The conception and simulation of artificial cells has been a highly successful piece of collective research. At the beginning of the project, the available literature on the protocell lifecycle did not included a single piece of truly embodied cellular replication with a physically reasonable implementation. Although some interesting results were available on Alife systems, they were not connected to the fundamental processes involved in the growth-deformation-splitting loop required for a macromolecular assembly to divide. Every existing model four years ago was thus lacking a crucial part of the cell division process. PACE research has generated several successful models of whole cell replication showing that such a phenomenon is expected to occur under a wide range of conditions, provided that some well-defined requirements are at work. The outcome of this section is also a good example of multilevel approaches to a given problem. Models have taken into account every possible scale, from the quantum MD simulation approach to Monte Carlo, DPD or mesoscopic approximations. Many lessons have emerged from these models and PACE has been able to open a new field of computational analysis of protocellular systems.
The simulation results have guided experimental work in diverse ways. The design of an appropriate photometabolic system has benefited from quantum mechanical simulations, the association of novel genetic molecules such as PNA with lipid membranes has been calculated with molecular dynamics. The dynamical phase behavior of amphiphile systems, including conditions for micelle and vesicle formation and fission, have been explored with extended DPD at the mesoscale. This work has resulted in a complete simulation of chemistry and physics in the life-cycle of two kinds of artificial cells: micellar and vesicular. For example, the implementation of a micellar system including information, containment and metabolism (Los Alamos bug) has shown the usefulness of these approaches in providing insight to experiments. Lattice (spin-lattice) and continuous macroscopic models (e.g. Ginzberg-Landau) have complemented the microscopic picture and all levels have been linked to chemical kinetic descriptions. Furthermore, the physico-chemical simulations have laid the groundwork for evolutionary studies (see the following section) and for simulations of programmed vesicle self-assembly. They have also been important in creating an understanding of timescales and robustness of compartmental solutions to artificial cells for implementation of the omega machine to optimize artificial cell architecture.