Thermodynamic properties are often modeled by classical force fields which describe the interactions on the atomistic scale. Molecular simulations are used for retrieving thermodynamic data from such models, and many simulation techniques and computer codes are available for that purpose. In the present round robin study, the following fundamental question is addressed: Will different user groups working with different simulation codes obtain coinciding results within the statistical uncertainty of their data? A set of 24 simple simulation tasks is defined and solved by five user groups working with eight molecular simulation codes: DL_POLY, GROMACS, IMC, LAMMPS, ms2, NAMD, Tinker, and TOWHEE. Each task consists of the definition of (1) a pure fluid that is described by a force field and (2) the conditions under which that property is to be determined. The fluids are four simple alkanes: ethane, propane, n-butane, and iso-butane. All force fields consider internal degrees of freedom: OPLS, TraPPE, and a modified OPLS version with bond stretching vibrations. Density and potential energy are determined as a function of temperature and pressure on a grid which is specified such that all states are liquid. The user groups worked independently and reported their results to a central instance. The full set of results was disclosed to all user groups only at the end of the study. During the study, the central instance gave only qualitative feedback. The results reveal the challenges of carrying out molecular simulations. Several iterations were needed to eliminate gross errors. For most simulation tasks, the remaining deviations between the results of the different groups are acceptable from a practical standpoint, but they are often outside of the statistical errors of the individual simulation data. However, there are also cases where the deviations are unacceptable. This study highlights similarities between computer experiments and laboratory experiments, which are both subject not only to statistical error but also to systematic error.

The calculation of solvation free energies ΔG by molecular simulations is of great interest as they are linked to other physical properties such as relative solubility, partition coefficient, and activity coefficient. However, shortcomings in molecular models can lead to ΔG deviations from experimental data. Various studies have demonstrated the impact of partial charges on free energy results. Consequently, calculation methods for partial charges aimed at more accurate ΔG predictions are the subject of various studies in the literature. Here we compare two methods to derive partial charges for the general AMBER force field (GAFF), i.e. the default RESP as well as the physically motivated IPolQ-Mod method that implicitly accounts for polarization costs. We study 29 solutes which include characteristic functional groups of drug-like molecules, and 12 diverse solvents were examined. In total, we consider 107 solute/solvent pairs including two water models TIP3P and TIP4P/2005. Comparison with experimental results yields better agreement for TIP3P, regardless of the partial charge scheme. The overall performance of GAFF/RESP and GAFF/IPolQ-Mod is similar, though specific shortcomings in the description of ΔG for both RESP and IPolQ-Mod can be identified. However, the high correlation between free energies obtained with GAFF/RESP and GAFF/IPolQ-Mod demonstrates the compatibility between the modified charges and remaining GAFF parameters.

No abstract

<p class="ADMETabstracttext">Rational drug design featuring explicit solubility considerations can greatly benefit from molecular dynamics simulations, as they allow for the prediction of the Gibbs free energy of solvation and thus relative solubilities. In our previous work (A. Mecklenfeld, G. Raabe. J. Chem. Theory Comput. <strong>13 </strong>no. 12 (2017) 6266–6274), we have compared solvation free energy results obtained with the General Amber Force Field (GAFF) and its default restrained electrostatic potential (RESP) partial charges to those obtained by modified implicitly polarized charges (IPolQ-Mod) for an implicit representation of impactful polarization effects. In this work, we have adapted Lennard-Jones parameters for GAFF atom types in combination with IPolQ-Mod to further improve the accuracies of solvation free energy and liquid density predictions. We thereby focus on prominent atom types in common drugs. For the refitting, 357 respectively 384 systems were considered for free energies and densities and validation was performed for 142 free energies and 100 densities of binary mixtures. By the in-depth comparison of simulation results for default GAFF, GAFF with IPolQ-Mod and our new set of parameters, which we label GAFF/IPolQ-Mod+LJ-Fit, we can clearly highlight the improvements of our new model for the description of both relative solubilities and fluid phase behaviour.</p>

No abstract