A hierarchical, building-block-based computational scheme for protein structure prediction

Protein folding is a hierarchical event in which transiently formed local structural elements assemble to yield the native conformation. We outline the hierarchical building block protein folding model, which is based on two premises. First, while the local building block elements may be unstable, they nevertheless have higher population times than all alternate conformations; second, protein folding progresses through a combinatorial assembly of these elements. In accordance with this model, we describe a building block cutting algorithm, implementing its rationale. Through its automated iterative application to the native structure we obtain an anatomy tree, in terms of protein folding. The anatomy tree automatically yields the most likely folding pathway. In particular, we describe how, by using this algorithm and the building blocks which are obtained, we expect to reduce substantially the computational time involved in simulations of protein folding.