spheresDT/Mpacts-PiCS: cell tracking and shape retrieval in membrane-labeled embryos.

Area of Science:

• Developmental biology
• Biophysics
• Computational biology

Background:

• Accurate 3D cell geometry measurement over time is crucial for understanding embryo formation.
• Automated cell shape extraction from time-lapse microscopy, especially with only membrane labeling, presents significant challenges.

Purpose of the Study:

• To develop and validate an automated image analysis framework for 3D cell segmentation and tracking in confocal time-lapse microscopy.
• To enable quantitative analysis of cell geometry during early embryonic development.

Main Methods:

• A novel framework combining sphere clustering for initial segmentation and tracking with discrete element method simulations for refinement.
• Utilizes a biomechanical cell shape model to constrain segmentation accuracy.
• Applied to Caenorhabditis elegans embryos during early developmental stages.

Main Results:

• Successful automated tracking and 3D segmentation of cells in confocal time-lapse microscopy.
• Quantitative analysis of cell volume, contact area, and shape in 7- and 8-cell stage Caenorhabditis elegans embryos.
• Demonstrated framework's capability to handle variable cell shapes and membrane-only labeling.

Conclusions:

• The developed framework provides an accurate and automated method for analyzing 3D cell geometries during embryonic development.
• This tool facilitates the study of cellular and mechanical processes driving embryo formation.
• The open availability of the code and data promotes reproducibility and further research in the field.