clustering.py

This script contains classes and functions for implementing synaptic clustering in the build step of the simulation.

class L5NeuronSimulation.clustering.Cell(n_syns)[source]

A class used to an input cell within a functional group.

n_syns

number of synapses from the cell

Type: int

syn_segs

list of Segments where the cell’s synapses should be placed

Type: list

num_set

number of synapses that the build step has already created. Functions as a way to iterate through synapses.

Type: int

add_syn(seg: pandas.Series)[source]: Adds the given segment to syn_segs

get_seg()[source]: Returns the next synapse to set

add_syn(seg)[source]

Adds the given segment to syn_segs

Parameters: seg (pandas.Series) – represents the segment to added to syn_segs

get_seg()[source]: Returns the next Segment where a synapses should be created with this cell as the presynaptic cell.

class L5NeuronSimulation.clustering.Cluster(center, clustering_func, group_segs)[source]

A class used to represent a cluster of synapses within a functional group.

center

the segment that the cluster is centered around

Type: pd.Series

n_syns

number of synapses attached to the cluster

Type: int

cluster_segs

contains the segments that are within the cluster

Type: pd.DataFrame

random_seg() → pd.Series[source]: Returns a random segment in cluster_segs

random_seg()[source]: Returns a random segment in cluster_segs and increases the count of synapses in the cluster

class L5NeuronSimulation.clustering.FunctionalGroup(seg_list, center, num_cells, num_clusters, name, start_id, grouping_func, clustering_func, syn_per_cell=[2, 8])[source]

A class used to represent a functional group of input cells. Each functional group has some number of clusters that it places its cells’ synapses into.

center

the segment that the functional group is centered around

Type: pd.Series

n_cells

number of cells

Type: int

syn_per_cell

list of length two that contains the lower and upper boundary for the number of synapses from each cell. [low, high]

Type: list

n_clusters

number of clusters

Type: int

name

the name associated with the group. Used in the build step of the simulation for proper assignment.

Type: str

start_id

the node_id that is associated with the first cell in the cells list

Type: int

clustering_func

how segments are chosen for the clusters (formatted like make_seg_sphere above)

Type: func

group_segs

DataFrame of segments that fall within the group

Type: pd.DataFrame

cells

list of Cell objects that belong to the group

Type: list

clusters

list of Cluster objects that belong to the group

Type: list

make_clusters()[source]: Creates the group’s clusters

create_cells()[source]: Creates the group’s cells

create_cells()[source]: Creates the group’s cells. Gives each cell and random number of synapses within syn_per_cell. Distributes each cell’s synapses onto the clusters.

make_clusters()[source]: Initializes the clusters of the group. Cluster centers are randomly selected (uniquely) from group_segs.

class L5NeuronSimulation.clustering.Segment(bmtk_id, x)[source]

Simple class that represents a single segment in BMTK.

bmtk_id

the id that BMTK associates with the section (not the section id)

Type: int

x

distance along the section

Type: float

L5NeuronSimulation.clustering.calc_dist(p1, p2)[source]

Returns the distance between two points.

Parameters

p1 (list) – [x, y, z] representation of the first coordinate
p2 (list) – [x, y, z] representation of the second coordinate

Returns

The distance between the two given points

Return type

float

L5NeuronSimulation.clustering.make_seg_sphere(segments, center, radius=50)[source]

Returns segments within a sphere of the given radius centered at the given center.

Parameters

segments (pandas.DataFrame) – DataFrame with each row representing a segment that can be in the sphere
center (pandas.Series) – Series representing the segment that is the center of the created sphere
radius (int, optional) – The radius of sphere to be made in um (default is 50)

Returns

DataFrame contiaining the subset of segments that is within the sphere

Return type

pandas.DataFrame

L5NeuronSimulation.clustering.plot_group(segs, group, ax, psoma=[0, 0, 0])[source]

Plots the given group’s center, cells, and clusters on the given plt axes.

Parameters

segs (pd.DataFrame) – all segments that can possibly be plotted
group (FunctionalGroup) – the group to be plotted
ax (Axes3d) – what to plot on
psoma (list, optional) – used to shift the coordinates of the segments to match the morphology plot. formatted as [x, y, z] (default is [0, 0, 0])