Quantifying 3D Cell Shape During Early Development

For the last three decades, biologists and computer scientists have dedicated significant effort to quantifying the 3D shape of cells and analyzing their interactions. A fundamental process in developmental biology is embryogenesis. The complexity and dynamic nature of 3D embryonic cells makes effective and simple shape quantification a difficult task.

Commonly used descriptors like volume, surface area, and mean curvature are frequently insufficient as they offer only a broad overview, lack localized details, and cannot be used for reconstruction.

Recent progress in imaging technology has allowed for the creation of dynamic, time-lapse fluorescence images. These images facilitate the mapping of 3D cell morphology in developing organisms such as Caenorhabditis elegans (C. elegans).

Transferring genetic components to organisms that are not typically used as models is difficult because of mechanisms specific to the host, such as gene expression, metabolism, and differences in DNA vectors.

The lack of sufficient high-quality data on the shapes of living 3D cells has limited the investigation into the relationship between cell shape and cell fate during C. elegans embryogenesis. Therefore, there is a strong need for an integrated method of quantification and analysis.

A collaborative team from the City University of Hong Kong, Peking University, and the Centre for Intelligent Multidimensional Data Analysis Limited at Hong Kong Science Park recently published a study in the journal Quantitative Biology.

In this work, they presented a powerful integrated method called 3D Cell Shape Quantification (3DCSQ). This method converts digitized 3D cell shapes into analytical feature vectors, which they term eigengrid, eigenharmonic, and eigenspectrum. The approach uniquely integrates spherical grids, spherical harmonics, and principal component analysis to quantify cell shape.

The authors demonstrate 3DCSQ's ability to identify cellular morphological phenotypes and cluster cells. When applied to 29 living C. elegans embryos, ranging from the 4- to 350-cell stages, 3DCSQ successfully identified and quantified biologically reproducible cellular patterns, including specific deformations in skin cells.

The team also offers a program for automated cell shape lineage analysis. Their method represents a significant step forward in biological imaging and analysis, as it not only provides a systematic way to describe and evaluate cell shape but also allows for the monitoring of cell differentiation through changes in shape.

The digitized 3D cellular images, 3DCSQ, and the visualization of 17 compressed quantitative embryos, as well as examples of their lineage cell shape visualization, are presented. The shapes of cells over time are quantified and visualized during embryogenesis, following the cell lineage.

3DCSQ is specifically developed for the quantitative evaluation, visualization, and analysis of cellular morphologies, with a particular emphasis on the embryogenesis of C. elegans from the 4- to 350-cell stages.