Inconsistency of area_range and optimal choice of bin size

[claire98han]

I noticed that there are different area_range used across analyses so just want to double-check on this.

Expectations:

• nwbfile.acquisition['boundaries']['x_range'] and nwbfile.acquisition['boundaries']['z_range'] represent the x_boundary and z_boundary of the arena, which is fixed and identical in all nwb files.

Issues:

• Player positions during the navigation period fall outside of the x_boundary and z_boundary range. See example in THF-wv004-0, where x_boundary, y_boundary = array([349.8549, 419.8549]), array([308. , 409.3358]). But from x_range, z_range = get_pos_ranges(positions_nav), x_range, z_range=[367.1234, 401.8272], [285.0, 433.0].

• z_range is defined as nwbfile.acquisition['boundaries']['z_range'] + np. array([-10, 10]) in script, nwbfile.acquisition['boundaries']['z_range'] + np. array([-15, 15]) in 02-PlaceCells.ipynb. For place cell analysis, because the idea is to include data from the navigation period only, the buffer for tower transportation is not needed.

@TomDonoghue - could you please double-check these? Thanks!

[claire98han]

For the optimal choice of bin size, I'm thinking about defining bins such that each bin is a square. Currently, we are using bins = [6, 10] and chest_bins = [5, 7] which gives each bin a rectangular shape. Based on the range of the arena and chest_arena, it seems reasonable to have a 10-by-10 bin size.

[TomDonoghue]

Copying in notes from #13, as it's relevant here:

02-PlaceCells & 03-SpatialTargetCells: I don't think we need to add a buffer to area_range for the actual analysis, even though it would be helpful to add it for visualization purposes.
Related to unit_analysis.py, line 201:
area_range should be the chest area [[360, 410], [320, 400]] instead of the whole field range, right?

Bin sizes:

• [5, 7] for all bins (place analysis)
• [3, 5] for chest bins (target analysis)