`paleo` ¶

Modules:

Name	Description
`graph_edits`
`level2_graph`
`networkdelta`
`replay`
`skeletons`
`synapses`
`utils`

Classes:

Name	Description
`NetworkDelta`

Functions:

Name	Description
`compare_graphs`	Compare two graphs and return the differences.
`get_detailed_change_log`	Get a detailed change log for a root ID.
`get_metadata_table`	Retrieve metadata for a list of operations or root IDs.
`get_metaedits`	Combine edits into meta-edits based on shared nodes.
`get_operation_level2_edit`	Extract changes to the level2 graph for a specific operation.
`get_operations_level2_edits`	Extract changes to the level2 graph for a list of operations.
`get_root_level2_edits`	Extract changes to the level2 graph for all operations on a root.
`get_initial_graph`	Get the initial graph for a given `root_id`, including objects that could become
`get_node_aliases`	For a given supervoxel, get the node that it was part of at `stop_layer` for
`get_component_masks`	From a list of components, get a node by component boolean DataFrame of masks.
`get_nucleus_supervoxel`	Get the supervoxel corresponding to the nucleus of a neuron by looking it up
`get_nodes_aliases`	For a list of supervoxels, get all of the nodes at `stop_layer` that they were
`get_supervoxel_mappings`	For a set of supervoxels and edits, get a mapping between the supervoxels and
`get_changed_nodes`	From a set of edits, get the nodes that have changed (added or removed).
`get_used_node_ids`	Starting from an initial graph and a series of edits, get the nodes that are
`apply_edit`	Apply the edit described by the networkdelta to the graph.
`resolve_edit`	Apply the edit described by the networkdelta and return the connected component
`find_anchor_node`	Find the first anchor node that is in the graph.
`apply_edit_sequence`	Apply a sequence of edits to the graph in order, storing information about
`get_mutable_synapses`	Get all synapses that could have been part of this `root_id` across all states.
`map_synapses_to_sequence`	Map synapses (with level2 node information) to a sequence of level2 nodes.
`skeletonize_sequence`	Generate skeletons for a sequence of graphs.

`NetworkDelta(removed_nodes, added_nodes, removed_edges, added_edges, metadata={})` ¶

A class to represent a change to a network.

Parameters:

Name	Type	Description	Default
`removed_nodes`	`ndarray`	IDs of nodes that were removed by this operation.	required
`added_nodes`	`ndarray`	IDs of nodes that were added by this operation.	required
`removed_edges`	`ndarray`	Edges that were removed by this operation.	required
`added_edges`	`ndarray`	Edges that were added by this operation.	required
`metadata`	`dict`	A dictionary of arbitrary metadata about the operation.	`{}`

`compare_graphs(graph_before, graph_after, metadata=False)` ¶

Compare two graphs and return the differences.

Parameters:

Name	Type	Description	Default
`graph_before`	`Graph`	The graph before the operation. Can either be a tuple of (nodes, edges) stored as `np.ndarrays`, or just the edges as an `np.ndarray`.	required
`graph_after`	`Graph`	The graph after the operation. Can either be a tuple of (nodes, edges) stored as `np.ndarrays`, or just the edges as an `np.ndarray`.	required

Returns:

Type	Description
`NetworkDelta`	The differences between the two graphs.

`get_detailed_change_log(root_id, client, filtered=True)` ¶

Get a detailed change log for a root ID.

Parameters:

Name	Type	Description	Default
`root_id`	`int`	The root ID to get the change log for.	required
`client`	`CAVEclient`	The CAVEclient instance to use.	required
`filtered`	`bool`	Whether to filter the change log to only include changes which affect the final state of the root ID.	`True`

Returns:

Type	Description
`DataFrame`	A detailed change log for the root ID.

`get_metadata_table(operation_ids=None, root_ids=None, client=None)` ¶

Retrieve metadata for a list of operations or root IDs.

NOTE: aspirational, not yet implemented. To make this efficient, would probably need a server-side implementation

Parameters:

Name	Type	Description	Default
`operation_ids`	`list of int`	The operation IDs to retrieve metadata for.	`None`
`root_ids`	`list of int`	The root IDs to retrieve metadata for.	`None`
`client`	`CAVEclient`	The CAVEclient instance to use.	`None`

Returns:

Type Description

DataFrame

The metadata for the operations or root IDs. Metadata includes:

operation_id: The operation ID.
timestamp: The timestamp of the operation.
location: The approximate [x,y,z] centroid of the operation in nanometers.
volume_added: The volume added by the operation in cubic nanometers.
volume_removed: The volume removed by the operation in cubic nanometers.
n_added_nodes: The number of level2 nodes added by the operation.
n_removed_nodes: The number of level2 nodes removed by the operation.
n_modified_nodes: The number of level2 nodes modified by the operation.
n_added_edges: The number of level2 edges added by the operation.
n_removed_edges: The number of level2 edges removed by the operation.
n_modified_edges: The number of level2 edges modified by the operation.

`get_metaedits(networkdeltas)` ¶

Combine edits into meta-edits based on shared nodes.

Meta-edits are groups of one or more edits which affected a local region in the chunkedgraph. More specifically, they are defined as groups of edits which are connected components in a graph where nodes are edits and edges are shared nodes between edits.

Parameters:

Name	Type	Description	Default
`networkdeltas`	`dict[Integer, NetworkDelta]`	The changes to the level2 graph from each operation.	required

Returns:

Type	Description
`dict[Integer, NetworkDelta]`	The changes to the level2 graph from each meta-operation.
`dict[Integer, list[Integer]]`	A mapping of meta-operation IDs to the operation IDs that make them up.

`get_operation_level2_edit(operation_id, client, before_root_ids=None, after_root_ids=None, timestamp=None, point=None, radius=20000, metadata=False)` ¶

Extract changes to the level2 graph for a specific operation.

Parameters:

Name	Type	Description	Default
`operation_id`	`int`	The operation ID to extract changes for.	required
`client`	`CAVEclient`	The CAVEclient instance to use.	required
`before_root_ids`	`Optional[Collection[int]]`	The root ID(s) that were involved in the operation prior to it happening. If None, these will be looked up.	`None`
`after_root_ids`	`Optional[Collection[int]]`	The root ID(s) that were created by the operation. If None, these will be looked up.	`None`
`timestamp`	`Optional[datetime]`	The timestamp of the operation. Only used if `before_root_ids` is not provided. If None, this will be looked up.	`None`
`point`	`Optional[ndarray]`	The point to center the bounding box on. If None, will compare the entire level2 graphs of the objects before and after the operation.	`None`
`radius`	`Number`	The radius of the bounding box to use.	`20000`
`metadata`	`bool`	Whether to include metadata about the changes in the output.	`False`

Returns:

Type	Description
`NetworkDelta`	The changes to the level2 graph from this operation.

`get_operations_level2_edits(operation_ids, client, radius=20000, metadata=False, n_jobs=-1, verbose=True)` ¶

Extract changes to the level2 graph for a list of operations.

Parameters:

Name	Type	Description	Default
`operation_ids`	`Union[Collection[Integer], Integer]`	The operation ID(s) to extract changes for.	required
`client`	`CAVEclient`	The CAVEclient instance to use.	required
`radius`	`Number`	The radius of the bounding box to use.	`20000`
`metadata`	`bool`	Whether to include metadata about the changes in the output.	`False`
`n_jobs`	`int`	The number of jobs to run in parallel. If -1, will use all available cores.	`-1`
`verbose`	`bool`	Whether to display a progress bar.	`True`

Returns:

Type	Description
`dict[Integer, NetworkDelta]`	The changes to the level2 graph from these operations

`get_root_level2_edits(root_id, client, radius=20000, metadata=False, filtered=False, n_jobs=-1, verbose=True)` ¶

Extract changes to the level2 graph for all operations on a root.

Parameters:

Name	Type	Description	Default
`root_id`	`Integer`	The root ID to extract changes for.	required
`client`	`CAVEclient`	The CAVEclient instance to use.	required
`radius`	`Number`	The radius of the bounding box to use.	`20000`
`metadata`	`bool`	Whether to include metadata about the changes in the output.	`False`
`filtered`	`bool`	Whether to filter the change log to only include changes which affect the final state of the root ID.	`False`
`n_jobs`	`int`	The number of jobs to run in parallel. If -1, will use all available cores.	`-1`
`verbose`	`bool`	Whether to display a progress bar.	`True`

Returns:

Type	Description
`dict[Integer, NetworkDelta]`	The changes to the level2 graph from each operation

`get_initial_graph(root_id, client, verbose=True, return_as='networkx', n_jobs=-1)` ¶

Get the initial graph for a given root_id, including objects that could become part of the neuron in the future.

`get_node_aliases(supervoxel_id, client, stop_layer=2, return_as='list')` ¶

For a given supervoxel, get the node that it was part of at stop_layer for each timestamp.

`get_component_masks(components)` ¶

From a list of components, get a node by component boolean DataFrame of masks.

`get_nucleus_supervoxel(root_id, client)` ¶

Get the supervoxel corresponding to the nucleus of a neuron by looking it up in the soma table.

`get_nodes_aliases(supervoxel_ids, client, stop_layer=2, verbose=True, return_as='list')` ¶

For a list of supervoxels, get all of the nodes at stop_layer that they were part of across time.

`get_supervoxel_mappings(supervoxel_ids, edits, client, n_jobs=-1)` ¶

For a set of supervoxels and edits, get a mapping between the supervoxels and any level2 nodes they could have been part of across time.

`get_changed_nodes(edits)` ¶

From a set of edits, get the nodes that have changed (added or removed).

`get_used_node_ids(initial_graph, edits, anchor_nodes)` ¶

Starting from an initial graph and a series of edits, get the nodes that are used in at least one state of the graph throughout its history.

Parameters:

Name	Type	Description	Default
`initial_graph`	`Graph`	The initial graph to start from.	required
`edits`	`dict`	A dictionary of edits where the key is the `operation_id` and the value is a `NetworkDelta` object.	required
`anchor_nodes`	`list`	A list of nodes that are on the object of interest, used to pick the connected component to consider at each point in the history.	required

Returns:

Type	Description
	Nodes that are ever used in the history of the graph.

`apply_edit(graph, networkdelta)` ¶

Apply the edit described by the networkdelta to the graph.

`resolve_edit(graph, networkdelta, anchor_nodes)` ¶

Apply the edit described by the networkdelta and return the connected component containing the anchor node.

`find_anchor_node(graph, anchor_nodes)` ¶

Find the first anchor node that is in the graph.

`apply_edit_sequence(graph, edits, anchor_nodes, return_graphs=False, include_initial=True, verbose=True)` ¶

Apply a sequence of edits to the graph in order, storing information about intermediate states.

`get_mutable_synapses(root_id, edits, client, sides='both', synapse_table=None, remove_self=True, verbose=False, n_jobs=-1)` ¶

Get all synapses that could have been part of this root_id across all states.

`map_synapses_to_sequence(synapses, nodes_by_state, side='pre')` ¶

Map synapses (with level2 node information) to a sequence of level2 nodes.

`skeletonize_sequence(graphs_by_state, client=None, root_id=None, root_point=None, level2_data=None)` ¶

Generate skeletons for a sequence of graphs.

paleo ¶

NetworkDelta(removed_nodes, added_nodes, removed_edges, added_edges, metadata={}) ¶

compare_graphs(graph_before, graph_after, metadata=False) ¶

get_detailed_change_log(root_id, client, filtered=True) ¶

get_metadata_table(operation_ids=None, root_ids=None, client=None) ¶

get_metaedits(networkdeltas) ¶

get_operation_level2_edit(operation_id, client, before_root_ids=None, after_root_ids=None, timestamp=None, point=None, radius=20000, metadata=False) ¶

get_operations_level2_edits(operation_ids, client, radius=20000, metadata=False, n_jobs=-1, verbose=True) ¶

get_root_level2_edits(root_id, client, radius=20000, metadata=False, filtered=False, n_jobs=-1, verbose=True) ¶

get_initial_graph(root_id, client, verbose=True, return_as='networkx', n_jobs=-1) ¶

get_node_aliases(supervoxel_id, client, stop_layer=2, return_as='list') ¶

get_component_masks(components) ¶

get_nucleus_supervoxel(root_id, client) ¶

get_nodes_aliases(supervoxel_ids, client, stop_layer=2, verbose=True, return_as='list') ¶

get_supervoxel_mappings(supervoxel_ids, edits, client, n_jobs=-1) ¶

get_changed_nodes(edits) ¶

get_used_node_ids(initial_graph, edits, anchor_nodes) ¶

apply_edit(graph, networkdelta) ¶

resolve_edit(graph, networkdelta, anchor_nodes) ¶

find_anchor_node(graph, anchor_nodes) ¶

apply_edit_sequence(graph, edits, anchor_nodes, return_graphs=False, include_initial=True, verbose=True) ¶

get_mutable_synapses(root_id, edits, client, sides='both', synapse_table=None, remove_self=True, verbose=False, n_jobs=-1) ¶

map_synapses_to_sequence(synapses, nodes_by_state, side='pre') ¶

skeletonize_sequence(graphs_by_state, client=None, root_id=None, root_point=None, level2_data=None) ¶

`paleo` ¶

`NetworkDelta(removed_nodes, added_nodes, removed_edges, added_edges, metadata={})` ¶

`compare_graphs(graph_before, graph_after, metadata=False)` ¶

`get_detailed_change_log(root_id, client, filtered=True)` ¶

`get_metadata_table(operation_ids=None, root_ids=None, client=None)` ¶

`get_metaedits(networkdeltas)` ¶

`get_operation_level2_edit(operation_id, client, before_root_ids=None, after_root_ids=None, timestamp=None, point=None, radius=20000, metadata=False)` ¶

`get_operations_level2_edits(operation_ids, client, radius=20000, metadata=False, n_jobs=-1, verbose=True)` ¶

`get_root_level2_edits(root_id, client, radius=20000, metadata=False, filtered=False, n_jobs=-1, verbose=True)` ¶

`get_initial_graph(root_id, client, verbose=True, return_as='networkx', n_jobs=-1)` ¶

`get_node_aliases(supervoxel_id, client, stop_layer=2, return_as='list')` ¶

`get_component_masks(components)` ¶

`get_nucleus_supervoxel(root_id, client)` ¶

`get_nodes_aliases(supervoxel_ids, client, stop_layer=2, verbose=True, return_as='list')` ¶

`get_supervoxel_mappings(supervoxel_ids, edits, client, n_jobs=-1)` ¶

`get_changed_nodes(edits)` ¶

`get_used_node_ids(initial_graph, edits, anchor_nodes)` ¶

`apply_edit(graph, networkdelta)` ¶

`resolve_edit(graph, networkdelta, anchor_nodes)` ¶

`find_anchor_node(graph, anchor_nodes)` ¶

`apply_edit_sequence(graph, edits, anchor_nodes, return_graphs=False, include_initial=True, verbose=True)` ¶

`get_mutable_synapses(root_id, edits, client, sides='both', synapse_table=None, remove_self=True, verbose=False, n_jobs=-1)` ¶

`map_synapses_to_sequence(synapses, nodes_by_state, side='pre')` ¶

`skeletonize_sequence(graphs_by_state, client=None, root_id=None, root_point=None, level2_data=None)` ¶