This Title All WIREs
How to cite this WIREs title:
WIREs Comput Mol Sci
Impact Factor: 16.778

Computational prediction of protein–protein binding affinities

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract Protein–protein interactions form central elements of almost all cellular processes. Knowledge of the structure of protein–protein complexes but also of the binding affinity is of major importance to understand the biological function of protein–protein interactions. Even weak transient protein–protein interactions can be of functional relevance for the cell during signal transduction or regulation of metabolism. The structure of a growing number of protein–protein complexes has been solved in recent years. Combined with docking approaches or template‐based methods, it is possible to generate structural models of many putative protein–protein complexes or to design new protein–protein interactions. In order to evaluate the functional relevance of putative or predicted protein–protein complexes, realistic binding affinity prediction is of increasing importance. Several computational tools ranging from simple force‐field or knowledge‐based scoring of single protein–protein complexes to ensemble‐based approaches and rigorous binding free energy simulations are available to predict relative and absolute binding affinities of complexes. With a focus on molecular mechanics force‐field approaches the present review aims at presenting an overview on available methods and discussing advantages, approximations, and limitations of the various methods. This article is categorized under: Molecular and Statistical Mechanics > Molecular Interactions Molecular and Statistical Mechanics > Free Energy Methods Software > Molecular Modeling
Illustration of the rotational and axial angles that are typically used to restrain the rotation of one partner and to restrain the axial placement (relative to the second partner) in PMF‐based free energy simulations. For the definition, three centers in each partner need to be selected (indicated as blue and green spheres). These can be centers‐of‐mass of groups of atoms. The three Euler angles α, χ, and γ are used to restrain the rotation of one partner and the two axial angles θ and ϕ restrict the direction of r to separate the protein partners. Additionally, the conformation of both partner proteins is restrained, preferably via root mean square deviation (RMSD) restraints with respect to a reference structure. The distance r is typically used as reaction coordinate for the PMF calculation to induce dissociation (or guide association) of the complex
[ Normal View | Magnified View ]
Binding free energy calculations including geometrical restraints. The conformational flexibility as well as the relative rotational/axial degrees of freedom of the binding partners are restrained during the calculation of the potential of mean force (PMF) along a separation coordinate (typically a distance r). This limits the necessary sampling of relevant states during the PMF calculation (third row). The contributions of the restricted mobility can be calculated at the end points (bound and fully separated states) of the PMF simulation using either analytical or free energy perturbation (FEP) methods. The absolute binding free energy between the unrestrained proteins ΔGbind (top row) is calculated by accounting for several free energy contributions through the illustrated thermodynamic cycle. It requires the separate calculation of ΔGconf (site,bulk) (second row), orientation ΔGorient (site,bulk), and direction of separation ΔGaxial_site (third row)
[ Normal View | Magnified View ]
Evaluation of protein–protein complexes based on a continuum solvent model during MM‐PBSA or MM‐GBSA calculations. The binding process consists of an interaction contribution indicated in the lower panel (interaction energy is calculated as difference in the vacuum energies of the complex and the separate partners). The transfer of the partners and the complex into the aqueous environment (upper panel) adds a solvation contribution (also calculated as difference between complex and partner contributions). The solvation part consists typically of separate cavity terms and van der Waals interaction with the solvent plus an electrostatic reaction field (solvation) contribution either based on the generalized born (GB) method or based on solving the finite difference Poisson–Boltzmann (FDPB) equation numerically
[ Normal View | Magnified View ]
Computational protein–protein docking starting from separate unbound partners typically results in several putative sterically possible complex geometries. Task of a scoring step is to identify a most realistic geometry and to estimate the relative binding affinity of putative docked geometries
[ Normal View | Magnified View ]
Protein–protein association can induce different types of conformational changes. Conformational changes can involve side chain flips (a) indicated for a tyrosine side chain flip in the complex of RNAseA (yellow) inhibitor (green: unbound structure, pink: bound structure) complex, pdb1DFJ). For a protease‐inhibitor complex (pdb1GL1), a loop refolding is observed (b, blue: unbound inhibitor; yellow: bound inhibitor). Besides local changes also global adaptations can accompany binding (c) demonstrated for the bound (pink cartoon) and unbound (green cartoon) of an RNAseA inhibitor (yellow cartoon: RNAseA)
[ Normal View | Magnified View ]

Browse by Topic

Software > Molecular Modeling
Molecular and Statistical Mechanics > Free Energy Methods
Molecular and Statistical Mechanics > Molecular Interactions

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts