Abstract
Objective: The selection of appropriate metal parameters for molecular dynamics simulations is a significant challenge. Therefore, this review aims to provide in-depth insights valuable for the optimization of parameter selection in the context of chemical simulations.
Methods: A total of 550 scientific articles were collected from pubmed and science direct databases from 2009 to 2024, resulting in the inclusion of 60 full studies for review. The selection process of preferred reporting items for systematic reviews and meta-analyses (PRISMA) was utilized, enabling the conduction of an initial screening of articles by use of the Rayyan web-based application.
Results: This study found that the modeling and parameterization of metal proteins were categorized into bonded and non-bonded models. The Bonded Model incorporates MCPB, a Python-based software that facilitates parameter construction for over 80 metal ions and force fields in molecular dynamics simulations. The non-bonded model evaluates metals in proteins, such as zinc, nickel, magnesium, cobalt, iron, and cadmium by using AMBER force field and the Seminario method. The 12-6 lennard-Jones (LJ) non-bonded model is suitable for divalent, trivalent, and tetravalent metals, with Zinc parameters being compared for accuracy. Additionally, the force fields suitable for modeling unbound metal proteins include AMBER FF19SB, FF14SB, ff9X, CHARMM36, CHARMM22, CHARMM27, and CHARMM-Metal.
Conclusion: This study found that the modeling and parameterization of metal proteins were categorized into bonded and non-bonded models. molecular dynamics (MD) simulations can be conducted using various methods, such as classical molecular dynamics, umbrella sampling, quantum mechanics-discrete molecular dynamics (QM/DMD), stochastic boundary molecular dynamics (SBMD), steered molecular dynamics (SMD), gaussian accelerated molecular dynamics (GaMD) and random acceleration molecular dynamics (RAMD).
Publisher
Innovare Academic Sciences Pvt Ltd