Optimal prediction of folding rates and transition state placement from native state geometry

Abstract

A variety of experimental and theoretical studies have established that the folding process of monomeric proteins is strongly influenced by the topology of the native state. In particular, folding times have been shown to correlate well with the contact order, a measure of contact locality. Our investigation focuses on identifying additional topologic properties that correlate with experimentally measurable quantities, such as folding rates and transition state placement, for both two- and three-state folders. The validation against data from forty experiments shows that a particular topologic property which measures the interdepedence of contacts, termed cliquishness or clustering coefficient, can account with significant accuracy both for the transition state placement and especially for folding rates, the linear correlation coefficient being $r=0.71$. This result can be further improved to $r=0.74$, by optimally combining the distinct topologic information captured by cliquishness and contact order.