Devin Sullivan’s PhD thesis in the Murphy Lab at Carnegie Mellon University extends generative statistical modeling of subcellular organization into three dimensions and across time. Building on the CellOrganizer framework, the work learns models of cell geometry, organelle morphology, and spatial distribution directly from fluorescence microscopy images — and introduces temporal modeling to capture how cellular organization evolves dynamically.
What the Thesis Covers
The thesis addresses a core challenge in computational cell biology: how to represent the spatial organization of a cell in a form that is both statistically rigorous and practically useful for downstream simulation. Key contributions include:
- 3D generative models of nuclear, cell, and organelle shape learned from image data
- Temporal extensions that model how organization changes over time
- Integration of learned geometries with spatial simulation frameworks
- Evaluation methods for assessing model quality and biological fidelity
Connection to CellOrganizer
This work is a direct extension of the CellOrganizer project, which I contributed to as a software developer and research programmer in the Murphy Lab. The infrastructure, tooling, and earlier model implementations that the CellOrganizer team built formed the foundation on which this thesis was developed.
Citation
Image-Derived Generative Modeling of Complex Cellular Organization in Both Space and Time Devin P. Sullivan PhD Thesis, Carnegie Mellon University Computational Biology Department, 2015 Technical Report CMU-CB-15-102 PDF