Relative Binding Free Energies Research Articles

Fully automated screening of compound libraries in drug discovery using docking and molecular dynamics.

In the initial stage of drug discovery, hundreds of thousands of compounds are screened in high-throughput assays to find a small number of hits. To save time and resources, we develop a computational workflow that ranks drug candidates by their binding affinity to the target protein. First, all compounds in a library are docked to the protein. After this filtering step, the drug candidates with the poses scoring best are grouped into pairs, and their relative free energy of binding to the protein is calculated using the accelerated weight histogram method (AWH) for alchemical transformations. Thanks to random moves between λ-windows, a single AWH simulation samples the complete alchemical path connecting two ligands. The probability to visit a λ-window depends on its estimated free energy such that the sampling of the alchemical transformation and the free energy estimate improve each other in a positive feedback loop. To fully integrate docking and free energy calculations, we implemented routines that generate hybrid topologies for drug candidates without human intervention, i.e., they assign force field parameters (AMBER and CHARMM) to the compound and determine the smallest set of atoms to be alchemically transformed for a given compound pair. Being able to convert docking poses into input files for the molecular dynamics simulations in an automated fashion, we extensively test AWH's robustness to (sub-optimal) initial configurations on a benchmark with 13 proteins and approximately 500 ligand pairs for which experimental relative binding free energies are available. In return, free energies obtained from MD for protein-ligand complexes in the PDBbind data set are used as training data for a machine-learning model improving the scoring of docking poses such that both docking and free energy calculations benefit from integration in the workflow.

Regioselective metabolism of cannabinoids by cytochrome P450 2J2.

Cannabichromene (CBC) is a major non-psychoactive phytocannabinoid extracted from the cannabis plant. CBC has been shown to have anti-inflammatory effects; hence it can serve as a non-narcotic alternative to tetrahydrocannabinol (THC). Nevertheless, the metabolic pathways and metabolites of CBC have not been fully characterized. Metabolism of phytocannabinoids typically occurs via a superfamily of hemoproteins called cytochrome P450s (CYP450s). Our liquid chromatography-mass spectrometry (LC/MS) experiments show that CBC is metabolized to three different products by various human CYP450s. We identified that CYP450 2J2 (CYP2J2) predominantly produces 8'-hydroxy-CBC, an allylic hydroxide, out of the three metabolites. To investigate the CBC metabolic regioselectivity in CYP2J2, we leveraged atomistic molecular dynamics (MD) and modeling of the system. Using 200 unique conformations of CYP2J2 obtained from MD, we performed ensemble molecular docking of CBC to reveal three putative binding modes. We then assessed the stability of the obtained CBC binding modes by simulating them individually for 1 μs. Our simulations captured an average distance of 4.8 Å between the CYP2J2 heme and the 8’ carbon atom on CBC for one of the predominant binding modes, at which the allylic hydroxide metabolite occurs experimentally. At the regioselective binding mode, we observe conserved hydrogen bonding networks, mainly between the CBC hydroxyl group and the backbone oxygen of I487, as well as N379, stabilized by a water molecule. We further corroborated the stability of the regioselective binding modes by calculating relative binding free energies using alchemical free energy perturbation calculations. In conjunction, our computational and experimental methodologies elucidate a likely mode for regioselective binding and metabolism of CBC by CYP2J2, furthering the way towards accessible plant-based non-psychoactive anti-inflammatory therapeutics.

TIES 2.0: A Dual-Topology Open Source Relative Binding Free Energy Builder with Web Portal

Relative binding free energy (RBFE) calculations are widely used to aid the process of drug discovery. TIES, Thermodynamic Integration with Enhanced Sampling, is a dual-topology approach to RBFE calculations with support for NAMD and OpenMM molecular dynamics engines. The software has been thoroughly validated on publicly available datasets. Here we describe the open source software along with a web portal (https://ccs-ties.org) that enables users to perform such calculations correctly and rapidly.

Anesthesia, Coming of Age in the World of Modern In Silico Drug Design.

Anesthesia, Coming of Age in the World of Modern In Silico Drug Design.

ChemFlow─From 2D Chemical Libraries to Protein-Ligand Binding Free Energies.

The accurate prediction of protein-ligand binding affinities is a fundamental problem for the rational design of new drug entities. Current computational approaches are either too expensive or inaccurate to be effectively used in virtual high-throughput screening campaigns. In addition, the most sophisticated methods, e.g., those based on configurational sampling by molecular dynamics, require significant pre- and postprocessing to provide a final ranking, which hinders straightforward applications by nonexpert users. We present a novel computational platform named ChemFlow to bridge the gap between 2D chemical libraries and estimated protein-ligand binding affinities. The software is designed to prepare a library of compounds provided in SMILES or SDF format, dock them into the protein binding site, and rescore the poses by simplified free energy calculations. Using a data set of 626 protein-ligand complexes and GPU computing, we demonstrate that ChemFlow provides relative binding free energies with an RMSE < 2 kcal/mol at a rate of 1000 ligands per day on a midsize computer cluster. The software is publicly available at https://github.com/IFMlab/ChemFlow.

Active Learning Guided Drug Design Lead Optimization Based on Relative Binding Free Energy Modeling.

In silico identification of potent protein inhibitors commonly requires prediction of a ligand binding free energy (BFE). Thermodynamics integration (TI) based on molecular dynamics (MD) simulations is a BFE calculation method capable of acquiring accurate BFE, but it is computationally expensive and time-consuming. In this work, we have developed an efficient automated workflow for identifying compounds with the lowest BFE among thousands of congeneric ligands, which requires only hundreds of TI calculations. Automated machine learning (AutoML) orchestrated by active learning (AL) in an AL-AutoML workflow allows unbiased and efficient search for a small set of best-performing molecules. We have applied this workflow to select inhibitors of the SARS-CoV-2 papain-like protease and were able to find 133 compounds with improved binding affinity, including 16 compounds with better than 100-fold binding affinity improvement. We obtained a hit rate that outperforms that expected of traditional expert medicinal chemist-guided campaigns. Thus, we demonstrate that the combination of AL and AutoML with free energy simulations provides at least 20× speedup relative to the naïve brute force approaches.

The role of loop dynamics in the prediction of ligand-protein binding enthalpy.

The enthalpic and entropic components of ligand-protein binding free energy reflect the interactions and dynamics between ligand and protein. Despite decades of study, our understanding and hence our ability to predict these individual components remains poor. In recent years, there has been substantial effort and success in the prediction of relative and absolute binding free energies, but the prediction of the enthalpic (and entropic) contributions in biomolecular systems remains challenging. Indeed, it is not even clear what kind of performance in terms of accuracy could currently be obtained for such systems. It is, however, relatively straight-forward to compute the enthalpy of binding. We thus evaluated the performance of absolute enthalpy of binding calculations using molecular dynamics simulation for ten inhibitors against a member of the bromodomain family, BRD4-1, against isothermal titration calorimetry data. Initial calculations, with the AMBER force-field showed good agreement with experiment (R2 = 0.60) and surprisingly good accuracy with an average of root-mean-square error (RMSE) = 2.49 kcal mol-1. Of the ten predictions, three were obvious outliers that were all over-predicted compared to experiment. Analysis of various simulation factors, including parameterization, buffer concentration and conformational dynamics, revealed that the behaviour of a loop (the ZA loop on the periphery of the binding site) strongly dictates the enthalpic prediction. Consistent with previous observations, the loop exists in two distinct conformational states and by considering one or the other or both states, the prediction for the three outliers can be improved dramatically to the point where the R2 = 0.95 and the accuracy in terms of RMSE improves to 0.90 kcal mol-1. However, performance across force-fields is not consistent: if OPLS and CHARMM are used, different outliers are observed and the correlation with the ZA loop behaviour is not recapitulated, likely reflecting parameterization as a confounding problem. The results provide a benchmark standard for future study and comparison.

Computational Evaluation of Peptide-Protein Binding Affinities: Application of Potential of Mean Force Calculations to SH2 Domains.

Many biological functions are mediated by protein-protein interactions (PPIs), often involving specific structural modules, such as SH2 domains. Inhibition of PPIs is a pharmaceutical strategy of growing importance. However, a major challenge in the design of PPI inhibitors is the large interface involved in these interactions, which, in many cases, makes inhibition by small organic molecules ineffective. Peptides, which cover a wide range of dimensions and can be opportunely designed to mimic protein sequences at PPI interfaces, represent a valuable alternative to small molecules. Computational techniques able to predict the binding affinity of peptides for the target domain or protein represent a crucial stage in the workflow for the design of peptide-based drugs. This chapter describes a protocol to obtain the potential of mean force (PMF) for peptide-SH2 domain binding, starting from umbrella sampling (US) molecular dynamics (MD) simulations. The PMF profiles can be effectively used to predict the relative standard binding free energies of different peptide sequences.

An Open Source Graph-Based Weighted Cycle Closure Method for Relative Binding Free Energy Calculations.

Free energy perturbation-relative binding free energy (FEP-RBFE) prediction has shown its reliability and accuracy in the prediction of protein-ligand binding affinities, which plays a fundamental role in structure-based drug design. In FEP-RBFE predictions, the calculation of each mutation path is associated with a statistical error, and cycle closure (cc) has proven to be an effective method in improving the calculation accuracy by correcting the hysteresis (summation of errors) of each closed cycle to the theoretical value 0. However, a primary hypothesis was made in the current cycle closure method that the hysteresis is evenly distributed to all paths, which is unlikely to be true in practice and may limit the further improvement of the calculation accuracy when better error estimation methods are available. Moreover, being a closed source software makes the current cycle closure method unachievable in many studies. In this paper, a newly implemented open source graph-based weighted cycle closure (wcc) algorithm was developed and introduced, not only including functions from the original cc method but also containing a new wcc method which can consider different error contributions from different paths and further improve the calculation accuracy. The wcc program also provides a new path-independent molecular error calculation method, which can be quite useful in many studies (like structure-activity relationship (SAR)) compared with the path-dependent method of the original cc program.

Study of potential inhibition of the estrogen receptor α by cannabinoids using an in silico approach: Agonist vs antagonist mechanism

Breast cancer is the main cancer type with more than 2.2 million cases in 2020, and is the principal cause of death in women; with 685000 deaths in 2020 worldwide. The estrogen receptor is involved at least in 70% of breast cancer diagnoses, and the agonist and antagonist properties of the drug in this receptor play a pivotal role in the control of this illness. This work evaluated the agonist and antagonist mechanisms of 30 cannabinoids by employing molecular docking and dynamic simulations. Compounds with docking scores < −8 kcal/mol were analyzed by molecular dynamic simulation at 300 ns, and relevant insights are given about the protein's structural changes, centered on the helicity in alpha-helices H3, H8, H11, and H12. Cannabicitran was the cannabinoid that presented the best relative binding-free energy (−34.96 kcal/mol), and based on rational modification, we found a new natural-based compound with relative binding-free energy (−44.83 kcal/mol) better than the controls hydroxytamoxifen and acolbifen. Structure modifications that could increase biological activity are suggested.

Protein-Ligand Binding Free-Energy Calculations with ARROW─A Purely First-Principles Parameterized Polarizable Force Field.

Protein-ligand binding free-energy calculations using molecular dynamics (MD) simulations have emerged as a powerful tool for in silico drug design. Here, we present results obtained with the ARROW force field (FF)─a multipolar polarizable and physics-based model with all parameters fitted entirely to high-level ab initio quantum mechanical (QM) calculations. ARROW has already proven its ability to determine solvation free energy of arbitrary neutral compounds with unprecedented accuracy. The ARROW FF parameterization is now extended to include coverage of all amino acids including charged groups, allowing molecular simulations of a series of protein-ligand systems and prediction of their relative binding free energies. We ensure adequate sampling by applying a novel technique that is based on coupling the Hamiltonian Replica exchange (HREX) with a conformation reservoir generated via potential softening and nonequilibrium MD. ARROW provides predictions with near chemical accuracy (mean absolute error of ∼0.5 kcal/mol) for two of the three protein systems studied here (MCL1 and Thrombin). The third protein system (CDK2) reveals the difficulty in accurately describing dimer interaction energies involving polar and charged species. Overall, for all of the three protein systems studied here, ARROW FF predicts relative binding free energies of ligands with a similar accuracy level as leading nonpolarizable force fields.

Optimizing active learning for free energy calculations

While Relative Binding Free Energy (RBFE) calculations have become a mainstay in lead optimization programs, the computational expense of performing these calculations has limited their broader application. Active learning (AL), a machine learning method used to direct a search iteratively, has explored larger chemical libraries using RBFE calculations. While AL has been successfully applied, there has not been a systematic study of the impact of parameter settings on the performance of AL. To address this gap, we have generated an exhaustive dataset of RBFE calculations on 10,000 congeneric molecules. We used this dataset to explore the impact of several AL design choices, including the number of molecules sampled at each iteration, the method used to select an initial sample, the method used to build a machine learning model, and the acquisition function that defines the balance between exploration and exploitation in the search. Our studies demonstrated that the performance of AL is largely insensitive to the specific machine learning method and acquisition functions used. In our studies, the most significant factor impacting performance was the number of molecules sampled at each iteration where selecting too few molecules hurts performance. Under the best conditions, we were able to identify 75% of the 100 top scoring molecules by sampling only 6% of the dataset. We hope that the dataset of 10K molecules will provide the basis for future studies exploring additional AL strategies. The source code and supporting data for the work are available at https://github.com/google-research/google-research/tree/master/al_for_fep.

Encapsulation dynamics of neuromuscular blocking drugs by sugammadex.

The clinical actions of sugammadex have been well studied, but the detailed molecular mechanism of the drug encapsulation process has not been systematically documented. We hypothesized that sugammadex would attract rocuronium and vecuronium via interaction with the sugammadex side-chain "tentacles", as previously suggested. Computational molecular dynamics simulations were done to investigate docking of sugammadex with rocuronium and vecuronium. To validate these methods, strength of binding was assessed between sugammadex and a heterogeneous group of 9 other drugs, whose binding affinities have been experimentally determined. These observations hinted that high concentrations of unbound sugammadex could bind to propofol, potentially altering its pharmacokinetic profile. We tested this experimentally in in vitro cortical slices. Sugammadex encapsulation of rocuronium involved a sequential progression down a series of metastable states. After initially binding beside the sugammadex molecule (mean±SD center-of-mass distance = 1.17 ± 0.13 nm), rocuronium then moved to the opposite side to that hypothesized, where it optimally aligned with the 16 hydroxyl groups (distance 0.82 ± 0.04 nm); before entering the sugammadex cavity to achieve energetically stable encapsulation by ~120 ns (distance 0.35 ± 0.12 nm). Vecuronium formed fewer hydrogen bonds with sugammadex than did rocuronium, hence was less avidly bound. For the other molecules, our computational results showed good agreement with the available experimental data, showing a clear bi-logarithmic relation between the relative binding free energy and the association constant (R2=0.98). Weaker binding was manifest by periodic unbinding. The brain slice results confirmed the presence of a weak propofol-sugammadex interaction. Computational simulations demonstrate the dynamics of neuromuscular blocking drug encapsulation by sugammadex occurring from the opposite direction to that hypothesized, and also how high concentrations of unbound sugammadex can potentially weakly bind to other drugs given during general anesthesia.

Practical Guidance for Consensus Scoring and Force Field Selection in Protein-Ligand Binding Free Energy Simulations.

The advances in ligand binding affinity prediction have been fostered by system generation tools and improved force fields (FFs). CHARMM-GUI Free Energy Calculator provides input and postprocessing scripts for AMBER-TI free energy calculations with various FFs. In this study, we used 12 different FF combinations (ff14SB and ff19SB for protein, GAFF2.2 and OpenFF for ligand, and TIP3P, TIP4PEW, and OPC for water) to calculate relative binding free energies (ΔΔGbind) for 80 alchemical transformations (among the JACS benchmark set) with different numbers of λ windows (12 or 21) and simulation times (1, 5, or 10 ns). Our results show that 12 λ windows and 5 ns simulation time for each window are sufficient to obtain reliable ΔΔGbind with 4 independent runs for the current benchmark set. While there is no statistically noticeable performance difference among 12 different FF combinations compared to the experimental values, a combination of ff14SB + GAFF2.2 + TIP3P FFs appears to be best with a mean unsigned error of 0.87 [0.69, 1.07] kcal/mol, a root-mean-square error of 1.22 [0.94, 1.50] kcal/mol, a Pearson's correlation of 0.64 [0.52, 0.76], a Spearman's correlation of 0.73 [0.58, 0.83], and a Kendell's correlation of 0.54 [0.42, 0.64]. This large-scale ΔΔGbind calculation study provides useful information about ΔΔGbind prediction with different AMBER FF combinations and presents valuable suggestions for FF selection in AMBER-TI ΔΔGbind calculations.

Open Force Field BespokeFit: Automating Bespoke Torsion Parametrization at Scale.

The development of accurate transferable force fields is key to realizing the full potential of atomistic modeling in the study of biological processes such as protein-ligand binding for drug discovery. State-of-the-art transferable force fields, such as those produced by the Open Force Field Initiative, use modern software engineering and automation techniques to yield accuracy improvements. However, force field torsion parameters, which must account for many stereoelectronic and steric effects, are considered to be less transferable than other force field parameters and are therefore often targets for bespoke parametrization. Here, we present the Open Force Field QCSubmit and BespokeFit software packages that, when combined, facilitate the fitting of torsion parameters to quantum mechanical reference data at scale. We demonstrate the use of QCSubmit for simplifying the process of creating and archiving large numbers of quantum chemical calculations, by generating a dataset of 671 torsion scans for druglike fragments. We use BespokeFit to derive individual torsion parameters for each of these molecules, thereby reducing the root-mean-square error in the potential energy surface from 1.1 kcal/mol, using the original transferable force field, to 0.4 kcal/mol using the bespoke version. Furthermore, we employ the bespoke force fields to compute the relative binding free energies of a congeneric series of inhibitors of the TYK2 protein, and demonstrate further improvements in accuracy, compared to the base force field (MUE reduced from 0.560.390.77 to 0.420.280.59 kcal/mol and R2 correlation improved from 0.720.350.87 to 0.930.840.97).

Dual A1/A3 Adenosine Receptor Antagonists: Binding Kinetics and Structure-Activity Relationship Studies Using Mutagenesis and Alchemical Binding Free Energy Calculations.

Drugs targeting adenosine receptors (AR) can provide treatment for diseases. We report the identification of 7-(phenylamino)-pyrazolo[3,4-c]pyridines L2-L10, A15, and A17 as low-micromolar to low-nanomolar A1R/A3R dual antagonists, with 3-phenyl-5-cyano-7-(trimethoxyphenylamino)-pyrazolo[3,4-c]pyridine (A17) displaying the highest affinity at both receptors with a long residence time of binding, as determined using a NanoBRET-based assay. Two binding orientations of A17 produce stable complexes inside the orthosteric binding area of A1R in molecular dynamics (MD) simulations, and we selected the most plausible orientation based on the agreement with alanine mutagenesis supported by affinity experiments. Interestingly, for drug design purposes, the mutation of L2506.51 to alanine increased the binding affinity of A17 at A1R. We explored the structure-activity relationships against A1R using alchemical binding free energy calculations with the thermodynamic integration coupled with the MD simulation (TI/MD) method, applied on the whole G-protein-coupled receptor-membrane system, which showed a good agreement (r = 0.73) between calculated and experimental relative binding free energies.

Computational molecular interaction between SARS-CoV-2 main protease and theaflavin digallate using free energy perturbation and molecular dynamics

Our objective was to identify the molecule which can inhibit SARS-CoV-2 main protease and can be easily procured. Natural products may provide such molecules and can supplement the current custom chemical synthesis-based drug discovery for this objective. A combination of docking approaches, scoring functions, classical molecular dynamic simulation, binding pose metadynamics, and free energy perturbation calculations have been employed in this study. Theaflavin digallate has been observed in top-scoring compounds after the three independent virtual screening simulations of 598435 compounds (unique 27256 chemical entities). The main protease-theaflavin digallate complex interacts with critical active site residues of the main protease in molecular dynamics simulation independent of the explored computational framework, simulation time, initial structure, and force field used. Theaflavin digallate forms approximately three hydrogen bonds with Glutamate166 of main protease, primarily through hydroxyl groups in the benzene ring of benzo(7)annulen-6-one, along with other critical residues. Glu166 is the most critical amino acid for main protease dimerization, which is necessary for catalytic activity. The estimated binding free energy, calculated by Amber and Schrodinger MMGBSA module, reflects a high binding free energy between theaflavin digallate and main protease. Binding pose metadynamics simulation shows the highly persistent H-bond and a stable pose for the theaflavin digallate-main protease complex. Using method control, experimental controls, and test set, alchemical transformation studies confirm high relative binding free energy of theaflavin digallate with the main protease. Computational molecular interaction suggests that theaflavin digallate can inhibit the main protease of SARS-CoV-2.

Leveraging the sampling efficiency of RE-EDS in OpenMM using a shifted reaction-field with an atom-based cutoff.

Replica-exchange enveloping distribution sampling (RE-EDS) is a pathway-independent multistate free-energy method currently implemented in the GROMOS software package for molecular dynamics (MD) simulations. It has a high intrinsic sampling efficiency as the interactions between the unperturbed particles have to be calculated only once for multiple end-states. As a result, RE-EDS is an attractive method for the calculation of relative solvation and binding free energies. An essential requirement for reaching this high efficiency is the separability of the nonbonded interactions into solute-solute, solute-environment, and environment-environment contributions. Such a partitioning is trivial when using a Coulomb term with a reaction-field (RF) correction to model the electrostatic interactions but not when using lattice-sum schemes. To avoid cutoff artifacts, the RF correction is typically used in combination with a charge-group-based cutoff, which is not supported by most small-molecule force fields as well as other MD engines. To address this issue, we investigate the combination of RE-EDS simulations with a recently introduced RF scheme including a shifting function that enables the rigorous calculation of RF electrostatics with atom-based cutoffs. The resulting approach is validated by calculating solvation free energies with the generalized AMBER force field in water and chloroform using both the GROMOS software package and a proof-of-concept implementation in OpenMM.

Relative binding free energy calculations with transformato: A molecular dynamics engine-independent tool.

We present the software package transformato for the setup of large-scale relative binding free energy calculations. Transformato is written in Python as an open source project (https://github.com/wiederm/transformato); in contrast to comparable tools, it is not closely tied to a particular molecular dynamics engine to carry out the underlying simulations. Instead of alchemically transforming a ligand L 1 directly into another L 2, the two ligands are mutated to a common core. Thus, while dummy atoms are required at intermediate states, in particular at the common core state, none are present at the physical endstates. To validate the method, we calculated 76 relative binding free energy differences for five protein–ligand systems. The overall root mean squared error to experimental binding free energies is 1.17 kcal/mol with a Pearson correlation coefficient of 0.73. For selected cases, we checked that the relative binding free energy differences between pairs of ligands do not depend on the choice of the intermediate common core structure. Additionally, we report results with and without hydrogen mass reweighting. The code currently supports OpenMM, CHARMM, and CHARMM/OpenMM directly. Since the program logic to choose and construct alchemical transformation paths is separated from the generation of input and topology/parameter files, extending transformato to support additional molecular dynamics engines is straightforward.

In Silico Positional Analogue Scanning with Amber GPU-TI.

Positional analogue scanning (PAS) is an accepted strategy for multiparameter lead optimization (MPO) in drug discovery. Small structural changes as introduced by PAS can lead to 10-fold changes in binding potency in ∼10-20% of cases, a significant parameter shift irrespective of other MPO objectives. Sometimes performing a complete PAS is challenging due to resource and time constraints, building block availability, or difficulty in synthesis. Calculating relative binding free energies (RBFEs) for all positions can contribute to prioritizing the most promising analogues for synthesis. We tested a well-established RBFE calculation method, Amber GPU-TI, for 20 positional analogue scans in 14 test systems (cyclin-dependent kinase 8 (CDK8), hepatitis C virus nonstructural protein 5B (HCV NS5B), tankyrase, RAC-α serine/threonine-protein kinase (Akt), phosphodiesterase 1B (PDE1B), orexin/hypocretin receptor type 1 (OX1R), orexin/hypocretin receptor type 2 (OX2R), histone acetyltransferase K (lysine) acetyltransferase 6A (KAT6A), peroxisome proliferator-activated receptor γ (PPARγ), extracellular signal-regulated kinases (ERK1/2), coactivator-associated arginine methyltransferase 1 (PRMT4), αvβ6, bromodomain 1 (BD1), human immunodeficiency virus-1 (HIV-1) entry) involving nitrogen, methyl, halogen, methoxy, and hydroxyl scans with at least four analogues per set. Among the 66 analogue positions explored, we found that in 18 cases Amber GPU-TI calculations predicted a more than 10-fold change in potency. In all of these cases, the experimentally observed direction of potency changes agreed with the predictions. In 16 cases, more than 10-fold changes in experimental potency were observed. Again, in all of these cases, Amber GPU-TI predicted the direction of the potency changes correctly. In none of these cases would a decision made for or against synthesis based on a 10-fold change in potency have resulted in missing an important analogue. Therefore, in silico RBFE calculations using Amber GPU-TI can meaningfully contribute to the prioritization of positional analogues before synthesis.