CLICKABLE HELP
Quick guides
Use the question marks beside controls to jump to the choice you are making. Start with the source data, then check the time window before reading a result.
Check a new recording
Before interpreting a rhythm, make sure the imported time axis and measurements match the experiment.
- Check the source name and sample interval, then compare the dates shown on the plots with the acquisition log.
- Look at the quality control card for gaps and at the time series for unexpected flat or empty stretches.
- Confirm there are enough days for the question you want to ask. A clean import alone does not establish rhythmicity.
Need to rearrange columns or times? See the trace file guide.
Choose the analysis settings
The file's sampling interval, the analysis bin, the fold period and the period search have different jobs.
A protocol preset loads a starting set of settings; check every value against the actual experiment before interpreting a result.
Enter the lighting schedule
“Lights on” and “Lights off” are clock hours from midnight: 7 means 07:00 and 19.5 means 19:30.
For a schedule that changes, enter one line per change: start date, lights-on hour, lights-off hour.
Or turn on Edit lighting above the actogram and click a clock cell. The light/dark strip shows the proposed cycle: drag its on/off markers or enter exact times, then apply. The change begins on that date and lasts until the next dated change. Click an existing change date to remove it. Nothing is saved until you press Apply.
2026-05-01, 7, 19
2026-05-08, 0, 0
Equal on/off hours mean constant darkness. The optional light–dark and constant-dark date ranges in the sidebar override the figure background for those days. Use the recorded schedule; elapsed-hour files have no real calendar dates to align to it.
Exclude known bad intervals
Enter one full start and end timestamp per line, separated by a comma.
2026-05-03 08:00, 2026-05-03 10:00
2026-05-06 12:00, 2026-05-06 13:30
Exclusions leave the original file untouched. For whole days, you can also turn on “Brush day range” and assign the brushed days to an exclusion. Recheck the remaining duration before fitting a period.
Choose an actogram or time series
Like a calendar laid out one day per row, an actogram shows when activity occurs from day to day. A time series lays the same measurements along one continuous time axis.
- In a double-plotted view, the next day is repeated beside each day so a pattern crossing midnight stays visible.
- Choose the Time series tab in Temporal Structure to see the same recording along a continuous time axis; its trace ticks select the visible measurements.
- Turn on “Brush day range” to assign selected days as the analysis window, lighting range, exclusion, fit range or shift day.
- Check daily onset and offset markers against the trace before correcting them by hand. Hand corrections belong to that recording.
The Layout selector sits beside the actogram's export buttons and is repeated under Canvas in Edit / export actogram. “Separate lanes” is the older stacked-vertically view: tick “Separate lane” only on traces that need their own vertical lanes. Other visible traces remain overlays.
For behaviour categories, choose “Parts of total” and tick “Part of total” on two or more traces with matching units and bin aggregation, such as correct and incorrect runs. The whole bar represents the sum; each coloured segment represents its percentage. Do not include a precomputed total alongside its components. A bin with any missing component is blank. The hover label gives the total, each value and its percentage. If columns contain event counts, set each column to Sum during import so the total is total runs.
Choose a period estimate
A high peak is a candidate period, not a final biological conclusion.
- Use several observable cycles and a selected interval that answers your question.
- Open Rhythmicity beside Traces shown and Dates used to choose the periodogram method and significance threshold used for the rhythmicity labels. Choose Use as default to remember these settings for new recordings; saved projects retain their own settings.
- For a file with several trace columns, select the columns under Traces shown. The periodogram overlays all three methods for each selected trace. Each curve is scaled to its own peak, so compare peak positions rather than heights.
- Check whether a peak sits at the search boundary, at half or twice another peak, or changes with the selected interval.
- Use Period estimates by method to compare the chosen methods on the same data and read their warnings.
Choosing a period for folding changes the display reference; it does not change what range the estimators searched.
Analyse recordings as a batch
Share analysis settings while retaining each recording's trace identity and manual edits.
- Import a folder, or select loaded recording tabs and create a batch.
- Use each recording tab's menu to set its condition and subject identifier. Repeated recordings from one subject must share that identifier.
- Use Traces shown, Dates used and the section's Edit traces control to choose the data included.
- Open Metrics and plots, then Metric bar plots. By condition compares a trace across conditions; By trace compares two traces within subjects and then their change across conditions.
- Choose a test, every condition pair or comparisons against a control, and the adjustment. Three or more conditions show the overall test plus the selected follow-up comparisons.
Clock-phase values always use circular permutation across midnight. With an intrinsically adjusted linear test, the adjustment selector is labelled Clock-phase correction. Small samples stay visible; calculations run when their mathematical requirements allow. Stars use adjusted p values and cap at four. Cell-level results do not establish animal-level effects.
Compare recordings and traces
Use the same settings and explicit subject identities for comparisons.
Primary trace identifies the trace used by analyses requiring one reference. Traces shown chooses the visible traces; selecting a primary trace does not turn the others into independent animals.
In metric plots, paired A/B brackets test trace differences within each condition. Between-condition brackets compare each subject's log B/A ratio when values are positive, differences when they are not, and circular lags for clock phases. The selected overall and follow-up tests support two or more conditions.
Each subject is counted once after averaging repeated recordings. Independent condition tests are unavailable when the same subject identifier appears in multiple conditions. Reordering bars changes their display order, not the chosen control or the numerical test. Projects and recovery save these choices; exported plots retain their brackets and workspace archives include the statistical results.
Run a question-specific analysis
Use the Workbench tab when you have a specific scientific question beyond the overview plots.
- Search for a question and read the inputs shown for it.
- Fill required fields, then choose “Run analysis”.
- Read the conclusion together with its values, warnings and full result. Use the question mark beside the selected question for the method behind it.
Save or export the right thing
A project is a recipe; a bundle is a set of finished outputs.
Edit and export a figure
Figure controls change how results are drawn; keep the underlying analysis settings recorded separately.
Use the actogram or time-series figure workspace for canvas size, layers, colours and labels. The time series also offers grid, legend and line controls. Download PNG for a fixed image, SVG for an editable vector figure, transparent SVG, or all selected layers as separate SVG files in a ZIP. Use Publication to arrange existing figures into a page.
Figure images are browser downloads, not files stored in the Workbench project. Open your browser’s Downloads list to find the named file and show its folder. Depending on your browser settings, it may go to your usual Downloads folder or the browser may ask you to choose a location.