v0.0.4

🚀 CellColoc v0.0.4

June 23, 2026

This release expands CellColoc from a pure multi-channel colocalization
workflow into a broader interactive microscopy-analysis toolkit by adding a
dedicated single-channel mode, a first full set of usage tutorials, and
tutorial-derived notebook counterparts for the interactive example scripts.

✨ Features

• add a dedicated single-channel segmentation and counting workflow that can
  analyze one microscopy channel without any colocalization step while still
  reusing CellColoc's existing core capabilities:
  • Cellpose and threshold-based segmentation backends
  • ROI-based or whole-image analysis
  • prefiltering and postfiltering
  • global z-cropping and optional z projection
  • cached Cellpose refinement
  • manual napari relabeling and reanalysis
  • standardized mask and table export
• add a dedicated 2D DAPI nuclei demo script for the new single-channel
  workflow
• extend the single-channel object export with morphology metrics:
  • 2D area, perimeter, roundness, and eccentricity
  • 3D volume, voxel-surface area, sphericity, and ellipticity-like
    elongation
  • a separate voxel plausibility sheet in the Excel export
  • per-ROI averages of the new morphology metrics

📃 Changes

• allow VOXEL_SCALE_ZYX to be provided either as a full (Z, Y, X)
  tuple or, for 2D workflows, as a shorter (Y, X) tuple that is expanded
  internally to (1.0, Y, X)
• add the first full usage tutorials to the Read the Docs documentation:
  • a 2D tutorial based on the DAPI-stained nuclei example workflow
  • a 3D tutorial based on the microglia example workflow
  • a three-channel tutorial
  • a three-channel z-projection tutorial
  • a 2D single-channel nuclei tutorial
• generate notebook counterparts for the interactive example workflows from
  the tutorial structure itself, including local figure references inside the
  user_scripts folder
• expand the documentation with mathematical definitions of object-based
  colocalization and occupancy metrics
• improve the Read the Docs configuration so copy buttons are shown on all
  standard highlighted code blocks instead of only Python code snippets
• extend the 2D DAPI example user script with:
  • whole-image-as-single-ROI mode
  • automatic reuse of an existing saved ROI mask from the results directory
• add a dedicated three-channel 3D microglia demo script that demonstrates:
  • active segmentation of the third channel
  • separate visualization of cells positive for channel 0+1
  • separate visualization of cells positive for channel 0+2
  • separate visualization of cells positive for channel 0+1+2
• add a dedicated three-channel z-projection demo script that demonstrates:
  • global z projection before segmentation
  • projected three-channel analysis
  • projected positivity views for 0+1, 0+2, and 0+1+2
• extend cache-based Cellpose refinement so the optional third analysis channel
  can also be rebuilt from cached Cellpose outputs, including optional
  threshold changes and postfiltering
• keep manual reanalysis after napari label edits consistent with the active
  analysis z-bounds in the 3D workflows
• surface the new single-channel workflow explicitly in the README and the
  general documentation overview as a first-class CellColoc feature