Skip to content
Image processing for developmental biology.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
Measurements First commit of working EDGE build Apr 7, 2012
doc more broken links Jul 21, 2016
java1 added java binaries and removed special character from java src Jan 26, 2017
screenshots added screenshots and changed link Apr 19, 2016
DATA_INFO_DEFAULT.csv got default data info and gitignore Aug 6, 2012
EDGE_paper.bib added bib file Jul 21, 2016
install.m updated instal and wrap files Aug 3, 2012
install_nopermissions.m small change to install file Jun 6, 2012
license.txt add license file Mar 16, 2015 added class files, will remove. also changed button text Aug 7, 2012


Embryo Development Geometry Explorer (EDGE) is a software package for analyzing microscope images of developing embryos. It takes as its input a stack of membrane-tagged images (or, optionally, a "4D" stack that includes a temporal dimension) and builds a 3D reconstruction of the cells. It then can compute various geometric properties of the cells, including user-specified properties through a Matlab API. EDGE was developed specifically for Drosophila embryo development data but can be used for a variety of systems.

EDGE was created by Michael Gelbart, Matthias Kaschube, Adam Martin and Eric Wieschaus at Princeton University.


Download EDGE here or clone the git repository using the link on the right-hand side of this page.



Unfortunately, EDGE is no longer supported by the developer. You could try opening an Issue in this repository but it is unlikely to be answered.

Citing EDGE

If you use EDGE in your work, please cite the following paper:

Michael A. Gelbart, Bing He, Adam C. Martin, Stephan Thiberge, Eric F. Wieschaus, Matthias Kaschube. Volume conservation principle involved in cell lengthening and nucleus movement during tissue morphogenesis. Proc Natl Acad Sci, 109(47):19298-19303, 2012. BibTeX


screenshots/Screenshot_semiauto_cells.png A screenshot of the EDGE Importer showing the polygon-membranes (pink) overlaid over the raw image (grayscale). The red dots show the centroids of all the identified cells. The buttons on the right can be used to fix any errors, and the many other settings allow for control over other aspects of the processing.

screenshots/screenshot_neighbors.png A screenshot of the EDGE Browser showing that the membrane detection (polygons, pink) works very well even for noisy data sets. The membranes were produced with minimal manual error-correction.The lower left plot shows area vs. depth for the selected cell (red) and its neighbors (green). A 3D reconstruction of the selected cell and its neighbors are plotted in the lower right.

hscreenshots/Screenshot_nuclei_intensity.png Another screenshot of the EDGE Browser. This fixed data set has been stained for the nuclei. The nuclei (green) and the polygon membranes (pink) are overlaid on top of the raw data (grayscale). The bottom plot shows the intensity of the nuclei channel inside the selected cell as a function of depth. The intensity curve exhibits a clear maximum, showing that the nucleus is spatially localized inside the cell.

You can’t perform that action at this time.