Assessment of software methods for estimating protein-protein relative binding affinities

Abstract

A growing number of computational tools have been developed to accurately and rapidly predict the impact of amino acid mutations on protein-protein relative binding affinities. Such tools have many applications, for example, designing new drugs and studying evolutionary mechanisms. In the search for accuracy, many of these methods employ expensive yet rigorous molecular dynamics simulations. By contrast, non-rigorous methods use less exhaustive statistical mechanics, allowing for more efficient calculations. However, it is unclear if such methods retain enough accuracy to replace rigorous methods in binding affinity calculations. This trade-off between accuracy and computational expense makes it difficult to determine the best method for a particular system or study. Here, eight non-rigorous computational methods were assessed using eight antibody-antigen and eight non-antibody-antigen complexes for their ability to accurately predict relative binding affinities (ΔΔG) for 654 single mutations. In addition to assessing accuracy, we analyzed the CPU cost and performance for each method using a variety of physico-chemical structural features. This allowed us to posit scenarios in which each method may be best utilized. Most methods performed worse when applied to antibody-antigen complexes compared to non-antibody-antigen complexes. Rosetta-based JayZ and EasyE methods classified mutations as destabilizing (ΔΔG < -0.5 kcal/mol) with high (83–98%) accuracy and a relatively low computational cost for non-antibody-antigen complexes. Some of the most accurate results for antibody-antigen systems came from combining molecular dynamics with FoldX with a correlation coefficient (r) of 0.46, but this was also the most computationally expensive method. Overall, our results suggest these methods can be used to quickly and accurately predict stabilizing versus destabilizing mutations but are less accurate at predicting actual binding affinities. This study highlights the need for continued development of reliable, accessible, and reproducible methods for predicting binding affinities in antibody-antigen proteins and provides a recipe for using current methods.

Atomistic Simulations and In Silico Mutational Profiling of Protein Stability and Binding in the SARS-CoV-2 Spike Protein Complexes with Nanobodies: Molecular Determinants of Mutational Escape Mechanisms

“…This approach is comparable to other knowledge-based structural methods such as Dcomplex which is based on distance-specific contact potentials trained on the monomer structures. By combining structural and evolutionary profile analyses derived from the binding interfaces, the BindProfX approach showed potential for improvements over both the statistics-based and physics-based FoldX potentials. − The BeAtMuSiC approach adapted in our study that was further enhanced through ensemble-based averaging of binding energy computations yielded a good correlation between prediction and experiments and was successfully used for the binding free analysis of the SARS-CoV-2 S binding with the host receptor and antibodies. − The recent analysis of computational tools for estimating protein–protein binding probed several approaches or their predictive ability of relative binding affinities for 654 single mutations on antibody–antigen and nonantibody–antigen complexes . Instructively, this analysis showed context-dependent and system-dependent strengths and weaknesses of these tools.…”

“…Instructively, this analysis showed context-dependent and system-dependent strengths and weaknesses of these tools. Moreover, physics-based methods often display a similar predictive power to comparable but much faster knowledge-based methods which are indispensable for massive mutational scanning experiments . Importantly, it was concluded that most methods can quickly and accurately differentiate between favorable and unfavorable mutations rather than accurately predicting the actual binding affinities …”

Molecular modeling predicts novel antibody escape mutations in respiratory syncytial virus fusion glycoprotein

“…Although this modeling approach remarkably identified six MARMs, there is still a need for better modeling strategies to improve correlation between experimental and predicted ΔΔ G Bind values for antigen-mAb complexes. Use of MD simulation to provide conformation sampling is the key reason behind the performance of the MD+FoldX approach (56). However, empirical validation of the MD+FoldX approach carried out in this study further emphasizes the need for improved conformational strategies for large protein-protein complexes.…”

Defining the HIV Capsid Binding Site of Nucleoporin 153

“…This discrepancy between modeling and empirical data is likely due to the water-mediated interaction of T1416 with CA residues R173 and E63 (21). These modeling inaccuracies are expected because one of the significant limitations of fast protein-protein binding affinity prediction tools (FoldX and PyRosetta) is that they ignore the explicit presence of bridging water molecules (31).…”