Zebrafish embryo

Click in the image to view a video

ZEBRA is an embryology dataset

The ZEBRA dataset provides a feature representation (154 features) of cells of zebrafish embryo to determine whether they are in division (meiosis) or not. All the examples are manually annotated. The dataset was prepared by Emmanuel Faure (Institut des systèmes complexes, France).
The embryomics project is devoted to the morphodynamical “reconstruction” of the cell lineage tree underlying the processes of animal embryogenesis. A set of strategies, methods and algorithms to “sequence” the cell lineage tree as a branching process annotated in space and time was designed. The final goal is to fully reconstruct the dynamics of cell divisions and movements from time-lapse series of high-resolution optical sections obtained by multiphoton laser scanning microscopy throughout embryonic development of live animals. Embryomics allows the automated tracking of events such as cell division and cell death in live embryos and give us access to parameters such as the rate of cell proliferation in time and space. Special thanks to all the project participants: Thierry Savy, Louise Duloquin, Miguel Luengo Oroz, Benoit Lombardot, Camilo Melani, Paul Bourgine, and Nadine Peyriéras.

CausalityThis dataset is used in the Active Learning Challenge by the Causality Workbench