Replication and Optimization of a Multicellular Spatial Model of IFN Response to Viral Infection

This respository holds materials for our replication and optimization of the simulation reported in the paper Multicellular Spatial Model of RNA Virus Replication and Interferon Responses Reveals Factors Controlling Plaque Growth Dynamics: https://www.biorxiv.org/content/10.1101/2021.03.16.435618v1.

The original model code can be found here on GitHub in the repository Multicellular Spatial Model of RNA Virus Replication: https://github.com/ImmuSystems-Lab/Multicellular_Spatial_Model_of_RNA_Virus_Replication.

This model is designed to be accessible regardless of experience with computer programming, though its use requires a working installation of CompuCell3D. Running a simulated plaque assay is as simple as starting the simulation; parameters including virus and interferon diffusivity and decay, the rate at which cells infected with the virus lose health, the rate of STAT activation, RIG-I activity (which is set to zero by default to reflect antagonism by influenza and SARS-CoV), and multiplicity of infection for the seeding of the simulated cell culture with the virus can be changed by dragging sliders in a window which appears upon simulation start.

Model details and biological background are given in this paper. To facilitate exploration for the curious, we suggest the following exercises:

1. In this model, diffusion of virions is the only way the virus can move around the host cell culture. The tendency of the virions to diffuse is modeled mathematically by the viral diffusion constant, which you can change while the simulation runs using the sliders. How would the infection progress if the virus diffused more than the default? Or less? Use the sliders to change the value of the viral diffuison constant – as one would expect, a higher viral diffusion constant means more diffusion, while a lower constant means less diffusion – and compare.
2. Virions don't just move through the extracellular matrix by diffusion in this model; they are also uptaken by host cells, which we capture mathematically as virion decay. What if host cells were to take up greater concentrations of virions at once, or something were to destroy the virions in the extracellular matrix? You can experiment with this situation by using the sliders to intensify the viral decay. Try increasing viral decay by a factor of 10. How does the simulation's behavior change? Why?
3. Once simulated cells enter the viral-releasing stage, the likelihood that they will transition to the death state increases as their health decreases. What would happen if the virus were more virulent than the default? Increase the rate at which infected cells lose health and compare what you see to the default simulation's behavior. Does the infection spread more or less? Why?
4. The cells in this model alert one another to the presence of the virus by releasing interferons. The interferon-releasing process within the cell is accelerated when the cell detects extracellular interferon, and this detection is accomplished by the molecule STAT, which "activates" and becomes STATP upon interaction with interferon. Increase STAT activation by a factor of 100 and compare the infection outcomes. What is the difference? Why?
5. The cells in this model alert one another to the presence of the virus by releasing interferons. What if the interferon signal were more potent, spreading throughout more of the simulated cell culture? Increase the diffuison constant of extracellular interferon using the steering panels and compare the simulation's behavior with that of the default.
6. Researchers are working on treatments in which infected tissues are bathed in cytokines such as interferons prior to or at the onset of infection. It's possible to simulate such a treatment with this model: open the model in CompuCell3D's Twedit code editor, find the varible IFNWash, and set it to True. How does the infection progress under these conditions? Based on your observations alone, is this a good treatment? Would you expect this treatment to be as effective or as ineffective as this if it were done in vivo? Why or why not?