1869-5868/asset/olalertbanner.jpg?v=1&s=fe896002a886640f17721a3de4f11f71cd9ec948)
Simulating Biomolecular Folding and Function by Native-Structure-Based/Go-Type Models
Abstract
The 2013 Nobel Prize in Chemistry highlights how crucial computer simulations have become for many scientific and engineering fields. Nowadays, scientific progress is not only driven by the interplay of new experimental measurements and increasingly sophisticated theoretical frameworks, but also by an incredible toolbox of complex computational models meeting ubiquitously available computing power and data storage facilities. Quantum mechanical (QM) calculations can be condensed into molecular mechanics (MM) force fields and coupled QM/MM calculations can derive atomic and molecular properties of biomolecular or materials science systems with high accuracy. Pure MM simulations driven by Monte Carlo or molecular dynamics algorithms are widely applied in biological chemistry/physics and can investigate large biomolecular systems, such as proteins, DNA, or RNA. One coarse-grained class of these models, native-structure-based or Go models, are based on energy landscape theory and the principle of minimal frustration. Herein, an ensemble of converging pathways guide protein folding on a funnel-like shape of the entire energy landscape towards the native state. Simulations based on these ideas have been tremendously successful in explaining protein folding and function. Their history and recent application highlights are reviewed.