Toward a Whole-Cell Model of Escherichia coli

Derek Macklin, Stanford University

Whole-cell computational models comprehensively simulate the growth and division of single cells, explicitly accounting for the functions of all known gene products and their interactions. Such models have the potential to revolutionize biology by serving as a platform for interpreting complex behaviors, prioritizing experiments and enabling design. Recently, our lab completed the first whole-cell model of the simplest culturable organism, Mycoplasma genitalium. We are now embarking on the creation a whole-cell model of one of the foundational model organisms in biology, Escherichia coli. Because E. coli has 10 times more genes and 50 times more molecules than M. genitalium, whole-cell simulations present a number of computational challenges, particularly in execution time, due to the massive increase in the number and diversity of interactions that take place. Furthermore, modeling challenges also abound. We have made considerable progress on both of these fronts: We have performed state-of-the-art gene expression measurements to obtain high-quality model parameters; we have made a number of optimizations to our code that have improved run time by an order of magnitude; and we have implemented a model execution framework to, for the first time, fit, run and analyze multi-generation simulations. As we continue to incorporate gene functions, we expect the model to provide systems-level insight into E. coli physiology.

Abstract Author(s): Derek N. Macklin, Nicholas A. Ruggero, John Mason, Javier Carrera, Markus W. Covert