Abstract

As the number of atoms in proteins increases, so too does the complexity of the underlying potential energy surface. Hence treating sets of atoms as local rigid bodies becomes appealing, since this approach reduces the number of degrees of freedom, leading to shorter computer simulation times. The usefulness of local rigidication has been demonstrated in the context of basin-hopping global optimisation, where significant reductions in mean first encounter times were observed. However, the effects of this formalism on predicted protein folding kinetics and mechanisms have not yet been benchmarked. Investigating these effects requires a systematic study of the transformation of the energy landscape as increasingly larger domains of atoms are locally rigidified and is the focus of my presentation.