pcg_skel

Basic functions for generating chunk-resolution skeletons from a pychunkedgraph.

pcg_graph(root_id, client, cv=None, return_l2dict=False, nan_rounds=10, require_complete=False, level2_graph=None)

Compute the level 2 spatial graph (or mesh) of a given root id using the l2cache.

Parameters:

• root_id (int) –

  Root id of a segment

• client (caveclient) –

  Initialized CAVEclient for the dataset.

• cv (CloudVolume, default: None ) –

  Initialized CloudVolume object for the dataset. This does not replace the caveclient, but a pre-initizialized cloudvolume can save some time during batch processing.

• return_l2dict (bool, default: False ) –

  If True, returns the mappings between l2 ids and vertices.

• nan_rounds (int, default: 10 ) –

  If vertices are missing (or not computed), this sets the number of iterations for smoothing over them.

• require_complete (bool, default: False ) –

  If True, raise an Exception if any vertices are missing from the cache.

• level2_graph (ndarray, default: None ) –

  Level 2 graph for the root id as returned by client.chunkedgraph.level2_chunk_graph. A list of lists of edges between level 2 chunks, as defined by their chunk ids. If None, will query the chunkedgraph for the level 2 graph. Optional, by default None.

Returns:

• mesh ( Mesh ) –

  Object with a vertex for every level 2 id and edges between all connected level 2 chunks.

• l2dict ( (dict, optional) ) –

  Dictionary with keys as level 2 ids and values as mesh vertex index. Optional, only returned if return_l2dict is True.

• l2dict_reverse ( (dict, optional) ) –

  Dictionary with keys as mesh vertex indices and values as level 2 id. Optional, only returned if return_l2dict is True.

pcg_skeleton(root_id, client, datastack_name=None, cv=None, invalidation_d=7500, return_mesh=False, return_l2dict=False, return_l2dict_mesh=False, root_point=None, root_point_resolution=None, collapse_soma=False, collapse_radius=7500, nan_rounds=10, require_complete=False, level2_graph=None)

Produce a skeleton from the level 2 graph.

Parameters:

• root_id (int) –

  Root id of a segment

• client (CAVEclientFull) –

  Initialized CAVEclient for the dataset.

• datastack_name (string, default: None ) –

  If client is None, initializes a CAVEclient at this datastack.

• cv (CloudVolume, default: None ) –

  Initialized CloudVolume object for the dataset. This does not replace the caveclient, but a pre-initizialized cloudvolume can save some time during batch processing.

• invalidation_d (int, default: 7500 ) –

  Distance (in nanometers) for TEASAR skeleton invalidation.

• return_mesh (bool, default: False ) –

  If True, returns the mesh graph as well as the skeleton.

• return_l2dict (bool, default: False ) –

  If True, returns the mappings between l2 ids and skeleton vertices.

• return_l2dict_mesh (bool, default: False ) –

  If True, returns mappings between l2 ids and mesh graph vertices.

• root_point (ArrayLike, default: None ) –

  3-element list or array with the x,y,z location of the root point. If None, the most distant tip is set to root.

• root_point_resolution (ArrayLike, default: None ) –

  3-element list or array with the x,y,z resolution of the root point, in nanometers per voxel dimension.

• collapse_soma (bool, default: False ) –

  If True, collapse nearby vertices into the root point.

• collapse_radius (int, default: 7500 ) –

  Distance (in nanometers) for soma collapse.

• nan_rounds (int, default: 10 ) –

  If vertices are missing (or not computed), this sets the number of iterations for smoothing over them.

• require_complete (bool, default: False ) –

  If True, raise an Exception if any vertices are missing from the cache.

• level2_graph (ndarray, default: None ) –

  Level 2 graph for the root id as returned by client.chunkedgraph.level2_chunk_graph. A list of lists of edges between level 2 chunks, as defined by their chunk ids. If None, will query the chunkedgraph for the level 2 graph. Optional, by default None.

Returns:

• sk ( Skeleton ) –

  Skeleton for the root id

• mesh ( (Mesh, optional) ) –

  Mesh graph that the skeleton is based on, only returned if return_mesh is True.

• (l2dict_skel, l2dict_reverse): (dict, dict), optional –

  Dictionaries mapping l2 ids to skeleton vertices and skeleton vertices to l2 ids, respectively. Only returned if return_l2dict is True.

• (l2dict_mesh, l2dict_mesh): (dict, dict), optional –

  Dictionaries mapping l2 ids to mesh graph vertices and mesh_graph vertices to l2 ids, respectively. Only returned if return_l2dict is True.

pcg_skeleton_direct(vertices, edges, invalidation_d=DEFAULT_INVALIDATION_D, root_point=None, collapse_soma=False, collapse_radius=DEFAULT_COLLAPSE_RADIUS, root_id=None)

Produce a skeleton from an already-computed l2graph. This is effectively a wrapper for meshparty skeletonize with a consistent set of parameters and format.

Parameters:

• vertices (array) –

  Array of vertices for the mesh graph.

• edges (array) –

  Array of edges for the mesh graph.

• invalidation_d (int, default: DEFAULT_INVALIDATION_D ) –

  Distance (in nanometers) for TEASAR skeleton invalidation.

• root_point (array, default: None ) –

  3-element list or array with the x,y,z location of the root point in same units as vertices. If None, the most distant tip is set to root.

• collapse_soma (bool, default: False ) –

  If True, collapse nearby vertices into the root point.

• collapse_radius (int, default: DEFAULT_COLLAPSE_RADIUS ) –

  Distance (in nanometers) for soma collapse.

• root_id (int, default: None ) –

  Root id of the segment, used in metadata. Optional, by default None.

• level2_graph (ndarray) –

  Level 2 graph for the root id as returned by client.chunkedgraph.level2_chunk_graph. A list of lists of edges between level 2 chunks, as defined by their chunk ids. If None, will query the chunkedgraph for the level 2 graph. Optional, by default None.

Returns:

• sk ( Skeleton ) –

  Skeleton for the l2graph.

pcg_meshwork(root_id, datastack_name=None, client=None, cv=None, root_point=None, root_point_resolution=None, collapse_soma=False, collapse_radius=DEFAULT_COLLAPSE_RADIUS, synapses=None, synapse_table=None, remove_self_synapse=True, live_query=False, timestamp=None, invalidation_d=DEFAULT_INVALIDATION_D, require_complete=False, metadata=False, synapse_partners=False, synapse_point_resolution=[1, 1, 1], synapse_representative_point_pre='ctr_pt_position', synapse_representative_point_post='ctr_pt_position', level2_graph=None)

Generate a meshwork file based on the level 2 graph.

Parameters:

• root_id (int) –

  Root id of an object in the pychunkedgraph.

• datastack_name (str or None, default: None ) –

  Datastack name to use to initialize a client, if none is provided. By default None.

• client (CAVEclientFull or None, default: None ) –

  Initialized CAVE client. If None is given, will use the datastack_name to create one. By default None

• cv (CloudVolume or None, default: None ) –

  Initialized cloudvolume. If none is given, the client info will be used to create one. By default None

• root_point (array - like or None, default: None ) –

  3 element xyz location for the location to set the root in units set by root_point_resolution, by default None. If None, a distal tip is selected.

• root_point_resolution (array - like, default: None ) –

  Resolution in euclidean space of the root_point, by default [4, 4, 40]

• collapse_soma ((bool, optional), default: False ) –

  If True, collapses vertices within a given radius of the root point into the root vertex, typically to better represent primary neurite branches. Requires a specified root_point. Default if False.

• collapse_radius (float, default: DEFAULT_COLLAPSE_RADIUS ) –

  Max distance in euclidean space for soma collapse. Default is 10,000 nm (10 microns).

• synapses ('pre', 'post', 'all', True, or None, default: None ) –

  If not None, queries the synapse_table for presynaptic synapses (if 'pre'), postsynaptic sites (if 'post'), or both (if 'all' or True). By default None

• synapse_table (str, default: None ) –

  Name of the synapse table to query if synapses are requested, by default None

• remove_self_synapse (bool, default: True ) –

  If True, filters out synapses whose pre- and postsynaptic root ids are the same neuron, by default True

• live_query (bool, default: False ) –

  If True, expect a timestamp for querying at a give point in time. Otherwise, use the materializatio set by the client. Optional, by default False.

• timestamp (Optional[datetime], default: None ) –

  If set, acts as the time at which all root ids and annotations are found at.

• invalidation_d (int, default: DEFAULT_INVALIDATION_D ) –

  Invalidation radius in hops for the mesh skeletonization along the chunk adjacency graph, by default 3

• require_complete (bool, default: False ) –

  If True, raise an Exception if any vertices are missing from the cache, by default False

• metadata (bool, default: False ) –

  If True, adds metadata to the meshwork annotations. By default False.

• synapse_partners (bool, default: False ) –

  If True, includes the partner root id to the synapse annotation. By default False, because partner roots can change across time.

• synapse_point_resolution (array - like, default: [1, 1, 1] ) –

  Resolution in euclidean space of the synapse points, by default None. If None, the resolution will be the default of the synapse table.

• synapse_representative_point_pre (str, default: 'ctr_pt_position' ) –

  If set, uses the specified column in the synapse table for the pre-synaptic points. By default 'ctr_pt_position'.

• synapse_representative_point_post (str, default: 'ctr_pt_position' ) –

  If set, uses the specified column in the synapse table for the post-synaptic points. By default 'ctr_pt_position'.

Returns:

• Meshwork –

  Meshwork object with skeleton based on the level 2 graph. See documentation for details.