Skip to content

Settings

The Settings tab is the central place for persisting and reloading analysis configuration across tabs.

It uses a unified JSON bundle (senoquant_settings.json) so the same file format can be reused by:

  • Manual save/load from the Settings tab.
  • Batch output metadata generated after batch runs.
  • Quantification feature exports that include settings context.

Controls

  • Save settings: Writes a senoquant_settings.json file.
  • Load settings: Reads a senoquant_settings.json file and applies supported settings.

What is saved

Segmentation tab state

  • Selected nuclear model.
  • Current nuclear model settings.
  • Selected cytoplasmic model.
  • Current cytoplasmic model settings.

Spots tab state

  • Selected detector.
  • Current detector settings.
  • Spot diameter filters (minimum and maximum diameter values in pixels).

Batch tab state

  • Current batch configuration is saved into the bundle under batch_job.
  • This allows the same file to restore batch UI state later.

Bundle key layout

  • Segmentation + Spots UI state is stored under tab_settings.
  • Batch UI state is stored under batch_job.
  • Quantification feature exports use separate per-feature feature_settings.json files and are not intended for Settings-tab reload.

What is restored on load

  • Segmentation model selections and settings.
  • Spots detector settings and diameter filters.
  • Batch tab state, when the loaded JSON contains a non-empty batch_job section.

What is not restored

  • Prediction tab model selection and model-interface settings.
  • Quantification tab feature configuration.
  • Visualization tab plot configuration.
  • Runtime viewer state (active layer selection, visibility, colormaps).

Typical workflow

  1. Configure Segmentation and Spots parameters in their tabs.
  2. Optionally configure Batch settings.
  3. Open Settings and click Save settings.
  4. Reopen later and click Load settings to restore the saved setup.

Notes

  • The UI stores and restores settings by key, so unknown keys in a JSON file are ignored safely.
  • If a model or detector from the file is unavailable in the current installation, that specific selection cannot be applied.