Free Energy Perturbation Research Articles

Application of the Linear Interaction Energy Method to Nitric Oxide Synthase Structure-Based Inhibitor Design.

The overproduction of nitric oxide by neuronal nitric oxide synthase (nNOS) is associated with several neuropathological conditions. As a result, inhibition of nNOS is a desirable therapeutic goal while avoiding the inhibition of endothelial NOS (eNOS) given its essential role in maintaining cardiovascular tone. Designing inhibitors with high specificity for nNOS over eNOS is challenging given the close similarity in the active site structure of all mammalian NOS isoforms. Computational methods like free energy perturbation (FEP) and thermodynamic integration (TI) offer attractive avenues for rational drug design, but application of these methods to NOS is hindered by several challenges, including proper handling of highly charged inhibitors with diverse structures as well as computational expense. To address these issues, we present a simplified approach combining continuum dielectric generalized born (GB) solvent models with linear interaction energy (LIE) calculations. Our method demonstrates excellent agreement with experimental data for charged inhibitors targeting mammalian NOS isoforms (mNOS). Our results highlight the utility of the GB-LIE method as a promising tool for screening NOS inhibitors and potentially other protein targets with charged active sites and diverse inhibitor structures.

The Discovery of MORF-627, a Highly Selective Conformationally-Biased Zwitterionic Integrin αvβ6 Inhibitor for Fibrosis.

Inhibition of integrin αvβ6 is a promising approach to the treatment of fibrotic disease such as idiopathic pulmonary fibrosis. Screening a small library combining head groups that stabilize the bent-closed conformation of integrin αIIbβ3 with αv integrin binding motifs resulted in the identification of hit compounds that bind the bent-closed conformation of αvβ6. Crystal structures of these compounds bound to αvβ6 and related integrins revealed opportunities to increase potency and selectivity, and these efforts were accelerated using accurate free energy perturbation (FEP+) calculations. Optimization of PK parameters including permeability, bioavailability, clearance, and half-life resulted in the discovery of development candidate MORF-627, a highly selective inhibitor of αvβ6 that stabilizes the bent-closed conformation and has good oral PK. Unfortunately, the compound showed toxicity in a 28-day NHP safety study, precluding further development. Nevertheless, MORF-627 is a useful tool compound for studying the biology of integrin αvβ6.

"Pseudosubstrate Envelope"/Free Energy Perturbation-Guided Design and Mechanistic Investigations of Benzothiazole HIV Capsid Modulators with High Ligand Efficiency.

Based on our proposed "pseudosubstrate envelope" concept, 25 benzothiazole-bearing HIV capsid protein (CA) modulators were designed and synthesized under the guidance of free energy perturbation technology. The most potent compound, IC-1k, exhibited an EC50 of 2.69 nM against HIV-1, being 393 times more potent than the positive control PF74. Notably, IC-1k emerged as the highest ligand efficiency (LE = 0.32) HIV CA modulator, surpassing that of the approved drug lenacapavir (LE = 0.21). Surface plasmon resonance assay and crystallographic analysis confirmed that IC-1k targeted HIV-1 CA within the chemical space of the "pseudosubstrate envelope". Further mechanistic studies revealed a dual-stage inhibition profile: IC-1k disrupted early-stage capsid-host-factor interactions and promoted late-stage capsid misassembly. Preliminary pharmacokinetic evaluations demonstrated significantly improved metabolic stability in human liver microsomes for IC-1k (T1/2 = 91.3 min) compared to PF74 (T1/2 = 0.7 min), alongside a favorable safety profile. Overall, IC-1k presents a promising lead compound for further optimization.

Polymorph sampling with coupling to extended variables: enhanced sampling of polymorph energy landscapes and free energy perturbation of polymorph ensembles.

A novel approach to computationally enhance the sampling of molecular crystal structures is proposed and tested. This method is based on the use of extended variables coupled to a Monte Carlo based crystal polymorph generator. Inspired by the established technique of quasi-random sampling of polymorphs using the rigid molecule constraint, this approach represents molecular clusters as extended variables within a thermal reservoir. Polymorph unit-cell variables are generated using pseudo-random sampling. Within this framework, a harmonic coupling between the extended variables and polymorph configurations is established. The extended variables remain fixed during the inner loop dedicated to polymorph sampling, enforcing a stepwise propagation of the extended variables to maintain system exploration. The final processing step results in a polymorph energy landscape, where the raw structures sampled to create the extended variable trajectory are re-optimized without the thermal coupling term. The foundational principles of this approach are described and its effectiveness using both a Metropolis Monte Carlo type algorithm and modifications that incorporate replica exchange is demonstrated. A comparison is provided with pseudo-random sampling of polymorphs for the molecule coumarin. The choice to test a design of this algorithm as relevant for enhanced sampling of crystal structures was due to the obvious relation between molecular structure variables and corresponding crystal polymorphs as representative of the inherent vapor to crystal transitions that exist in nature. Additionally, it is shown that the trajectories of extended variables can be harnessed to extract fluctuation properties that can lead to valuable insights. A novel thermodynamic variable is introduced: the free energy difference between ensembles of Z' = 1 and Z' = 2 crystal polymorphs.

Coarse-Graining the Recognition of a Glycolipid by the C-Type Lectin Mincle Receptor.

Macrophage inducible Ca2+-dependent lectin (Mincle) receptor recognizes Mycobacterium tuberculosis glycolipids to trigger an immune response. This host membrane receptor is thus a key player in the modulation of the immune response to infection by M. tuberculosis and has emerged as a promising target for the development of new vaccines against tuberculosis. The recent development of the Martini 3 force field for coarse-grained (CG) molecular modeling allows the study of interactions of soluble proteins with small ligands which was not typically modeled well with the previous Martini 2 model. Here, we present a refined approach detailing a protocol for modeling interactions between a glycolipid and its receptor at a CG level using the Martini 3 force field. Using this approach, we studied Mincle and identified critical parameters governing ligand recognition, such as loop flexibility and the regulation of hydrophobic groove formation by calcium ions. In addition, we assessed ligand affinity using free energy perturbation calculations. Our results offer mechanistic insight into the interactions between Mincle and glycolipids, providing a basis for the rational design of molecules targeting this type of membrane receptors.

Alchemical Transformations and Beyond: Recent Advances and Real-World Applications of Free Energy Calculations in Drug Discovery.

Computational methods constitute efficient strategies for screening and optimizing potential drug molecules. A critical factor in this process is the binding affinity between candidate molecules and targets, quantified as binding free energy. Among various estimation methods, alchemical transformation methods stand out for their theoretical rigor. Despite challenges in force field accuracy and sampling efficiency, advancements in algorithms, software, and hardware have increased the application of free energy perturbation (FEP) calculations in the pharmaceutical industry. Here, we review the practical applications of FEP in drug discovery projects since 2018, covering both ligand-centric and residue-centric transformations. We show that relative binding free energy calculations have steadily achieved chemical accuracy in real-world applications. In addition, we discuss alternative physics-based simulation methods and the incorporation of deep learning into free energy calculations.

In silico study of some plant compounds as potential anticancer agents targeting MALT1 allosteric domain

Mucosa-associated lymphoid tissue lymphoma translocation protein 1 (MALT1) is the only human paracaspase, that serves as an adaptor protein and controls substantial genes expressed in the activation, proliferation of lymphocyte, and immune reactions by triggering the IKK/NF-kB signaling pathway. However, unusual MALT1-mediated NF-kB signaling pathway has been identified in multiple diseases like cancer, therefore making MALT1 a promising therapeutic target. There are scanty numbers of MALT1 inhibitors, thus the need to discover more compounds with less or no toxicity issue, that are cheap and pharmacologically efficient is of pertinence. Hence, our present study was to identify phyto-small molecules that could bind the allosteric interface of MALT1 using in silico methods. Total of 34 plant molecules were selected and screened for druglikeness, after which they were docked via Maestro 11.1 against the allosteric site of MALT1. The molecule with a binding score (kcal/mol) better than the control drug was subjected to molecular dynamics (MD) simulations of 100 ns via Desmond, free energy perturbations, principal component and Pearson correlation analyses. Our findings from this computational study presents cyanidin (–8.822 kcal/mol) as better binder to the allosteric site of MALT1 based on the molecular docking and pharmacokinetic profiling than thioridazine. Similarly, cyanidin-MALT1 complex showed significant stability and exhibiting contacts with critical amino acid residues in the site of interest than thioridazine-MALT1 complex. Hence, cyanidin is a potential allosteric inhibitor of MALT1. However, an urgent need for in vitro and in vivo validations is required to ascertain the efficacy of cyanidin in the fight against cancer and other MALT1-related diseases.

Benzylhydroxamic acids as inhibitors of insulin regulated aminopeptidase (IRAP)

With the objective of finding new classes of cognitive enhancers with potential for the treatment of neurodegenerative disorders, such as Alzheimer's disease, small molecule inhibitors of insulin-regulated aminopeptidase (IRAP) were designed and synthesized. IRAP is a member of the M1 family of zinc aminopeptidases and is abundantly expressed in areas of the brain associated with cognition, such as the amygdala, hippocampus and cerebral cortex. IRAP inhibitors were previously shown to enhance memory and learning in animal models. A comprehensive high throughput screening of 400,000 small molecules from the European Lead Factory library provided a series of 50 promising compounds in a qualified hit list (QHL). More than 30 IRAP inhibitors with an IC50 below 3.5 μM were identified. Herein, selected compounds from this QHL were assayed for solubility and permeability. Most of the selected compounds displayed good solubility, but further permeability studies on the best compounds revealed low blood brain barrier (BBB) permeability and high efflux in cells overexpressing P-gp pumps, rendering them less promising as starting points in drug discovery processes. Two compounds from the QHL were prioritized for further structural optimization; the pyridyl-substituted isoxazole 1a (QHL27) and the benzylhydroxamic acid derivative 1b (QHL1), both demonstrating fair BBB permeability and no indication of efflux. While our attempts to improve the isoxazole derivative 1a were not fruitful, a structural modification of 1b to provide the chloro-substituted benzylhydroxamic acid 14b resulted in a ten-fold improvement of the IRAP inhibition with an IC50 value of 60 nM. The binding modes of 1b and 14b were determined by free energy perturbation (FEP) analysis performed on candidate docking poses, determining a binding mode that accurately explained the experimental SAR. Further FEP studies of compound 14b suggested that it exhibits selectivity towards IRAP over Aminopeptidase N (APN), indicating its potential for targeted therapeutic applications.

Effects of Normal and Lateral Electric Fields on Membrane Mechanical Properties.

As a core component of biological and synthetic membranes, lipid bilayers are key to compartmentalizing chemical processes. Bilayer morphology and mechanical properties are heavily influenced by electric fields, such as those caused by biological ion concentration gradients. We present atomistic simulations exploring the effects of electric fields applied normally and laterally to lipid bilayers. We find that normal fields decrease membrane tension, while lateral fields increase it. Free energy perturbation calculations indicate the importance of dipole-dipole interactions to these tension changes, especially for lateral fields. We additionally show that membrane area compressibilities can be related to their cohesive energies, allowing us to estimate changes in membrane bending rigidity under applied fields. We find that normal and lateral fields decrease and increase bending rigidity, respectively. These results point to the use of directed electric fields to locally control membrane stiffness, thereby modulating associated cellular processes.

Identifying Artifacts from Large Library Docking.

While large library docking has discovered potent ligands for multiple targets, as the libraries have grown the hit lists can become dominated by rare artifacts that cheat our scoring functions. Here, we investigate rescoring top-ranked docked molecules with orthogonal methods to identify these artifacts, exploring implicit solvent models and absolute binding free energy perturbation as cross-filters. In retrospective studies, this approach deprioritized high-ranking nonbinders for nine targets while leaving true ligands relatively unaffected. We tested the method prospectively against hits from docking against AmpC β-lactamase. We prioritized 128 high-ranking molecules for synthesis and testing, a mixture of 39 molecules flagged as likely cheaters and 89 that were plausible inhibitors. None of the predicted cheating compounds inhibited AmpC detectably, while 57% of the 89 plausible compounds did so. As our libraries continue to grow, deprioritizing docking artifacts by rescoring with orthogonal methods may find wide use.

A Novel Rational PROTACs Design and Validation via AI-Driven Drug Design Approach.

The rational design of novel drug candidates presents a formidable challenge in modern drug discovery. Proteolysis-targeting chimeras (PROTACs) drug design is particularly demanding due to their limited crystal structure availability and design of a viable small molecule to bridge the protein of interest (POI) and ubiquitin-protein ligase (E3). An integrated approach that combines superimposition techniques and deep neural networks is demonstrated in this study to leverage the power of deep learning and structural biology to generate structurally diverse molecules with enhanced binding affinities. The superimposition technique ensures the congruence of initial and new protein-ligand pairs, which are evaluated via subsequent comprehensive screening using the root-mean-square deviation (RMSD), binding free energy (BFE), and buried solvent-accessible surface area (SASA). The final candidates are subjected to the incorporation of molecular dynamics (MD) and free energy perturbation (FEP) simulations to provide a quantitative evaluation of relative binding energies, reinforcing the efficacy and reliability of the generated molecules. The outcomes of the generated novel PROTACs molecules exhibit comparable structural attributes while demonstrating superior binding affinities within the binding pockets when contrasted with those of the established cocrystal ternary complexes. To enhance the generalizability of the workflow, we chose the ternary structure of the cellular inhibitor of apoptosis protein 1 (cIAP1) and Bruton's Tyrosine Kinase (BTK) for validating the chemical properties generated from the processes. The new linker molecules additionally showed superior affinity from the simulations. In summary, this methodology serves as an effective workflow to align computational predictions with current limitations, thereby introducing a novel paradigm in AI-driven drug design.

Relative genotoxicity of polycyclic aromatic hydrocarbons inferred from free energy perturbation approaches

Utilizing molecular dynamics and free energy perturbation, we examine the relative binding affinity of several covalent polycyclic aromatic hydrocarbon - DNA (PAH-DNA) adducts at the central adenine of NRAS codon-61, a mutational hotspot implicated in cancer risk. Several PAHs classified by the International Agency for Research on Cancer as probable, possible, or unclassifiable as to carcinogenicity are found to have greater binding affinity than the known carcinogen, benzo[a]pyrene (B[a]P). van der Waals interactions between the intercalated PAH and neighboring nucleobases, and minimal disruption of the DNA duplex drive increases in binding affinity. PAH-DNA adducts may be repaired by global genomic nucleotide excision repair (GG-NER), hence we also compute relative free energies of complexation of PAH-DNA adducts with RAD4-RAD23 (the yeast ortholog of human XPC-RAD23) which constitutes the recognition step in GG-NER. PAH-DNA adducts exhibiting the greatest DNA binding affinity also exhibit the least RAD4-RAD23 complexation affinity and are thus predicted to resist the GG-NER machinery, contributing to their genotoxic potential. In particular, the fjord region PAHs dibenzo[a,l]pyrene, benzo[g]chrysene, and benzo[c]phenanthrene are found to have greater binding affinity while having weaker RAD4-RAD23 complexation affinity than their respective bay region analogs B[a]P, chrysene, and phenanthrene. We also find that the bay region PAHs dibenzo[a,j]anthracene, dibenzo[a,c]anthracene, and dibenzo[a,h]anthracene exhibit greater binding affinity and weaker RAD4-RAD23 complexation affinity than B[a]P. Thus, the study of PAH genotoxicity likely needs to be substantially broadened, with implications for public policy and the health sciences. This approach can be broadly applied to assess factors contributing to the genotoxicity of other unclassified compounds.

Monte Carlo QM/MM simulation studies of the Cannizzaro reaction in ionic liquids for improved biofuel production.

The conversion of biomass to 5-hydroxymethylfurfural (HMF) holds substantial promise as a renewable energy source. Notably, HMF can be transformed into 2,5-bis(hydroxymethyl)furan (BHMF), a crucial reactant in biofuel production, but requires harsh operating conditions, H2, and precious metal catalysts. A recently reported Cannizzaro reaction of HMF to BHMF, characterized by its efficiency, mild conditions, and eco-friendliness, instead employed ionic liquids (ILs) to achieve high yields. In this study, combined quantum mechanical and molecular mechanical (QM/MM) simulations in conjunction with Metropolis Monte Carlo statistical mechanics and free-energy perturbation theory utilized M06-2X/6-31+G(d), PDDG/PM3, and the OPLS-VSIL force field to uncover important solute-solvent interactions present in the HMF to BHMF reaction pathway. The Cannizzaro reaction was examined in water and in five ILs composed of the 1-butyl-3-methylimidazolium [BMIM] cation coupled to hexafluorophosphate, tetrafluoroborate, thiocyanate, chloride, and bromide. Energetic and structural analysis of the rate-determining hydride transfer between HMF and the hydride-donor anion HMFOH- attributed the enhanced reactivity to highly organized solvent interactions featuring (1) hydrogen bonding between the ring protons of [BMIM] and the negatively charged carbonyl oxygen atoms on the transition structure, (2) favorable electrostatic interactions between the IL anions and solute hydroxyl groups, and (3) beneficial π-π stacking interactions between [BMIM] and the two aromatic rings present in HMF and HMFOH-.

A computational study of buckyballs as potential inhibitors of BACE1 in Alzheimer’s treatment

Alzheimer’s disease, marked by cognitive decline and memory loss, involves BACE1 in generating Aβ peptides, making it a key drug target. This study screened 129 buckyball ligands using molecular docking, identifying three promising compounds (4_C26_d3hc, 8_C20_Ih, and 16_C24_d6d) that fit within the BACE1 binding site. Molecular dynamics simulations highlighted thirteen BACE1 residues crucial for binding to buckyball ligands. The binding affinities of these ligands were also revealed via free energy perturbation (FEP), with 4_C26_d3hc demonstrating the greatest potential. This work lays the foundation for developing new BACE1 inhibitors using buckyball structures, offering a promising approach for Alzheimer’s therapy.

Machine Learning Guided AQFEP: A Fast and Efficient Absolute Free Energy Perturbation Solution for Virtual Screening.

Structure-based methods in drug discovery have become an integral part of the modern drug discovery process. The power of virtual screening lies in its ability to rapidly and cost-effectively explore enormous chemical spaces to select promising ligands for further experimental investigation. Relative free energy perturbation (RFEP) and similar methods are the gold standard for binding affinity prediction in drug discovery hit-to-lead and lead optimization phases, but have high computational cost and the requirement of a structural analog with a known activity. Without a reference molecule requirement, absolute FEP (AFEP) has, in theory, better accuracy for hit ID, but in practice, the slow throughput is not compatible with VS, where fast docking and unreliable scoring functions are still the standard. Here, we present an integrated workflow to virtually screen large and diverse chemical libraries efficiently, combining active learning with a physics-based scoring function based on a fast absolute free energy perturbation method. We validated the performance of the approach in the ranking of structurally related ligands, virtual screening hit rate enrichment, and active learning chemical space exploration; disclosing the largest reported collection of free energy simulations to date.

Model Mechanism for Lipid Uptake by the Human STARD2/PC-TP Phosphatidylcholine Transfer Protein.

The human StAR-related lipid transfer domain protein 2 (STARD2), also known as phosphatidylcholine (PC) transfer protein, is a single-domain lipid transfer protein thought to transfer PC lipids between intracellular membranes. We performed extensive μs-long molecular dynamics simulations of STARD2 of its apo and holo forms in the presence or absence of complex lipid bilayers. The simulations in water reveal ligand-dependent conformational changes. In the 2 μs-long simulations of apo STARD2 in the presence of a lipid bilayer, we observed spontaneous reproducible PC lipid uptake into the protein hydrophobic cavity. We propose that the lipid extraction mechanism involves one to two metastable states stabilized by choline-tyrosine or choline-tryptophane cation-π interactions. Using free energy perturbation, we evaluate that PC-tyrosine cation-π interactions contribute 1.8 and 2.5 kcal/mol to the affinity of a PC-STARD2 metastable state, thus potentially providing a significant decrease of the energy barrier required for lipid desorption.

Evolutionary sequence and structural basis for the distinct conformational landscapes of Tyr and Ser/Thr kinases

Protein kinases are molecular machines with rich sequence variation that distinguishes the two main evolutionary branches – tyrosine kinases (TKs) from serine/threonine kinases (STKs). Using a sequence co-variation Potts statistical energy model we previously concluded that TK catalytic domains are more likely than STKs to adopt an inactive conformation with the activation loop in an autoinhibitory folded conformation, due to intrinsic sequence effects. Here we investigate the structural basis for this phenomenon by integrating the sequence-based model with structure-based molecular dynamics (MD) to determine the effects of mutations on the free energy difference between active and inactive conformations, using a thermodynamic cycle involving many (n = 108) protein-mutation free energy perturbation (FEP) simulations in the active and inactive conformations. The sequence and structure-based results are consistent and support the hypothesis that the inactive conformation DFG-out Activation Loop Folded, is a functional regulatory state that has been stabilized in TKs relative to STKs over the course of their evolution via the accumulation of residue substitutions in the activation loop and catalytic loop that facilitate distinct substrate binding modes in trans and additional modes of regulation in cis for TKs.

Fluorine Domains Induced Ultrahigh Nitrogen Solubility in Ionic Liquids.

Fluorinated ionic liquids (ILs) are well-known as electrolytes in the nitrogen (N2) electroreduction reaction due to their exceptional gas solubility. However, the influence of fluorinated functional group on N2 solvation and solubility enhancement remains unclear. Massive molecular dynamics simulations and free energy perturbation methods are conducted to investigate the N2 solubility in 11 traditional and 9 fluorinated ILs. It shows that the fluorinated IL of 1-Ethyl-3-methylimidazolium tris(pentafluoroethyl) trifluorophosphate ([Emim]FAP) exhibits ultrahigh solubility, 4.844 × 10-3, approximately 118 times higher than that of traditional IL 1-Ethyl-3-methylimidazolium nitrate ([Emim]NO3). Moreover, fluorinated ILs with more than 10 C-F bonds possess higher N2 solubility than others and show an exothermic nature during solvation. As the C-F bonds number in ILs decreases, the N2 solubility decreases significantly and displays the opposite endothermic behavior. To understand the ultrahigh N2 solubility in fluorinated ILs, we propose a concept of fluorine densification energy (FDE), referring to the average strength of interaction between atoms per unit volume in ILs with fluorine domains, demonstrating a linear relationship with C-F bonds. Physically, lower FDE results in lower N2-anion pair dissociation energy and higher free volume, finally enhancing the N2 solubility. Consequently, medium to long alkyl fluorine tails within a polar environment defines a distinct fluorine domain, emphasizing FDE's role in enhancing N2 solubility. Overall, these quantitative results will not only deepen the understanding of N2 solvation in ILs but may also shed light on the rational design of IL-based high-performance N2 capture and conversion technologies.

Computational discovery of tripeptide inhibitors targeting monkeypox virus A42R profilin-like protein

Monkeypox is an infectious disease caused by the monkeypox virus (MPXV), a member of the Orthopoxvirus genus closely related to smallpox. The structure of the A42R profilin-like protein is the first and only available structure among MPXV proteins. Biochemical studies of A42R were conducted in the 1990s and later work also analyzed the protein's function in viral replication in cells. This study aims to screen tripeptides for their potential inhibition of the A42R profilin-like protein using computational methods, with implications for MPXV therapy. A total of 8000 tripeptides underwent molecular docking simulations, resulting in the identification of 20 compounds exhibiting strong binding affinity to A42R. To validate the docking results, molecular dynamics simulations and free energy perturbation calculations were performed. These analyses revealed two tripeptides with sequences TRP-THR-TRP and TRP-TRP-TRP, which displayed robust binding affinity to A42R. Markedly, electrostatic interactions predominated over van der Waals interactions in the binding process between tripeptides and A42R. Three A42R residues, namely Glu9, Ser12, and Arg38, appear to be pivotal in mediating the interaction between A42R and the tripeptide ligands. Notably, tripeptides containing two or three tryptophan residues demonstrate a pronounced binding affinity, with the tripeptide comprising three tryptophan amino acids showing the highest level of affinity. These findings offer valuable insights for the selection of compounds sharing a similar structure and possessing a high affinity for A42R, potentially capable of inhibiting its enzyme activity. The study highlights a structural advantage and paves the way for the development of targeted therapies against MPXV infections.

Identifying Artifacts from Large Library Docking.

While large library docking has discovered potent ligands for multiple targets, as the libraries have grown, the very top of the hit-lists can become populated with artifacts that cheat our scoring functions. Though these cheating molecules are rare, they become ever-more dominant with library growth. Here, we investigate rescoring top-ranked molecules from docking screens with orthogonal methods to identify these artifacts, exploring implicit solvent models and absolute binding free energy perturbation (AB-FEP) as cross-filters. In retrospective studies, this approach deprioritized high-ranking non-binders for nine targets while leaving true ligands relatively unaffected. We tested the method prospectively against results from large library docking AmpC β-lactamase. From the very top of the docking hit lists, we prioritized 128 molecules for synthesis and experimental testing, a mixture of 39 molecules that rescoring flagged as likely cheaters and another 89 that were plausible true actives. None of the 39 predicted cheating compounds inhibited AmpC up to in enzyme assays, while 57% of the 89 plausible true actives did do so, with 19 of them inhibiting the enzyme with apparent values better than . As our libraries continue to grow, a strategy of catching docking artifacts by rescoring with orthogonal methods may find wide use in the field.