PROTAC-DB 3.0: an updated database of PROTACs with extended pharmacokinetic parameters.

Proteolysis-targeting chimera (PROTAC) is an emerging therapeutic technology that leverages the ubiquitin-proteasome system to target protein degradation. Due to its event-driven mechanistic characteristics, PROTAC has the potential to regulate traditionally non-druggable targets. Recently, AI-aided drug design has accelerated the development of PROTAC drugs. However, the rational design of PROTACs remains a considerable challenge. Here, we present an updated online database, PROTAC-DB 3.0. In this third version, we have expanded the database to include 6111 PROTACs (87% increase compared to the 2.0 version). Additionally, the database now contains 569 warheads (small molecules targeting the protein), 2753 linkers, and 107 E3 ligands (small molecules recruiting E3 ligases). The number of target-PROTAC-E3 ternary complex structures has also increased to 959. Recognizing the importance of druggability in PROTAC design, we have incorporated pharmacokinetic data to PROTAC-DB 3.0. To enhance user experience, we have added features for sorting based on molecular similarity and literature publication date. PROTAC-DB 3.0 is accessible at http://cadd.zju.edu.cn/protacdb/.

PROTAC-DB 2.0: an updated database of PROTACs.

Proteolysis targeting chimeras (PROTACs), which harness the ubiquitin-proteasome system to selectively induce targeted protein degradation, represent an emerging therapeutic technology with the potential to modulate traditional undruggable targets. Over the past few years, this technology has moved from academia to industry and more than 10 PROTACs have been advanced into clinical trials. However, designing potent PROTACs with desirable drug-like properties still remains a great challenge. Here, we report an updated online database, PROTAC-DB 2.0, which is a repository of structural and experimental data about PROTACs. In this 2nd release, we expanded the number of PROTACs to 3270, which corresponds to a 96% expansion over the first version. Meanwhile, the numbers of warheads (small molecules targeting the proteins of interest), linkers, and E3 ligands (small molecules recruiting E3 ligases) have increased to over 360, 1500 and 80, respectively. In addition, given the importance and the limited number of the crystal target-PROTAC-E3 ternary complex structures, we provide the predicted ternary complex structures for PROTACs with good degradation capability using our PROTAC-Model method. To further facilitate the analysis of PROTAC data, a new filtering strategy based on the E3 ligases is also added. PROTAC-DB 2.0 is available online at http://cadd.zju.edu.cn/protacdb/.

Genetic Algorithm-Based Receptor Ligand: A Genetic Algorithm-Guided Generative Model to Boost the Novelty and Drug-Likeness of Molecules in a Sampling Chemical Space.

Deep learning-based de novo molecular design has recently gained significant attention. While numerous DL-based generative models have been successfully developed for designing novel compounds, the majority of the generated molecules lack sufficiently novel scaffolds or high drug-like profiles. The aforementioned issues may not be fully captured by commonly used metrics for the assessment of molecular generative models, such as novelty, diversity, and quantitative estimation of the drug-likeness score. To address these limitations, we proposed a genetic algorithm-guided generative model called GARel (genetic algorithm-based receptor-ligand interaction generator), a novel framework for training a DL-based generative model to produce drug-like molecules with novel scaffolds. To efficiently train the GARel model, we utilized dense net to update the parameters based on molecules with novel scaffolds and drug-like features. To demonstrate the capability of the GARel model, we used it to design inhibitors for three targets: AA2AR, EGFR, and SARS-Cov2. The results indicate that GARel-generated molecules feature more diverse and novel scaffolds and possess more desirable physicochemical properties and favorable docking scores. Compared with other generative models, GARel makes significant progress in balancing novelty and drug-likeness, providing a promising direction for the further development of DL-based de novo design methodology with potential impacts on drug discovery.

Accuracy or novelty: what can we gain from target-specific machine-learning-based scoring functions in virtual screening?

Machine-learning (ML)-based scoring functions (MLSFs) have gradually emerged as a promising alternative for protein-ligand binding affinity prediction and structure-based virtual screening. However, clouds of doubts have still been raised against the benefits of this novel type of scoring functions (SFs). In this study, to benchmark the performance of target-specific MLSFs on a relatively unbiased dataset, the MLSFs trained from three representative protein-ligand interaction representations were assessed on the LIT-PCBA dataset, and the classical Glide SP SF and three types of ligand-based quantitative structure-activity relationship (QSAR) models were also utilized for comparison. Two major aspects in virtual screening campaigns, including prediction accuracy and hit novelty, were systematically explored. The calculation results illustrate that the tested target-specific MLSFs yielded generally superior performance over the classical Glide SP SF, but they could hardly outperform the 2D fingerprint-based QSAR models. Although substantial improvements could be achieved by integrating multiple types of protein-ligand interaction features, the MLSFs were still not sufficient to exceed MACCS-based QSAR models. In terms of the correlations between the hit ranks or the structures of the top-ranked hits, the MLSFs developed by different featurization strategies would have the ability to identify quite different hits. Nevertheless, it seems that target-specific MLSFs do not have the intrinsic attributes of a traditional SF and may not be a substitute for classical SFs. In contrast, MLSFs can be regarded as a new derivative of ligand-based QSAR models. It is expected that our study may provide valuable guidance for the assessment and further development of target-specific MLSFs.

CovalentInDB: a comprehensive database facilitating the discovery of covalent inhibitors.

Inhibitors that form covalent bonds with their targets have traditionally been considered highly adventurous due to their potential off-target effects and toxicity concerns. However, with the clinical validation and approval of many covalent inhibitors during the past decade, design and discovery of novel covalent inhibitors have attracted increasing attention. A large amount of scattered experimental data for covalent inhibitors have been reported, but a resource by integrating the experimental information for covalent inhibitor discovery is still lacking. In this study, we presented Covalent Inhibitor Database (CovalentInDB), the largest online database that provides the structural information and experimental data for covalent inhibitors. CovalentInDB contains 4511 covalent inhibitors (including 68 approved drugs) with 57 different reactive warheads for 280 protein targets. The crystal structures of some of the proteins bound with a covalent inhibitor are provided to visualize the protein-ligand interactions around the binding site. Each covalent inhibitor is annotated with the structure, warhead, experimental bioactivity, physicochemical properties, etc. Moreover, CovalentInDB provides the covalent reaction mechanism and the corresponding experimental verification methods for each inhibitor towards its target. High-quality datasets are downloadable for users to evaluate and develop computational methods for covalent drug design. CovalentInDB is freely accessible at http://cadd.zju.edu.cn/cidb/.

PROTAC-DB: an online database of PROTACs.

Proteolysis-targeting chimeras (PROTACs), which selectively degrade targeted proteins by the ubiquitin-proteasome system, have emerged as a novel therapeutic technology with potential advantages over traditional inhibition strategies. In the past few years, this technology has achieved substantial progress and two PROTACs have been advanced into phase I clinical trials. However, this technology is still maturing and the design of PROTACs remains a great challenge. In order to promote the rational design of PROTACs, we present PROTAC-DB, a web-based open-access database that integrates structural information and experimental data of PROTACs. Currently, PROTAC-DB consists of 1662 PROTACs, 202 warheads (small molecules that target the proteins of interest), 65 E3 ligands (small molecules capable of recruiting E3 ligases) and 806 linkers, as well as their chemical structures, biological activities, and physicochemical properties. Except the biological activities of warheads and E3 ligands, PROTAC-DB also provides the degradation capacities, binding affinities and cellular activities for PROTACs. PROTAC-DB can be queried with two general searching approaches: text-based (target name, compound name or ID) and structure-based. In addition, for the convenience of users, a filtering tool for the searching results based on the physicochemical properties of compounds is also offered. PROTAC-DB is freely accessible at http://cadd.zju.edu.cn/protacdb/.

Structural view on the role of the TRD loop in regulating DNMT3A activity: a molecular dynamics study.

DNA methyltransferase 3A (DNMT3A) has been regarded as a potential epigenetic target for the development of cancer therapeutics. A number of DNMT3A inhibitors have been reported, but most of them do not have good potency, high selectivity and/or low cytotoxicity. It has been suggested that a non-conserved region around the target recognition domain (TRD) loop is implicated in the DNMT3A activity under the allosteric regulation of the ATRX-DNMT3-DNMT3L (ADD) domain, but the molecular mechanism of the regulation of the TRD loop on the DNMT3A activity needs to be elucidated. In this study, based on the reported crystal structures, the dynamics of the TRD loop in different multimerization with/without the bound guest molecule, namely the ADD domain or the DNA molecule, was investigated using conventional molecular dynamics (MD) and umbrella sampling simulations. The simulation results illustrate that the TRD loop exhibits relatively higher flexibility than the other components in the whole catalytic domain (CD), which could be well stabilized into different local minima through the binding with either the ADD domain or the DNA molecule by forming tight hydrogen-bond and salt-bridge networks involving distinct residues. Moreover, the movement of the TRD loop away from the catalytic loop upon activation could be triggered simply by the detachment of the ADD domain, but not necessarily induced by the ADD domain relocation on the CD. All these dynamic structural details could be a supplement to the previously reported crystal structure, which underlines the importance of the structural flexibility for the critical residues in the TRD loop, arousing more interest in the rational design of novel DNMT3A inhibitors targeting this region.

VAD-MM/GBSA: A Variable Atomic Dielectric MM/GBSA Model for Improved Accuracy in Protein-Ligand Binding Free Energy Calculations.

The molecular mechanics/generalized Born surface area (MM/GBSA) has been widely used in end-point binding free energy prediction in structure-based drug design (SBDD). However, in practice, it is usually being treated as a disputed method mostly because of its system dependence. Here, combining with machine-learning optimization, we developed a novel version of MM/GBSA, named variable atomic dielectric MM/GBSA (VAD-MM/GBSA), by assigning variable dielectric constants directly to the protein/ligand atoms. The new strategy exhibits markedly improved accuracy in binding affinity calculations for various protein-ligand systems and is promising to be used in the postprocessing of structure-based virtual screening. Moreover, VAD-MM/GBSA outperformed prime MM/GBSA in Schrödinger software and showed remarkable predictive performance for specific protein targets, such as POL polyprotein, human immunodeficiency virus type 1 (HIV-1) protease, etc. Our study showed that the VAD-MM/GBSA method with little extra computational overhead provides a potential replacement of the MM/GBSA in AMBER software. An online web server of VAD-MMGBSA has been developed and is now available at http://cadd.zju.edu.cn/vdgb.

HawkDock: a web server to predict and analyze the protein-protein complex based on computational docking and MM/GBSA.

Protein-protein interactions (PPIs) play an important role in the different functions of cells, but accurate prediction of the three-dimensional structures for PPIs is still a notoriously difficult task. In this study, HawkDock, a free and open accessed web server, was developed to predict and analyze the structures of PPIs. In the HawkDock server, the ATTRACT docking algorithm, the HawkRank scoring function developed in our group and the MM/GBSA free energy decomposition analysis were seamlessly integrated into a multi-functional platform. The structures of PPIs were predicted by combining the ATTRACT docking and the HawkRank re-scoring, and the key residues for PPIs were highlighted by the MM/GBSA free energy decomposition. The molecular visualization was supported by 3Dmol.js. For the structural modeling of PPIs, HawkDock could achieve a better performance than ZDOCK 3.0.2 in the benchmark testing. For the prediction of key residues, the important residues that play an essential role in PPIs could be identified in the top 10 residues for ∼81.4% predicted models and ∼95.4% crystal structures in the benchmark dataset. To sum up, the HawkDock server is a powerful tool to predict the binding structures and identify the key residues of PPIs. The HawkDock server is accessible free of charge at http://cadd.zju.edu.cn/hawkdock/.

Development and Evaluation of MM/GBSA Based on a Variable Dielectric GB Model for Predicting Protein-Ligand Binding Affinities.

In structure-based drug design (SBDD), the molecular mechanics generalized Born surface area (MM/GBSA) approach has been widely used in ranking the binding affinity of small molecule ligands. However, an accurate estimation of protein-ligand binding affinity still remains a challenge due to the intrinsic limitation of the standard generalized Born (GB) model used in MM/GBSA. In this study, we proposed and evaluated the MM/GBSA approach based on a variable dielectric generalized Born (VDGB) model using residue-type-based dielectric constants. In the VDGB model, different dielectric values were assigned for the three types of protein residues, and the magnitude of the dielectric constants for residue types follows this order: charged ≥ polar ≥ nonpolar. We found that MM/GBSA based on a VDGB model (MM/GBSAVDGB) with an optimal dielectric constant of 4.0 for the charged residues and 1.0 for the noncharged residues together with a net-charge-dependent dielectric value for ligands achieved better predictions as judged by Pearson's correlation coefficient than the standard MM/GBSA with a uniform solute dielectric constant of 4.0 for the training set of 130 protein-ligand complexes. The prediction on the test set with 165 protein-ligand complexes also validated the better performance of MM/GBSAVDGB. Moreover, this method exhibited potential in predicting the relative binding free energies for multiple ligands against the same target. Furthermore, we found that rational truncation of protein residues far from the binding site can significantly speed up the MM/GBSAVDGB calculations, while it almost does not influence the prediction accuracy. Therefore, it is feasible to implement the system-truncated MM/GBSAVDGB as a scoring function for SBDD.

Assessing the performance of MM/PBSA and MM/GBSA methods. 9. Prediction reliability of binding affinities and binding poses for protein-peptide complexes.

A significant number of protein-protein interactions (PPIs) are mediated through the interactions between proteins and peptide segments, and therefore determination of protein-peptide interactions (PpIs) is critical to gain an in-depth understanding of the PPI network and even design peptides or small molecules capable of modulating PPIs. Computational approaches, especially molecular docking, provide an efficient way to model PpIs, and a reliable scoring function that can recognize the correct binding conformations for protein-peptide complexes is one of the most important components in protein-peptide docking. The end-point binding free energy calculation methods, such as MM/GBSA and MM/PBSA, are theoretically more rigorous than most empirical and semi-empirical scoring functions designed for protein-peptide docking, but their performance in predicting binding affinities and binding poses for protein-peptide systems has not been systematically assessed. In this study, we first evaluated the capability of MM/GBSA and MM/PBSA with different solvation models, interior dielectric constants (εin) and force fields to predict the binding affinities for 53 protein-peptide complexes. For the 19 short peptides with 5-12 residues, MM/PBSA based on the minimized structures in explicit solvent with the ff99 force field and εin = 2 yields the best correlation between the predicted binding affinities and the experimental data (rp = 0.748), while for the 34 medium-size peptides with 20-25 residues, MM/GBSA based on 1 ns of molecular dynamics (MD) simulations in implicit solvent with the ff03 force field, the GBOBC1 model and a low interior dielectric constant (εin = 1) yields the best accuracy (rp = 0.735). Then, we assessed the rescoring capability of MM/PBSA and MM/GBSA to distinguish the correct binding conformations from the decoys for 112 protein-peptide systems. The results illustrate that MM/PBSA based on the minimized structures with the ff99 or ff14SB force field and MM/GBSA based on the minimized structures with the ff03 force field show excellent capability to recognize the near-native binding poses for the short and medium-size peptides, respectively, and they outperform the predictions given by two popular protein-peptide docking algorithms (pepATTRACT and HPEPDOCK). Therefore, MM/PBSA and MM/GBSA are powerful tools to predict the binding affinities and identify the correct binding poses for protein-peptide systems.

Assessing the performance of the MM/PBSA and MM/GBSA methods. 10. Impacts of enhanced sampling and variable dielectric model on protein-protein Interactions.

Enhanced sampling has been extensively used to capture the conformational transitions in protein folding, but it attracts much less attention in the studies of protein-protein recognition. In this study, we evaluated the impact of enhanced sampling methods and solute dielectric constants on the overall accuracy of the molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA) and molecular mechanics/generalized Born surface area (MM/GBSA) approaches for the protein-protein binding free energy calculations. Here, two widely used enhanced sampling methods, including aMD and GaMD, and conventional molecular dynamics (cMD) simulations with two AMBER force fields (ff03 and ff14SB) were used to sample the conformations for 21 protein-protein complexes. The MM/PBSA and MM/GBSA calculation results illustrate that the standard MM/GBSA based on the cMD simulations yields the best Pearson correlation (rp = -0.523) between the predicted binding affinities and the experimental data, which is much higher than that given by MM/PBSA (rp = -0.212). Two enhanced sampling methods (aMD and GaMD) are indeed more efficient for conformational sampling, but they did not improve the binding affinity predictions for protein-protein systems, suggesting that the aMD or GaMD sampling (at least in short timescale simulations) may not be a good choice for the MM/PBSA and MM/GBSA predictions of protein-protein complexes. The solute dielectric constant of 1.0 is recommended to MM/GBSA, but a higher solute dielectric constant is recommended to MM/PBSA, especially for the systems with higher polarity on the protein-protein binding interfaces. Then, a preliminary assessment of the MM/GBSA calculations based on a variable dielectric generalized Born (VDGB) model was conducted. The results highlight the potential power of VDGB in the free energy predictions for protein-protein systems, but more thorough studies should be done in the future.

RediscMol: Benchmarking Molecular Generation Models in Biological Properties.

Deep learning-based molecular generative models have garnered emerging attention for their capability to generate molecules with novel structures and desired physicochemical properties. However, the evaluation of these models, particularly in a biological context, remains insufficient. To address the limitations of existing metrics and emulate practical application scenarios, we construct the RediscMol benchmark that comprises active molecules extracted from 5 kinase and 3 GPCR data sets. A set of rediscovery- and similarity-related metrics are introduced to assess the performance of 8 representative generative models (CharRNN, VAE, Reinvent, AAE, ORGAN, RNNAttn, TransVAE, and GraphAF). Our findings based on the RediscMol benchmark differ from those of previous evaluations. CharRNN, VAE, and Reinvent exhibit a greater ability to reproduce known active molecules, while RNNAttn, TransVAE, and GraphAF struggle in this aspect despite their notable performance on commonly used distribution-learning metrics. Our evaluation framework may provide valuable guidance for advancing generative models in real-world drug design scenarios.

Rational design of supramolecular self-assembly sensor for living cell imaging of HDAC1 and its application in high-throughput screening.

Supramolecular chemistry offers new insights in bioimaging, but specific tracking of enzyme in living cells via supramolecular host-guest reporter pair remains challenging, largely due to the interference caused by the complex cellular environment on the binding between analytes and hosts. Here, by exploiting the principle of supramolecular tandem assay (STA) and the classic host-guest reporter pair (p-sulfonatocalix[4]arene (SC4A) and lucigenin (LCG)) and rationally designing artificial peptide library to screen sequence with high affinity of the target enzyme, we developed a "turn-on" fluorescent sensing system for intracellular imaging of histone deacetylase 1 (HDAC1), which is a potential therapeutic target for various diseases, including cancer, neurological, and cardiovascular diseases. Based on computational simulations and experimental validations, we verified that the deacetylated peptide by HDAC1 competed LCG, freeing it from the SC4A causing fluorescence increase. Enzyme kinetics experiments were further conducted to prove that this assay could detect HDAC1 specifically with high sensitivity (the LOD value is 0.015 µg/mL, ten times lower than the published method). This system was further applied for high-throughput screening of HDAC1 inhibitors over a natural compound library containing 147 compounds, resulting in the identification of a novel HDAC1 down-regulator (Ginsenoside RK3). Our results demonstrated the sensitivity and robustness of the assay system towards HDAC1. It should serve as a valuable tool for biochemical studies and drug screening.

Learning on topological surface and geometric structure for 3D molecular generation.

Highly effective de novo design is a grand challenge of computer-aided drug discovery. Practical structure-specific three-dimensional molecule generations have started to emerge in recent years, but most approaches treat the target structure as a conditional input to bias the molecule generation and do not fully learn the detailed atomic interactions that govern the molecular conformation and stability of the binding complexes. The omission of these fine details leads to many models having difficulty in outputting reasonable molecules for a variety of therapeutic targets. Here, to address this challenge, we formulate a model, called SurfGen, that designs molecules in a fashion closely resembling the figurative key-and-lock principle. SurfGen comprises two equivariant neural networks, Geodesic-GNN and Geoatom-GNN, which capture the topological interactions on the pocket surface and the spatial interaction between ligand atoms and surface nodes, respectively. SurfGen outperforms other methods in a number of benchmarks, and its high sensitivity on the pocket structures enables an effective generative-model-based solution to the thorny issue of mutation-induced drug resistance.

RELATION: A Deep Generative Model for Structure-Based De Novo Drug Design.

Deep learning (DL)-based de novo molecular design has recently gained considerable traction. Many DL-based generative models have been successfully developed to design novel molecules, but most of them are ligand-centric and the role of the 3D geometries of target binding pockets in molecular generation has not been well-exploited. Here, we proposed a new 3D-based generative model called RELATION. In the RELATION model, the BiTL algorithm was specifically designed to extract and transfer the desired geometric features of the protein-ligand complexes to a latent space for generation. The pharmacophore conditioning and docking-based Bayesian sampling were applied to efficiently navigate the vast chemical space for the design of molecules with desired geometric properties and pharmacophore features. As a proof of concept, the RELATION model was used to design inhibitors for two targets, AKT1 and CDK2. The calculation results demonstrated that the RELATION model could efficiently generate novel molecules with favorable binding affinity and pharmacophore features.

ReMODE: a deep learning-based web server for target-specific drug design.

Deep learning (DL) and machine learning contribute significantly to basic biology research and drug discovery in the past few decades. Recent advances in DL-based generative models have led to superior developments in de novo drug design. However, data availability, deep data processing, and the lack of user-friendly DL tools and interfaces make it difficult to apply these DL techniques to drug design. We hereby present ReMODE (Receptor-based MOlecular DEsign), a new web server based on DL algorithm for target-specific ligand design, which integrates different functional modules to enable users to develop customizable drug design tasks. As designed, the ReMODE sever can construct the target-specific tasks toward the protein targets selected by users. Meanwhile, the server also provides some extensions: users can optimize the drug-likeness or synthetic accessibility of the generated molecules, and control other physicochemical properties; users can also choose a sub-structure/scaffold as a starting point for fragment-based drug design. The ReMODE server also enables users to optimize the pharmacophore matching and docking conformations of the generated molecules. We believe that the ReMODE server will benefit researchers for drug discovery. ReMODE is publicly available at http://cadd.zju.edu.cn/relation/remode/ .

Deep learning approaches for de novo drug design: An overview.

De novo drug design is the process of generating novel lead compounds with desirable pharmacological and physiochemical properties. The application of deep learning (DL) in de novo drug design has become a hot topic, and many DL-based approaches have been developed for molecular generation tasks. Generally, these approaches were developed as per four frameworks: recurrent neural networks; encoder-decoder; reinforcement learning; and generative adversarial networks. In this review, we first introduced the molecular representation and assessment metrics used in DL-based de novo drug design. Then, we summarized the features of each architecture. Finally, the potential challenges and future directions of DL-based molecular generation were prospected.

Integrative Modeling of PROTAC-Mediated Ternary Complexes.

Proteolysis-targeting chimeras (PROTACs), which selectively induce targeted protein degradation, represent an emerging drug discovery technology. Although numerous PROTACs have been reported, designing potent PROTACs still remains a great challenge, to some extent, due to insufficient structural data of Target-PROTAC-E3 ternary complexes. In this work, PROTAC-Model, an integrative computational method by combining the FRODOCK-based protocol and RosettaDock-based refinement, was developed to predict PROTAC-mediated ternary complex structures and tested on 14 cases. The quality of the models was evaluated using the criteria of the critical assessment of predicted interactions (CAPRI). Using the unbound structures, the FRODOCK-based protocol can generate the ternary complex structures with medium or high quality for 8 cases out of 14. With the refinement by RosettaDock, the cases with medium or high quality increase to 12. Compared with PRosettaC and the method developed by Drummond et al., PROTAC-Model shows better performance. In summary, PROTAC-Model should be useful for the rational design of PROTACs.

ASFP (Artificial Intelligence based Scoring Function Platform): a web server for the development of customized scoring functions.

Virtual screening (VS) based on molecular docking has emerged as one of the mainstream technologies of drug discovery due to its low cost and high efficiency. However, the scoring functions (SFs) implemented in most docking programs are not always accurate enough and how to improve their prediction accuracy is still a big challenge. Here, we propose an integrated platform called ASFP, a web server for the development of customized SFs for structure-based VS. There are three main modules in ASFP: (1) the descriptor generation module that can generate up to 3437 descriptors for the modelling of protein-ligand interactions; (2) the AI-based SF construction module that can establish target-specific SFs based on the pre-generated descriptors through three machine learning (ML) techniques; (3) the online prediction module that provides some well-constructed target-specific SFs for VS and an additional generic SF for binding affinity prediction. Our methodology has been validated on several benchmark datasets. The target-specific SFs can achieve an average ROC AUC of 0.973 towards 32 targets and the generic SF can achieve the Pearson correlation coefficient of 0.81 on the PDBbind version 2016 core set. To sum up, the ASFP server is a powerful tool for structure-based VS.