In 2009, I completed my master's degree in biomedical engineering in the
Department of Systems and Computer Engineering
at Carleton University. Under the supervision of
Dr. Gabriel Wainer,
my research focused on two topics: the simulation of deformable
biological structures, and the design of simulations of biological
systems using a formalism called DEVS.
My master's thesis can be downloaded here: DEVS-Based Dynamic Simulation of Deformable Biological Structures.
The software for my project was written entirely in Python. You can download a copy of the source code.
If you are interested in my research, please email me at email@example.com.
My personal website is www.rhysgoldstein.com.
Tethered Particle System (TPS)Deformable biological structures are typically modeled with the finite element method, but we have designed a much simpler impulse-based method called the "tethered particle system" (TPS). The TPS involves the use of discrete event simulation to track the positions of a large number of particles. As would be expected, two approaching particles may collide and rebound off one another. What's different is that, provided they are "tethered", two separating particles may also collide and retract inwards. This constrains the distances between pairs of particles, allowing various deformable structures to be represented.
The video on the right shows a simulation of part of a nerve cell called the "presynaptic nerve terminal". Synaptic vesicles (green) are initially scattered inside a presynaptic compartment. Over time, clusters form as vesicles bind with synapsin protein (blue). Docking sites (red) fix vesicles to the active zone at the bottom of the membrane, eventually attracting a large cluster. An action potential occurs at 1:30, causing vesicles to fuse with the membrane (shown as vesicles escaping the compartment). The same action potential is shown at 2:00 from a different angle. Synapsins can be seen separating from the vesicles, which disrupts the cluster.
The model was designed in collaboration with Dr. James J. Cheetham, a biologist at Carleton. Among other things, it may help predict how a decreased concentration of synapsin affects the transmission of signals through the brain.
Discrete Event System Specification (DEVS)I doubt most biological system modelers have considered adopting the Discrete Event System Specification (DEVS), but this formalism offers a promising approach for those who want to design complex simulations without making a mess of their code. DEVS has two properties that help keep software manageable. First, the simulator, which advances time and invokes state transitions, is kept separate from the model. Once the simulator is implemented, it can be used for a wide range of different models. Second, DEVS supports hierarchical model design. A complex DEVS model can be defined as an aggregation of simpler submodels, which can in turn be defined as hierarchies.
We implemented a DEVS simulator and a DEVS model of the tethered particle system in the Python programming language, and used the code to obtain simulation results like those in the above video. Our DEVS TPS model features the hierarchical design illustrated at the left.
PapersR. Goldstein, DEVS-Based Dynamic Simulation of Deformable Biological Structures, MASc Thesis, Carleton University, Ottawa, ON, Canada, 2009.
R. Goldstein, G. Wainer, Impulse-Based Dynamic Simulation of Deformable Biological Structures, Transactions on Computational Systems Biology, Springer-Verlag, Volume 6575, Pages 39-60, 2011. The original publication is available at www.springerlink.com.
R. Goldstein, G. Wainer, DEVS-Based Design of Spatial Simulations of Biological Systems, Proceedings of the Winter Simulation Conference (WSC), Austin, TX, USA, 2009.
R. Goldstein, G. Wainer, Simulation of a Presynaptic Nerve Terminal with a Tethered Particle System Model, Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Minneapolis, MN, USA, 2009.
R. Goldstein, G. Wainer, Simulation of Deformable Biological Structures with a Tethered Particle System Model, Proceedings of the 32nd Conference of the Canadian Medical and Biological Engineering Society (CMBEC), Calgary, AB, Canada, 2009.
R. Goldstein, G. Wainer, J. Cheetham, R. Bain, Vesicle-Synapsin Interactions Modeled with Cell-DEVS, Proceedings of the Winter Simulation Conference (WSC), Miami, FL, USA. 2008.
R. Goldstein, G. Wainer, Modelling Tumor-Immune Systems with Cell-DEVS, Proceedings of the European Conference on Modelling and Simulation (ECMS), Nicosia, Cyprus, 2008.