Simulating microbial systems: addressing model uncertainty/incompleteness via multiscale and entropy methods.

Area of Science:

• Multiscale modeling
• Computational biology
• Systems science

Background:

• Natural and engineering systems are often multiscale.
• Existing models can be incomplete or uncertain, necessitating probabilistic approaches.
• Direct simulation of large systems (e.g., 10^6 atoms) is computationally infeasible.

Purpose of the Study:

• To present a deductive multiscale approach for modeling complex systems.
• To derive coarse-grained equations from underlying physical models for a calibration-free methodology.
• To demonstrate the approach using virus and cell systems.

Main Methods:

• Reducing resolution from N-atom descriptions to coarse-grained variables.
• Integrating coarse-grained equations with probabilistic descriptions of fine-scale states.
• Developing a computational platform (SimEntropics™) for microbial modeling.

Main Results:

• A methodology for modeling microbial systems is presented.
• The approach allows for a probabilistic description of underlying microstates.
• Deductive, calibration-free modeling is achieved.

Conclusions:

• The developed deductive multiscale approach provides a robust framework for complex systems.
• Probabilistic descriptions are essential for handling uncertainty in multiscale modeling.
• The SimEntropics™ platform offers prospects for advanced microbial modeling and applications.