Statistics overview

Statistics brings study-level comparisons into a saved dashboard. View multiple plots together, with separate stainings, co-expression pairs, metrics and filters for each plot.

Use Statistics after reviewing the underlying Results. A useful plot starts with trustworthy classifications and measurements.

Before you start

Set up the stainings you want to analyze as Classification or Detection & Classification in Study.
Select a classifier for each staining and check its output on representative slices in Results.
Select brain regions in Study for regional plots.
Check registration, slice inclusion and pixel size before interpreting measurements.

For Detail only projects, check direct brain-region assignments and hemisphere assignments. These projects support regional plots, but not atlas-space anatomical or region-free plots.

Start your dashboard

Open Statistics in your project.
If the dashboard is empty, choose Load default dashboard for a starting layout, or Edit plots to add your own.
Check the classifier selections at the top of the page.
Use Process statistics or Reprocess statistics when results are missing or outdated. Wait for calculation to finish.
Edit individual plots to answer your study questions.

The default dashboard includes the plot families supported by your project, such as anatomy, cell locations, density, heatmaps, bars and box/violin plots. It starts with up to four sources. You can add more sources in a plot's editor.

Loading defaults is a one-time starting point for an empty dashboard. The saved plots are fully editable.

See Using the statistics dashboard for editing, quick controls, fullscreen viewing and figure downloads.

Shared dashboard, personal exploration

Your project team shares the dashboard layout and saved plot settings. Moving, resizing, adding or removing plots updates that shared dashboard. Saving a plot updates its settings for the team.

Quick controls change the current viewing session. They let you explore a staining, group, metric or atlas coordinate without saving a change to the dashboard.

Classifier selections remain personal and carry across Training, Results, Statistics and classification exports. Confirm the selected classifiers when comparing figures with a colleague.

Choose a comparison

Each plot can use its own sources, metrics, grouping and filters. Keep an animal-level plot beside a group summary to check whether a difference is consistent across animals.

Read Understanding statistics plots to choose a plot, Metric semantics by workflow to choose a valid metric, and Co-expression analysis to compare marker populations.

When Hemisphere mode is Separate, regional plots use left and right measurements where available. Regions kept joint remain combined, matching Results.

Keep results current

Changing classifiers, registration, AP coordinates, animal tilt, slice inclusion or selected study regions can make measurements outdated. Direct brain-region assignments can also affect Detail only projects.

Reprocess after these changes. For tilted sections, the saved angle affects regional cell assignments, analyzed areas and atlas-space cell locations.

Changing a plot's layout, colors or viewing controls does not change the underlying classifications.

Custom analysis

For custom calculations or statistical tests, export the project and analyze it with NeuroQP Python.