Machine learning combines atomistic simulations to predict SARS-CoV-2 Mpro inhibitors from natural compounds

To date, the COVID-19 pandemic has still been infectious around the world, continuously causing social and economic damage on a global scale. One of the most important therapeutic targets for the treatment of COVID-19 is the main protease (Mpro) of SARS-CoV-2. In this study, we combined machine-learning (ML) model with atomistic simulations to computationally search for highly promising SARS-CoV-2 Mpro inhibitors from the representative natural compounds of the National Cancer Institute (NCI) Database. First, the trained ML model was used to scan the library quickly and reliably for possible Mpro inhibitors. The ML output was then confirmed using atomistic simulations integrating molecular docking and molecular dynamic simulations with the linear interaction energy scheme. The results turned out to show that there was evidently good agreement between ML and atomistic simulations. Ten substances were proposed to be able to inhibit SARS-CoV-2 Mpro. Seven of them have high-nanomolar affinity and are very potential inhibitors. The strategy has been proven to be reliable and appropriate for fast prediction of SARS-CoV-2 Mpro inhibitors, benefiting for new emerging SARS-CoV-2 variants in the future accordingly.

Binding free energy predictions of farnesoid X receptor (FXR) agonists using a linear interaction energy (LIE) approach with reliability estimation: application to the D3R Grand Challenge 2

Computational protein binding affinity prediction can play an important role in drug research but performing efficient and accurate binding free energy calculations is still challenging. In the context of phase 2 of the Drug Design Data Resource (D3R) Grand Challenge 2 we used our automated eTOX ALLIES approach to apply the (iterative) linear interaction energy (LIE) method and we evaluated its performance in predicting binding affinities for farnesoid X receptor (FXR) agonists. Efficiency was obtained by our pre-calibrated LIE models and molecular dynamics (MD) simulations at the nanosecond scale, while predictive accuracy was obtained for a small subset of compounds. Using our recently introduced reliability estimation metrics, we could classify predictions with higher confidence by featuring an applicability domain (AD) analysis in combination with protein-ligand interaction profiling. The outcomes of and agreement between our AD and interaction-profile analyses to distinguish and rationalize the performance of our predictions highlighted the relevance of sufficiently exploring protein-ligand interactions during training and it demonstrated the possibility to quantitatively and efficiently evaluate if this is achieved by using simulation data only.

eTOX ALLIES: an automated pipeLine for linear interaction energy-based simulations

BACKGROUND

Computational methods to predict binding affinities of small ligands toward relevant biological (off-)targets are helpful in prioritizing the screening and synthesis of new drug candidates, thereby speeding up the drug discovery process. However, use of ligand-based approaches can lead to erroneous predictions when structural and dynamic features of the target substantially affect ligand binding. Free energy methods for affinity computation can include steric and electrostatic protein-ligand interactions, solvent effects, and thermal fluctuations, but often they are computationally demanding and require a high level of supervision. As a result their application is typically limited to the screening of small sets of compounds by experts in molecular modeling.

RESULTS

We have developed eTOX ALLIES, an open source framework that allows the automated prediction of ligand-binding free energies requiring the ligand structure as only input. eTOX ALLIES is based on the linear interaction energy approach, an efficient end-point free energy method derived from Free Energy Perturbation theory. Upon submission of a ligand or dataset of compounds, the tool performs the multiple steps required for binding free-energy prediction (docking, ligand topology creation, molecular dynamics simulations, data analysis), making use of external open source software where necessary. Moreover, functionalities are also available to enable and assist the creation and calibration of new models. In addition, a web graphical user interface has been developed to allow use of free-energy based models to users that are not an expert in molecular modeling.

CONCLUSIONS

Because of the user-friendliness, efficiency and free-software licensing, eTOX ALLIES represents a novel extension of the toolbox for computational chemists, pharmaceutical scientists and toxicologists, who are interested in fast affinity predictions of small molecules toward biological (off-)targets for which protein flexibility, solvent and binding site interactions directly affect the strength of ligand-protein binding.

Using DFT methodology for more reliable predictive models: Design of inhibitors of Golgi α-Mannosidase II

Human Golgi α-mannosidase II (GMII), a zinc ion co-factor dependent glycoside hydrolase (E.C.3.2.1.114), is a pharmaceutical target for the design of inhibitors with anti-cancer activity. The discovery of an effective inhibitor is complicated by the fact that all known potent inhibitors of GMII are involved in unwanted co-inhibition with lysosomal α-mannosidase (LMan, E.C.3.2.1.24), a relative to GMII. Routine empirical QSAR models for both GMII and LMan did not work with a required accuracy. Therefore, we have developed a fast computational protocol to build predictive models combining interaction energy descriptors from an empirical docking scoring function (Glide-Schrödinger), Linear Interaction Energy (LIE) method, and quantum mechanical density functional theory (QM-DFT) calculations. The QSAR models were built and validated with a library of structurally diverse GMII and LMan inhibitors and non-active compounds. A critical role of QM-DFT descriptors for the more accurate prediction abilities of the models is demonstrated. The predictive ability of the models was significantly improved when going from the empirical docking scoring function to mixed empirical-QM-DFT QSAR models (Q(2)=0.78-0.86 when cross-validation procedures were carried out; and R(2)=0.81-0.83 for a testing set). The average error for the predicted ΔGbind decreased to 0.8-1.1kcalmol(-1). Also, 76-80% of non-active compounds were successfully filtered out from GMII and LMan inhibitors. The QSAR models with the fragmented QM-DFT descriptors may find a useful application in structure-based drug design where pure empirical and force field methods reached their limits and where quantum mechanics effects are critical for ligand-receptor interactions. The optimized models will apply in lead optimization processes for GMII drug developments.

A comparative study of binding affinities for 6,7-dimethoxy-4-pyrrolidylquinazolines as phosphodiesterase 10A inhibitors using the linear interaction energy method

The linear interaction energy (LIE) method was used to estimate the free energies of binding for a set of 27 pyrrolidylquinazoline derivatives as phosphodiesterase 10A inhibitors. Twenty-six X-ray crystal structures of phosphodiesterase 10A and two sampling methods, minimization and Hybrid Monte Carlo, were used to assess the affinity models based on the linear interaction energies. The best model was obtained based on the parameters α=0.16 and β=0.04, which represent non-polar and polar interactions, respectively, with a root mean square error (RMSE) of 0.42kcal/mol (R(2)=0.71) and 0.52kcal/mol (R(2)=0.86) for the training and test sets, respectively. In addition, the applicability domain of the model was investigated. After validation of the models, the best model was subsequently used in a virtual screening process, which resulted in a set of optimized compounds. The models developed in this study could be useful as filter for virtual screening and lead optimization processes for phosphodiesterase 10A drug developments.

Qgui: A high-throughput interface for automated setup and analysis of free energy calculations and empirical valence bond simulations in biological systems

Structural information and activity data has increased rapidly for many protein targets during the last decades. In this paper, we present a high-throughput interface (Qgui) for automated free energy and empirical valence bond (EVB) calculations that use molecular dynamics (MD) simulations for conformational sampling. Applications to ligand binding using both the linear interaction energy (LIE) method and the free energy perturbation (FEP) technique are given using the estrogen receptor (ERα) as a model system. Examples of free energy profiles obtained using the EVB method for the rate-limiting step of the enzymatic reaction catalyzed by trypsin are also shown. In addition, we present calculation of high-precision Arrhenius plots to obtain the thermodynamic activation enthalpy and entropy with Qgui from running a large number of EVB simulations.