Iris Calliope

Edit LL commit ES: unselecting lambda and bregma with middle click in RoiQuest now possible

Edit by LL, commit by ES: when using combined grid and eeg template roiquest moves only first 122 rois (eeg rois and lambda bregma)

ES: new template combining grid and eeg rings based on lambda bregma

rd: added for WF the automatic determination of psmooth window depending on the imaging rate (e.g. 100hz and 25hz require different windows in frame number)

Fix: save changes to modified Adata after 'Propagate and overwrite' ROIs using ROIquest -SM

LL - fixed a bug in rois2struct.m where activity for 2channel WF was not properly deleted

LL - Add F0 in proj_meta

Now generate_proj_meta.m saves also F0 (baseline fluorescence) for each ROI in ROIinfo(i).F0 (or .F0_ch525 or .F0_ch610)
This happens for WF and 2P data.
Changed generate_proj_meta.m for directly saving F0 and get_smoothed_ROIs.m for F0 in 2P data

"inhomogenous arrays" error when writing to matfile was caused by numpy version (we were at version 1.24.4, recommendations is numpy>=1.13.0, < 1.24). Reverting to numpy 1.23.5 resovled the error. Solution is to save to RAM instead of to a temp matfile, making the process independent of numpy versions.

removed the one layer case added on 08.09.2026 again.

Added a special case to deal with stacks that only have one z layer.

Problem: run_fissa was called with export_to_matfile, which routes the results through Python's savemat; for a single-layer/single-trial stack, FISSA returns per-ROI object arrays with differing component counts, and savemat crashed trying to pack those arrays into one homogeneous array (the (5,1) + inhomogeneous part ValueError).

Fix: branch when nLayers==1: instead of exporting to a matfile, it reads FISSA's returned tuple directly in MATLAB, pulling each ROI's separated neuron trace (row 1 of its 5×nFrames block) and storing it in activity_fissa_corrected. The existing 4-layer pipeline behaves exactly as before.