Lecture 2: Scalably Building, Simulating, Calibrating, and Validating Whole-Cell Models
Recently, we and others demonstrated the feasibility of WC models by developing the first model that captures the function of each individual gene. However, it took over 10 person-years to construct the model and, despite our best efforts, the model is difficult to understand, reuse, and extend. To enable more comprehensive and more predictive models that can guide bioengineering, we must develop scalable methods and tools for every aspect of WC modeling. This lecture will introduce several emerging methods for scalably building, simulating, calibrating, and validating WC models including data aggregation and integration from public repositories, pathway/genome databases for organizing model input data, multi-algorithmic simulation for modeling both well- and poorly-characterized biology at multiple grains, surrogate modeling for computationally-efficient approximate calibration, and unit testing for rigorous verification.