GPU-Accelerated All-Atom Particle-Mesh Ewald Continuous Constant pH Molecular Dynamics in Amber

Abstract

Constant pH molecular dynamics (MD) simulations sample protonation states on the fly according to the conformational environment and user specified pH conditions; however, the current accuracy is limited due to the use of implicitsolvent models or a hybrid solvent scheme. Here, we report the first GPU-accelerated implementation, parametrization, and validation of the all-atom continuous constant pH MD (CpHMD) method with particle-mesh Ewald (PME) electrostatics in the Amber22 pmemd.cuda engine. The titration parameters for Asp, Glu, His, Cys, and Lys were derived for the CHARMM c22 and Amber ff14sb and ff19sb force fields. We then evaluated the PME-CpHMD method using the asynchronous pH replicaexchange titration simulations with the c22 force field for six benchmark proteins, including BBL, hen egg white lysozyme (HEWL), staphylococcal nuclease (SNase), thioredoxin, ribonuclease A (RNaseA), and human muscle creatine kinase (HMCK). The root-mean-square deviation from the experimental pK a 's of Asp, Glu, His, and Cys is 0.76 pH units, and the Pearson's correlation coefficient for the pK a shifts with respect to model values is 0.80. We demonstrated that a finite-size correction or much enlarged simulation box size can remove a systematic error of the calculated pK a 's and improve agreement with experiment. Importantly, the simulations captured the relevant biology in several challenging cases, e.g., the titration order of the catalytic dyad Glu35/Asp52 in HEWL and the coupled residues Asp19/Asp21 in SNase, the large pK a upshift of the deeply buried catalytic Asp26 in thioredoxin, and the large pK a downshift of the deeply buried catalytic Cys283 in HMCK. We anticipate that PME-CpHMD will offer proper pH control to improve the accuracies of MD simulations and enable mechanistic studies of proton-coupled dynamical processes that are ubiquitous in biology but remain poorly understood due to the lack of experimental tools and limitation of current MD simulations.

Accurate Host–Guest Binding Free Energies Using the AMOEBA Polarizable Force Field

“…The effect of finite box size on binding free energy is well recognized. ,, A 50 Å box was used in our initial submissions for the SAMPL challenge. To consider box size dependence, we chose the TEMOA-H3 system and computed the binding free energy at 50, 60, 70, and 80 Å.…”

Basis for Accurate Protein pKa Prediction with Machine Learning

“…To explore the potential coupling between conformation and protonation state, it could be necessary to generate structural fluctuations for the target protein by molecular simulation and calculate the p K a ’s by DeepKa. Inspired by implicit-solvent CpHMD, which is, however, limited to soluble proteins, hybrid-solvent , or fully explicit-solvent − CpHMD could be employed to produce a p K a database exclusively for the development of AI-based membrane protein p K a predictors under the framework of DeepKa. Finally, to uncover the molecular determinants not captured by the present deep learning scheme, it is worth looking into structures that account for large p K a errors.…”

In Silico Identification of Lead Compounds for Pseudomonas Aeruginosa PqsA Enzyme: Computational Study to Block Biofilm Formation

“…All visualization was conducted in PyMol [ 34 ]. All-atom MD simulations and essential dynamics analysis were conducted in AMBER version 2022 [ 35 ]. The LEaP module was used to integrate hydrogen atoms into the crystal structure.…”