Protein Structure and Function: Application of Bioinformatics Methods - John Rigden 2014

Protein Dynamics: From Structure to Function
Collective Variable Sampling Algorithms
Collective Variable Sampling Algorithms

Analysis of MD calculations in terms of collective coordinates (obtained, for example, via Structure/140.html">Principal Component Analysis or Normal Mode Analysis) reveals that only a fraction of the total degrees of freedom dominate The Molecular Dynamics of Biomolecules. Because protein function is often closely linked to these essential subspace modes, dynamics within this low-dimensional space has been termed “essential dynamics” (ED). It not only facilitates the analysis and interpretation of MD trajectories, but also paves the way for enhanced sampling algorithms that explore the essential subspace either systematically or probing-wise (Grubmüller 1995; Amadei et al. 1996).

Early efforts in this direction aimed to establish a computational scheme in which the equations of motion were integrated exclusively along selected primary principal modes, thereby drastically reducing the number of degrees of freedom (Amadei et al. 1993). However, these attempts proved challenging due to nontrivial couplings between high- and low-amplitude modes, even though the modes became linearly independent (orthogonal) upon diagonalization. Consequently, preference was given to alternative methodologies that account for the full-dimensional system while enhancing motion along selected principal modes. The most prominent among these techniques are conformational flooding (Grubmüller 1995) and ED sampling (Amadei et al. 1996; de Groot et al. 1996a, b). In conformational flooding, an additional term is introduced into the potential energy function for selected principal components, pushing the simulated system into uncharted regions of phase space, whereas ED sampling achieves a similar objective by imposing geometrical constraints on the chosen principal modes. These approaches can yield an order-of-magnitude increase in sampling efficiency, provided that a reasonable approximation of the principal modes has been derived from a conventional calculation. Nevertheless, due to the resulting spatial or energetic distortion of the system, the ensembles generated by ED sampling and conformational flooding are non-canonical, thereby restricting The Scope of analysis primarily to structural questions.