Abstract:Protein mutations occur frequently in biological systems, which may impact, for example, the binding of drugs to their targets through impairing the critical H-bonds, changing the hydrophobic interactions, etc. Thus, accurately predicting the effects of mutations on biological systems is of great interests to various fields. Unfortunately, it is still unavailable to conduct large-scale wet-lab mutation experiments because of the unaffordable experimental time and financial costs. Alternatively, in silico computation can serve as a pioneer to guide the experiments. In fact, numerous pioneering works have been conducted from computationally cheaper machine-learning (ML) methods to the more expensive alchemical methods with the purpose to accurately predict the mutation effects. However, these methods usually either cannot result in a physically understandable model (ML-based methods) or work with huge computational resources (alchemical methods). Thus, compromised methods with good physical characteristics and high computational efficiency are expected. Therefore, here, we conducted a comprehensive investigation on the mutation issues of biological systems with the famous end-point binding free energy calculation methods represented by MM/GBSA and MM/PBSA. Different computational strategies considering different length of MD simulations, different value of dielectric constants and whether to incorporate entropy effects to the predicted total binding affinities were investigated to provide a more accurate way for predicting the energetic change upon protein mutations. Overall, our result shows that a relatively long MD simulation (e.g. 100 ns) benefits the prediction accuracy for both MM/GBSA and MM/PBSA (with the best Pearson correlation coefficient between the predicted ∆∆G and the experimental data of ~ 0.44 for a challenging dataset). Further analyses shows that systems involving large perturbations (e.g. multiple mutations and large number of atoms change in the mutation site) are much easier to be accurately predicted since the algorithm works more sensitively to the large change of the systems. Besides, system-specific investigation reveals that conformational adjustment is needed to refine the micro-environment of the manually mutated systems and thus lead one to understand why longer MD simulation is necessary to improve the predicting result. The proposed strategy is expected to be applied in large-scale mutation effects investigation with interpretation.

Predicting the mutation effects of protein-ligand interactions via end-point binding free energy calculations: strategies and analyses

Calculation of Relative Binding Free Energy for Mutations in Protein Complexes: The Alchemical Path

An Efficient Method to Predict Protein Thermostability in Alanine Mutation

Computational studies of protein–drug binding affinity changes upon mutations in the drug target

Robust prediction of relative binding energies for protein-protein complex mutations using free energy perturbation calculations

Community-wide evaluation of methods for predicting the effect of mutations on protein-protein interactions.

Quantification of biases in predictions of protein–protein binding affinity changes upon mutations

An evaluation of machine learning/molecular mechanics end-state corrections with mechanical embedding to calculate relative protein-ligand binding free energies

Reliable Prediction of the Protein-Ligand Binding Affinity Using a Charge Penetration Corrected AMOEBA Force Field: A Case Study of Drug Resistance Mutations in Abl Kinase.

An Efficient Approach to the Accurate Prediction of Mutational Effects in Antigen Binding to the MHC1

Comparing Supervised Learning and Rigorous Approach for Predicting Protein Stability upon Point Mutations in Difficult Targets

Systematic Comparison and Prediction of the Effects of Missense Mutations on Protein-Dna and Protein-Rna Interactions

Accurate and Efficient Calculation of Protein-Protein Binding Free Energy-Interaction Entropy with Residue Type-Specific Dielectric Constants.

A Method for Efficient Calculation of Thermal Stability of Proteins Upon Point Mutations.

BindProfX: Assessing Mutation-Induced Binding Affinity Change by Protein Interface Profiles with Pseudo-Counts

SAAMBE-MEM: A Sequence-Based Method for Predicting Binding Free Energy Change upon Mutation in Membrane Protein-Protein Complexes

Assessing the Performance of the MM/PBSA and MM/GBSA Methods. 1. The Accuracy of Binding Free Energy Calculations Based on Molecular Dynamics Simulations

Identifying and Overcoming the Sampling Challenges in Relative Binding Free Energy Calculations of a Model Protein:Protein Complex

Prediction of Binding Free Energy of Protein–Ligand Complexes with a Hybrid Molecular Mechanics/Generalized Born Surface Area and Machine Learning Method

Boltzmann-Aligned Inverse Folding Model as a Predictor of Mutational Effects on Protein-Protein Interactions