Many classical models of development, including amphibians, fish, and echinoderms, have embryos that are approximately spherical and contain concentric cell layers during early development. Neighbor relationships in such curved cell layers are not conveniently assayed or measured by conventional physical or optical sectioning techniques. To answer these challenges, we have constructed computational methods that correct for spherical distortion in 2D images and that allow extraction of concentric cell layers from 3D digital images. These methods for quantitative analysis and visualization of early development in spherical embryos are introduced by using them for the quantitative analysis of 2D and 3D images of gastrula stage Xenopus laevis.
Copyright 2005 Wiley-Liss, Inc.