Repository navigation
pyBer 0.59
Neurophotometrics recordings. Preprocessing reads Neurophotometrics (FP3001/FP3002) CSV files. Each region becomes a channel: green at 470 nm and red at 560 nm, with the 415 nm reference interpolated onto the signal frames. Split per-LED files are merged. Changing digital columns and TTL bits become triggers, and event lists in the same folder are matched to the right clock. The CLI and folder discovery recognize the format.
Event lists without durations. A table whose columns are behaviors and whose rows are event times loads directly with the default file type, in any time order. Choose Times in file are: Onsets / Offsets to align on either edge. Before/after comparisons, duration filters and the behavior panels handle these time-only events: the panels show rate and count, and the statistics show n and rate per minute.
Events drawer. One Events button opens or closes the behavior panel. Behaviors are loaded with Add events. Click the small + in a behavior chip, or in an Events cell of the comparison table, to list the single events. You can hide an event from plots and metrics, remove it or restore it; source files are never changed. Checking a behavior switches the PSTH alignment to that behavior.
Groups. In Group scope, the metric panels compare groups, one point per animal. Choose the representation under PSTH → Metrics → Group comparison or by right-clicking a metric panel: Post − Pre (the change), Post only, Pre only, or Pre and Post side by side with a line joining each animal. The statistics follow the choice. Between groups: Welch t or Mann-Whitney U for two groups, ANOVA or Kruskal-Wallis with Holm-corrected pairwise tests for three or more; significant pairs are marked with brackets. In the paired view, each group gets a paired Pre vs Post test (paired t or Wilcoxon signed-rank, Holm across groups) and the change is tested between groups. Summary style draws the same numbers as points with median and mean ± SEM, bars, or box plots. Groups with fewer than 3 animals are described, not tested. Results are exported as *_groups.csv, *_group_pairwise.csv, *_group_rows.csv and, for the paired view, *_group_paired.csv. Use the eye on a group chip, or the Hide/show groups menu, to leave a group out of the visuals. The group legend moves to the part of the PSTH that hides the least of the curves; right-click the PSTH to hide it.
Baseline and windows. Type any value in the event baseline and metric window fields, before or after the event; entries are no longer limited or blocked while typing. Baseline suggestions are computed only when you click Auto beside the baseline fields, and nothing changes until you pick a window.
Display. Heatmaps trim rows to their shared time span and fill interior gaps for display only; statistics and exports keep the original data. The part of the time axis without heatmap rows keeps the plot background instead of a gray block. The color scale has no frame and is exactly as tall as the heatmap. The PSTH behavior panels default to bout duration and cumulative duration. pyBer opens with the Events drawer closed and the event marker off. Drawers adapt to their width instead of clipping, and the behavior comparison takes less space.
Export default. Preprocessing exports only the processed output by default (first launch and Reset defaults). Raw, isobestic, DIO and baseline columns remain available as options. The CLI still exports every column.
Download pyBer-windows.exe for the desktop app or pyBer-cli-windows.exe for the command-line tools. SHA256SUMS.txt contains checksums for both executables.
Full Changelog: v0.58.1...v0.59