CellOrganizer is an open source system for using microscope images to learn statistical models of the structure of cell components and of how those components are organized relative to each other.
Since these models can be used to synthesize new images reflecting what the model learned, they are referred to as generative models.
For example, CellOrganizer can take fluorescence microscope images of many cells expressing a GFP-tagged lysosomal protein and construct a model of how the cell and nuclear shapes vary among those cells, how many lysosomes are typically found per cell, how they vary in size, and where they are typically located relative to the cell and nuclear boundaries.
Publications describing the development and use of various software components of CellOrganizer are located here. A recent overview of the principles behind CellOrganizer can be found here. And to download the most recent version of CellOrganizer, please visit our download page.
CellOrganizer provides tools for
learning generative models of cell organization directly from images;
storing and retrieving those models;
synthesizing cell images (or other representations) from one or more models.
Model learning captures variation among cells in a collection of images. Images used for model learning and instances synthesized from models can be two- or three-dimensional static images or movies.
CellOrganizer can learn models of
vesicular organelle size, shape and position;
These models can be conditional upon each other. For example, for a given synthesized cell instance, organelle position is dependent upon the cell and nuclear shape of that instance. Cell types for which generative models of at least some type of organelle have been built include human HeLa cells, mouse NIH 3T3 cells, and Arabidopsis protoplasts. Planned projects include mouse T lymphocytes and rat PC12 cells.
Support for CellOrganizer has been provided by grants GM075205, GM090033 and GM103712 from the National Institute of General Medical Sciences, grants MCB1121919 and MCB1121793 from the U.S. National Science Foundation, by a Forschungspreis from the Alexander von Humboldt Foundation, and by the Freiburg Institute for Advanced Studies.
CellOrganizer is research code. As such, it is under constant development. Although we do our best to ensure our code is reliable, we distribute this code under the GNU public license without any warranty. Despite our best efforts, a feature may sometimes fail to work as expected. If you experience any such issues or have any questions regarding CellOrganizer, please do not hesitate to contact us at
About this document¶
This document covers the minimum materials required for learning model distributions of cellular organization and synthesizing images from those distributions. For more in-depth examples, please refer to the demos and the templates directories contained within the CellOrganizer distribution.