spiketools.spatial.place.compute_trial_place_bins

spiketools.spatial.place.compute_trial_place_bins(spikes, position, timestamps, bins, start_times, stop_times, area_range=None, speed=None, min_speed=None, max_speed=None, time_threshold=None, trial_occupancy=None, flatten=False, orientation=None)

Compute the spatially binned spiking activity, across trials.

Parameters:

spikes1d array

Spike times, in seconds.

position1d or 2d array

Position values.

timestamps1d array

Timestamps, in seconds, corresponding to the position values.

binsint or list of [int, int]

The bin definition for dividing up the space. If 1d, can be integer. If 2d should be a list, defined as [number of x_bins, number of y_bins].

start_times, stop_times1d array

The start and stop times, in seconds, of each trial.

area_rangelist of list, optional

Edges of the area to bin, defined as [[x_min, x_max], [y_min, y_max]].

speed1d array, optional

Speed values corresponding to each position. Should be the same length as timestamps.

min_speed, max_speedfloat, optional

Minimum and/or maximum speed thresholds to apply. Any spikes with an associated speed below the minimum or above maximum are dropped.

time_thresholdfloat, optional

A maximum time threshold, per bin observation, to apply. If provided, any bin values with an associated time length above this value are dropped.

trial_occupancy2d or 3d array, optional

Computed occupancy across the space, across trials. If provided, used to normalize bin counts per trial.

flattenbool, optional, default: False

Whether the flatten the spatial bins per trial. Only used if position data are 2d.

orientation{‘row’, ‘column’}, optional

The orientation of the position data. If not provided, is inferred from the position data.

Returns:

place_bins_trial1d or 2d or 3d array

The spike activity per spatial bin, per trial. If flatten is True, for a 2d position input, the output is 2d, as [n_trial, n_bins]. Otherwise, for a 2d position input, the output is 3d, as [n_trials, n_ybins, n_xbins].

Examples

Compute spike activity, in 1d spatial bins across 2 trials:

>>> spikes = np.array([0.15, 0.22, 0.28, 0.41, 0.50, 0.65, 0.77, 0.81, 0.95])
>>> position = np.array([1.0, 3.5, 2.0, 1.5, 3.0, 3.5, 4.0, 5.0, 3.5, 2.5])
>>> timestamps = np.array([0.10, 0.20, 0.25, 0.35, 0.45, 0.55, 0.6, 0.7, 0.80, 0.95])
>>> bins = 2
>>> start_times, stop_times = np.array([0, 0.6]), np.array([0.4, 1.0])
>>> compute_trial_place_bins(spikes, position, timestamps, bins, start_times, stop_times)
array([[2., 1.],
       [3., 1.]])

Compute spike activity across trials, normalizing by trial-level occupancy:

>>> from spiketools.spatial.occupancy import compute_trial_occupancy
>>> trial_occ = compute_trial_occupancy(position, timestamps, bins, start_times, stop_times)
>>> compute_trial_place_bins(spikes, position, timestamps, bins,
...                          start_times, stop_times, trial_occupancy=trial_occ)
array([[10., 20.],
       [20.,  5.]])