GRID: A high‐resolution protein structure refinement algorithm

Abstract

The energy-based refinement of protein structures generated by fold prediction algorithms to atomic-level accuracy remains a major challenge in structural biology. Energy-based refinement is mainly dependent on two components: (1) sufficiently accurate force fields, and (2) efficient conformational space search algorithms. Focusing on the latter, we developed a high-resolution refinement algorithm called GRID. It takes a three-dimensional protein structure as input and, using an all-atom force field, attempts to improve the energy of the structure by systematically perturbing backbone dihedrals and side-chain rotamer conformations. We compare GRID to Backrub, a stochastic algorithm that has been shown to predict a significant fraction of the conformational changes that occur with point mutations. We applied GRID and Backrub to 10 high-resolution (≤ 2.8 Å) crystal structures from the Protein Data Bank and measured the energy improvements obtained and the computation times required to achieve them. GRID resulted in energy improvements that were significantly better than those attained by Backrub while expending about the same amount of computational resources. GRID resulted in relaxed structures that had slightly higher backbone RMSDs compared to Backrub relative to the starting crystal structures. The average RMSD was 0.25 ± 0.02 Å for GRID versus 0.14 ± 0.04 Å for Backrub. These relatively minor deviations indicate that both algorithms generate structures that retain their original topologies, as expected given the nature of the algorithms.

Protein Structure Refinement Algorithms

“…Good prediction techniques would allow us to obtain the structures of novel protein sequences without having to determine them experimentally and would be particularly useful for membrane and other proteins whose structures are experimentally difficult to obtain or are highly diverged from the set of proteins with existing structures 1 . The knowledge obtained from accurate protein structure prediction could improve our understanding of how proteins function, aid in the discovery of proteins with new and improved properties, and facilitate the development of new drugs 2,3 .…”

“…The knowledge obtained from accurate protein structure prediction could improve our understanding of how proteins function, aid in the discovery of proteins with new and improved properties, and facilitate the development of new drugs 2,3 . Many biological applications, including studies on enzymatic activity, virtual ligand screening, structure-based protein function prediction, and structure-based drug design require knowledge of protein structures at atomic resolution 1 . In addition, the expansion of highthroughput genome sequencing projects has led to a continually widening gap between protein sequence data and the number of solved structures.…”

“…However, the resulting models are still low-resolution and further refinement is needed in order to achieve the atomic-level accuracy required for most biological applications. Refinement of low-resolution structures to higher resolution models therefore remains a major challenge in the field of protein structure prediction 1,4,[9][10][11] .…”

“…Approaches to high-resolution refinement have included molecular mechanics energy minimization and Monte Carlo methods including replica exchange Monte Carlo 1,[13][14][15][16] .…”

“…GRID uses a deterministic approach to high-resolution refinement, improving the energy of a protein structure by systematically perturbing the backbone dihedrals and side-chain conformations. In this study, REMC GRID was used for two purposes: (1) structure refinement, and (2) generation of decoy structures. In both applications, REMC GRID takes a protein structure as input and generates a conformationally diverse set of low-energy perturbed structures.…”