over the total number of contact points.

connect(pre, post)[source]¶

Parameters:

presyn_collection (bsb.connectivity.strategy.HemitypeCollection) – presynaptic filtered cell population.
postsyn_collection (bsb.connectivity.strategy.HemitypeCollection) – postsynaptic filtered cell population.

get_node_name()¶

get_region_of_interest(chunk)[source]¶

Returns the list of chunks containing the potential postsynaptic neurons, based on a chunk containing the presynaptic neurons.

Parameters:: chunk (bsb.storage._chunks.Chunk) – Presynaptic chunk
Returns:: List of postsynaptic chunks
Return type:: list[bsb.storage._chunks.Chunk]

presynaptic: Hemitype¶: Presynaptic (source) neuron population.

pruning_ratio¶: Ratio of conections to keep over the total number of apositions.

bsb.connectivity.geometric.shape_shape_intersection¶

class bsb.connectivity.geometric.shape_shape_intersection.ShapeHemitype(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: Hemitype

Class representing a population of cells to connect with a ConnectionStrategy.

These cells’ morphology is implemented as a ShapesComposition.

get_mbb(chunks, chunk_dimension)[source]¶

Get the list of minimal bounding box containing all cells in the ShapeHemitype.

Parameters:

chunks (list[bsb.storage._chunks.Chunk]) – List of chunks containing the cell types (see bsb.connectivity.strategy.Hemitype.get_all_chunks)
chunk_dimension (float) – Size of a chunk

Returns:

List of bounding boxes in the form [min_x, min_y, min_z, max_x, max_y, max_z] for each chunk containing cells.

Return type:

list[numpy.ndarray[float, float, float, float, float, float]]

get_node_name()¶

shapes_composition¶: Composite shape representing the Hemitype.

class bsb.connectivity.geometric.shape_shape_intersection.ShapeToShapeIntersection(*args, _parent=None, _key=None, **kwargs)[source]¶

Bases: ConnectionStrategy

affinity¶: Ratio of apositions to keep over the total number of contact points.

connect(pre, post)[source]¶

Parameters:

presyn_collection (bsb.connectivity.strategy.HemitypeCollection) – presynaptic filtered cell population.
postsyn_collection (bsb.connectivity.strategy.HemitypeCollection) – postsynaptic filtered cell population.

get_node_name()¶

get_region_of_interest(chunk)[source]¶

Returns the list of chunks containing the potential postsynaptic neurons, based on a chunk containing the presynaptic neurons.

Parameters:: chunk (bsb.storage._chunks.Chunk) – Presynaptic chunk
Returns:: List of postsynaptic chunks
Return type:: list[bsb.storage._chunks.Chunk]

postsynaptic: Hemitype¶: Postsynaptic (target) neuron population.

presynaptic: Hemitype¶: Presynaptic (source) neuron population.

pruning_ratio¶: Ratio of conections to keep over the total number of apositions.