Performance comparison of generalized born and Poisson methods in the calculation of electrostatic solvation energies for protein structures

Abstract

This study compares generalized Born (GB) and Poisson (PB) methods for calculating electrostatic solvation energies of proteins. A large set of GB and PB implementations from our own laboratories as well as others is applied to a series of protein structure test sets for evaluating the performance of these methods. The test sets cover a significant range of native protein structures of varying size, fold topology, and amino acid composition as well as nonnative extended and misfolded structures that may be found during structure prediction and folding/unfolding studies. We find that the methods tested here span a wide range from highly accurate and computationally demanding PB-based methods to somewhat less accurate but more affordable GB-based approaches and a few fast, approximate PB solvers. Compared with PB solvation energies, the latest, most accurate GB implementations were found to achieve errors of 1% for relative solvation energies between different proteins and 0.4% between different conformations of the same protein. This compares to accurate PB solvers that produce results with deviations of less than 0.25% between each other for both native and nonnative structures. The performance of the best GB methods is discussed in more detail for the application for force field-based minimizations or molecular dynamics simulations.

Exploring protein native states and large‐scale conformational changes with a modified generalized born model

“…Specifically, ⌬G el computed by GB OBC differs from that computed by PB (0.5 Å grid) by 5.41 kcal/mol [root-mean-square deviation (RMSD)], whereas for GB HCT , this number is 7.16 kcal/mol. These numbers represent, respectively, 0.8% and 1.1% of the average electrostatic solvation energy of the protein (⌬G el ϭ Ϫ653 kcal/mol), and are similar 30 to those computed by the new GB(MV) model 17 implemented in CHARMM. We feel that a word of caution would be appropriate at this point:…”

“…Protein L (PDB ID: 1PTL) is represented by 215 snapshots from an explicit solvent high-temperature unfolding simulation, 17 spanning various degrees of compactness, from close to native to largely unfolded. The villin head-piece is represented by 120 structures that model compact near-native, compact misfolded, and extended conformations, produced 30 from the native coordinate set (PDB ID: 1VII). The ubiquitin native structure is PDB ID: UBQ, and thioredoxin is PDB ID: 2TRX.…”

“…ϳ0.6 (i.e., for I 3 ˜i Ϫ1 ), the formula diverges and can even change sign, which was observed in some large proteins. 30 Although the latter is a consequence of the approximations used to evaluate ⌿, and could not happen if the integral in Eq. (5) were computed exactly, the sensitivity of R i to conformational changes would still remain, since Eq.…”

“…For example, when applied to an ensemble of 120 structures representing 30 various conformational states of the small, rather loosely packed protein villin head-piece, the GB OBC shows only marginal improvement over GB HCT (Fig. 4).…”

Identification and characterization of novel mutations implicated in congenital fibrinogen disorders

“…Structures of the γ and Bβ chain mutants were constructed by altering the relevant wild‐type residues to the mutant form within Discovery Studio. Predicted mutation‐induced changes in energy and/or protein conformation of the fibrinogen complex and the fragment D dimer were determined by subjecting the wild‐type and mutant structures to solvent‐based energy minimization using the smart minimization protocol with a CHARMM forcefield21 and the Generalized Born with Simple Switching (GBSW) implicit solvent model 22. Predicted protein–protein interactions between the minimized wild‐type and/or mutant fibrinogen fragment D structures were calculated using the ZDock algorithm 23.…”