Directory Of Useful Decoys , Enhanced Research Articles

AutoDock-SS: AutoDock for Multiconformational Ligand-Based Virtual Screening.

Ligand-based virtual screening (LBVS) can be pivotal for identifying potential drug leads, especially when the target protein's structure is unknown. However, current LBVS methods are limited in their ability to consider the ligand conformational flexibility. This study presents AutoDock-SS (Similarity Searching), which adapts protein-ligand docking for use in LBVS. AutoDock-SS integrates novel ligand-based grid maps and AutoDock-GPU into a novel three-dimensional LBVS workflow. Unlike other approaches based on pregenerated conformer libraries, AutoDock-SS's built-in conformational search optimizes conformations dynamically based on the reference ligand, thus providing a more accurate representation of relevant ligand conformations. AutoDock-SS supports two modes: single and multiple ligand queries, allowing for the seamless consideration of multiple reference ligands. When tested on the Directory of Useful Decoys─Enhanced (DUD-E) data set, AutoDock-SS surpassed alternative 3D LBVS methods, achieving a mean AUROC of 0.775 and an EF1% of 25.72 in single-reference mode. The multireference mode, evaluated on the augmented DUD-E+ data set, demonstrated superior accuracy with a mean AUROC of 0.843 and an EF1% of 34.59. This enhanced performance underscores AutoDock-SS's ability to treat compounds as conformationally flexible while considering the ligand's shape, pharmacophore, and electrostatic potential, expanding the potential of LBVS methods.

Systematic Investigation of Docking Failures in Large-Scale Structure-Based Virtual Screening.

In recent years, large-scale structure-based virtual screening has attracted increasing levels of interest for identification of novel compounds corresponding to potential drug targets. It is critical to understand the strengths and weaknesses of docking algorithms to increase the success rate in practical applications. Here, we systematically investigated the docking successes and failures of two representative docking programs: UCSF DOCK 3.7 and AutoDock Vina. DOCK 3.7 performed better in early enrichment on the Directory of Useful Decoys: Enhanced (DUD-E) data set, although both docking methods were roughly comparable in overall enrichment performance. DOCK 3.7 also showed superior computational efficiency. Intriguingly, the Vina scoring function showed a bias toward compounds with higher molecular weights. Both the tested docking approaches yielded incorrectly predicted ligand binding poses caused by the limitations of torsion sampling. Based on a careful analysis of docking results from six representative cases, we propose the reasons underlying docking failures; furthermore, we provide a few solutions, representing practical guidance for large-scale virtual screening campaigns and future docking algorithm development.

CHARMM‐GUIhigh‐throughput simulator for efficient evaluation of protein–ligand interactions with different force fields

AbstractMolecular docking is one of the most popular computational tools for the hit discovery step in drug design. However, there is ample room for improvement of docking's ability to identify correct binding modes and discriminate active from decoy compounds. Molecular dynamics (MD) simulations of protein–ligand docking structures have been shown to be effective in improving docking results. Here, we present CHARMM‐GUI high‐throughput simulator (HTS) that prepares MD simulation systems and inputs for multiple protein–ligand complex structures in a high‐throughput manner. HTS supports commonly used MD programs (NAMD, GROMACS, AMBER, OpenMM, GENESIS, Desmond, LAMMPS, and Tinker) along with various force field combinations for protein and ligand, including CHARMM36m, Amber (ff19SB/ff14SB), OPLS‐AA/M, CGenFF, GAFF2, and OpenFF. Validation tests using Miller and the directory of useful decoys‐enhanced (DUD‐E) datasets demonstrate that short MD simulations using HTS‐generated systems and simple ligand RMSD calculations consistently outperform docking results. Specifically, MD simulations can better identify correct ligand‐binding modes among top 10 binding poses as compared to docking scores. In addition, MD simulations can better discriminate active from decoy compounds in the DUD‐E dataset than docking scores for both soluble and membrane proteins. We expect that HTS can be a useful tool to facilitate the hit discovery process in drug design by improving docking results.

CHARMM-GUI Ligand Designer for Template-Based Virtual Ligand Design in a Binding Site.

Rational drug design involves a task of finding ligands that would bind to a specific target protein. This work presents CHARMM-GUI Ligand Designer that is an intuitive and interactive web-based tool to design virtual ligands that match the shape and chemical features of a given protein binding site. Ligand Designer provides ligand modification capabilities with 3D visualization that allow researchers to modify and redesign virtual ligands while viewing how the protein-ligand interactions are affected. Virtual ligands can also be parameterized for further molecular dynamics (MD) simulations and free energy calculations. Using 8 targets from 8 different protein classes in the directory of useful decoys, enhanced (DUD-E) data set, we show that Ligand Designer can produce similar ligands to the known active ligands in the crystal structures. Ligand Designer also produces stable protein-ligand complex structures when tested using short MD simulations. We expect that Ligand Designer can be a useful and user-friendly tool to design small molecules in any given potential ligand binding site on a protein of interest.

Improving structure-based virtual screening performance via learning from scoring function components.

Scoring functions (SFs) based on complex machine learning (ML) algorithms have gradually emerged as a promising alternative to overcome the weaknesses of classical SFs. However, extensive efforts have been devoted to the development of SFs based on new protein-ligand interaction representations and advanced alternative ML algorithms instead of the energy components obtained by the decomposition of existing SFs. Here, we propose a new method named energy auxiliary terms learning (EATL), in which the scoring components are extracted and used as the input for the development of three levels of ML SFs including EATL SFs, docking-EATL SFs and comprehensive SFs with ascending VS performance. The EATL approach not only outperforms classical SFs for the absolute performance (ROC) and initial enrichment (BEDROC) but also yields comparable performance compared with other advanced ML-based methods on the diverse subset of Directory of Useful Decoys: Enhanced (DUD-E). The test on the relatively unbiased actives as decoys (AD) datasetalso proved the effectiveness of EATL. Furthermore, the idea of learning from SF components to yield improved screening power can also be extended to other docking programs and SFs available.

Predicting drug–protein interaction using quasi-visual question answering system

Identifying novel drug–protein interactions is crucial for drug discovery. For this purpose, many machine learning-based methods have been developed based on drug descriptors and one-dimensional protein sequences. However, protein sequences cannot accurately reflect the interactions in three-dimensional space. However, direct input of three-dimensional structure is of low efficiency due to the sparse three-dimensional matrix, and is also prevented by the limited number of co-crystal structures available for training. Here we propose an end-to-end deep learning framework to predict the interactions by representing proteins with a two-dimensional distance map from monomer structures (Image) and drugs with molecular linear notation (String), following the visual question answering mode. For efficient training of the system, we introduce a dynamic attentive convolutional neural network to learn fixed-size representations from the variable-length distance maps and a self-attentional sequential model to automatically extract semantic features from the linear notations. Extensive experiments demonstrate that our model obtains competitive performance against state-of-the-art baselines on the directory of useful decoys, enhanced (DUD-E), human and BindingDB benchmark datasets. Further attention visualization provides biological interpretation to depict highlighted regions of both protein and drug molecules. When predicting the interaction of proteins with potential drugs, the protein can be encoded as its one-dimensional sequence or a three-dimensional structure, which can capture more relevant features of the protein, but also makes the task to predict the interactions harder. A new method predicts these interactions using a two-dimensional distance matrix representation of a protein, which can be processed like a two-dimensional image, striking a balance between the data being simple to process and rich in relevant structures.

Hidden bias in the DUD-E dataset leads to misleading performance of deep learning in structure-based virtual screening.

Recently much effort has been invested in using convolutional neural network (CNN) models trained on 3D structural images of protein-ligand complexes to distinguish binding from non-binding ligands for virtual screening. However, the dearth of reliable protein-ligand x-ray structures and binding affinity data has required the use of constructed datasets for the training and evaluation of CNN molecular recognition models. Here, we outline various sources of bias in one such widely-used dataset, the Directory of Useful Decoys: Enhanced (DUD-E). We have constructed and performed tests to investigate whether CNN models developed using DUD-E are properly learning the underlying physics of molecular recognition, as intended, or are instead learning biases inherent in the dataset itself. We find that superior enrichment efficiency in CNN models can be attributed to the analogue and decoy bias hidden in the DUD-E dataset rather than successful generalization of the pattern of protein-ligand interactions. Comparing additional deep learning models trained on PDBbind datasets, we found that their enrichment performances using DUD-E are not superior to the performance of the docking program AutoDock Vina. Together, these results suggest that biases that could be present in constructed datasets should be thoroughly evaluated before applying them to machine learning based methodology development.

Ligity: A Non-Superpositional, Knowledge-Based Approach to Virtual Screening.

We present Ligity, a hybrid ligand-structure-based, non-superpositional method for virtual screening of large databases of small molecules. Ligity uses the relative spatial distribution of pharmacophoric interaction points (PIPs) derived from the conformations of small molecules. These are compared with the PIPs derived from key interaction features found in protein–ligand complexes and are used to prioritize likely binders. We investigated the effect of generating PIPs using the single lowest energy conformer versus an ensemble of conformers for each screened ligand, using different bin sizes for the distance between two features, utilizing triangular sets of pharmacophoric features (3-PIPs) versus chiral tetrahedral sets (4-PIPs), fusing data for targets with multiple protein–ligand complex structures, and applying different similarity measures. Ligity was benchmarked using the Directory of Useful Decoys-Enhanced (DUD-E). Optimal results were obtained using the tetrahedral PIPs derived from an ensemble of bound ligand conformers and a bin size of 1.5 Å, which are used as the default settings for Ligity. The high-throughput screening mode of Ligity, using only the lowest-energy conformer of each ligand, was used for benchmarking against the whole of the DUD-E, and a more resource-intensive, “information-rich” mode of Ligity, using a conformational ensemble of each ligand, were used for a representative subset of 10 targets. Against the full DUD-E database, mean area under the receiver operating characteristic curve (AUC) values ranged from 0.44 to 0.99, while for the representative subset they ranged from 0.61 to 0.86. Data fusion further improved Ligity’s performance, with mean AUC values ranging from 0.64 to 0.95. Ligity is very efficient compared to a protein–ligand docking method such as AutoDock Vina: if the time taken for the precalculation of Ligity descriptors is included in the comparason, then Ligity is about 20 times faster than docking. A direct comparison of the virtual screening steps shows Ligity to be over 5000 times faster. Ligity highly ranks the lowest-energy conformers of DUD-E actives, in a statistically significant manner, behavior that is not observed for DUD-E decoys. Thus, our results suggest that active compounds tend to bind in relatively low-energy conformations compared to decoys. This may be because actives—and thus their lowest-energy conformations—have been optimized for conformational complementarity with their cognate binding sites.

Improved Method of Structure-Based Virtual Screening via Interaction-Energy-Based Learning.

Virtual screening is a promising method for obtaining novel hit compounds in drug discovery. It aims to enrich potentially active compounds from a large chemical library for further biological experiments. However, the accuracy of current virtual screening methods is insufficient. In this study, we develop a new virtual screening method named Similarity of Interaction Energy VEctor Score (SIEVE-Score), in which protein-ligand interaction energies are extracted to represent docking poses for machine learning. SIEVE-Score offers substantial improvements compared to other state-of-the-art virtual screening methods, namely, other machine-learning-based scoring functions, interaction fingerprints, and docking software, for the enrichment factor 1% results on the Directory of Useful Decoys, Enhanced (DUD-E). The screening results are also human-interpretable in the form of important interactions for distinguishing between active and inactive compounds. The source code is available at https://github.com/sekijima-lab/SIEVE-Score .

HybridSim-VS: a web server for large-scale ligand-based virtual screening using hybrid similarity recognition techniques.

Molecular-similarity searches based on two-dimensional (2D) fingerprint and three-dimensional (3D) shape represent two widely used ligand-based virtual screening (VS) methods in computer-aided drug design. 2D fingerprint-based VS utilizes the binary fragment information on a known ligand, whereas 3D shape-based VS takes advantage of geometric information for predefined features from a 3D conformation. Given their different advantages, it would be desirable to hybridize 2D fingerprint and 3D shape molecular-similarity approaches in drug discovery. Here, we presented a general hybrid molecular-similarity protocol, referred to as HybridSim, obtained by combining the 2D fingerprint- and 3D shape-based similarity search methods and evaluated its performance on 595,036 actives and decoys for 40 pharmaceutically relevant targets available in the Directory of Useful Decoys Enhanced (DUD-E). Our results showed that HybridSim significantly improved the overall performance in 40 VS projects as compared with using only 2D fingerprint and 3D shape methods. Furthermore, HybridSim-VS, the first online platform using the proposed HybridSim method coupled with 17,839,945 screenable and purchasable compounds, was developed to provide large-scale and proficient VS capabilities to experts and nonexperts in the field. HybridSim-VS web server is freely available at http://www.rcidm.org/HybridSim-VS/. lingwang@scut.edu.cn. Supplementary data are available at Bioinformatics online.

Improving Screening Efficiency through Iterative Screening Using Docking and Conformal Prediction.

High-throughput screening, where thousands of molecules rapidly can be assessed for activity against a protein, has been the dominating approach in drug discovery for many years. However, these methods are costly and require much time and effort. In order to suggest an improvement to this situation, in this study, we apply an iterative screening process, where an initial set of compounds are selected for screening based on molecular docking. The outcome of the initial screen is then used to classify the remaining compounds through a conformal predictor. The approach was retrospectively validated using 41 targets from the Directory of Useful Decoys, Enhanced (DUD-E), ensuring scaffold diversity among the active compounds. The results show that 57% of the remaining active compounds could be identified while only screening 9.4% of the database. The overall hit rate (7.6%) was also higher than when using docking alone (5.2%). When limiting the search to the top scored compounds from docking, 39.6% of the active compounds could be identified, compared to 13.5% when screening the same number of compounds solely based on docking. The use of conformal predictors also gives a clear indication of the number of compounds to screen in the next iteration. These results indicate that iterative screening based on molecular docking and conformal prediction can be an efficient way to find active compounds while screening only a small part of the compound collection.

AutoDock-GIST: Incorporating Thermodynamics of Active-Site Water into Scoring Function for Accurate Protein-Ligand Docking.

Water plays a significant role in the binding process between protein and ligand. However, the thermodynamics of water molecules are often underestimated, or even ignored, in protein-ligand docking. Usually, the free energies of active-site water molecules are substantially different from those of waters in the bulk region. The binding of a ligand to a protein causes a displacement of these waters from an active site to bulk, and this displacement process substantially contributes to the free energy change of protein-ligand binding. The free energy of active-site water molecules can be calculated by grid inhomogeneous solvation theory (GIST), using molecular dynamics (MD) and the trajectory of a target protein and water molecules. Here, we show a case study of the combination of GIST and a docking program and discuss the effectiveness of the displacing gain of unfavorable water in protein-ligand docking. We combined the GIST-based desolvation function with the scoring function of AutoDock4, which is called AutoDock-GIST. The proposed scoring function was assessed employing 51 ligands of coagulation factor Xa (FXa), and results showed that both scoring accuracy and docking success rate were improved. We also evaluated virtual screening performance of AutoDock-GIST using FXa ligands in the directory of useful decoys-enhanced (DUD-E), thus finding that the displacing gain of unfavorable water is effective for a successful docking campaign.

Ligand Pose and Orientational Sampling in Molecular Docking

Molecular docking remains an important tool for structure-based screening to find new ligands and chemical probes. As docking ambitions grow to include new scoring function terms, and to address ever more targets, the reliability and extendability of the orientation sampling, and the throughput of the method, become pressing. Here we explore sampling techniques that eliminate stochastic behavior in DOCK3.6, allowing us to optimize the method for regularly variable sampling of orientations. This also enabled a focused effort to optimize the code for efficiency, with a three-fold increase in the speed of the program. This, in turn, facilitated extensive testing of the method on the 102 targets, 22,805 ligands and 1,411,214 decoys of the Directory of Useful Decoys - Enhanced (DUD-E) benchmarking set, at multiple levels of sampling. Encouragingly, we observe that as sampling increases from 50 to 500 to 2000 to 5000 to 20000 molecular orientations in the binding site (and so from about 1×1010 to 4×1010 to 1×1011 to 2×1011 to 5×1011 mean atoms scored per target, since multiple conformations are sampled per orientation), the enrichment of ligands over decoys monotonically increases for most DUD-E targets. Meanwhile, including internal electrostatics in the evaluation ligand conformational energies, and restricting aromatic hydroxyls to low energy rotamers, further improved enrichment values. Several of the strategies used here to improve the efficiency of the code are broadly applicable in the field.

Directory of Useful Decoys, Enhanced (DUD-E): Better Ligands and Decoys for Better Benchmarking

A key metric to assess molecular docking remains ligand enrichment against challenging decoys. Whereas the directory of useful decoys (DUD) has been widely used, clear areas for optimization have emerged. Here we describe an improved benchmarking set that includes more diverse targets such as GPCRs and ion channels, totaling 102 proteins with 22886 clustered ligands drawn from ChEMBL, each with 50 property-matched decoys drawn from ZINC. To ensure chemotype diversity, we cluster each target’s ligands by their Bemis–Murcko atomic frameworks. We add net charge to the matched physicochemical properties and include only the most dissimilar decoys, by topology, from the ligands. An online automated tool (http://decoys.docking.org) generates these improved matched decoys for user-supplied ligands. We test this data set by docking all 102 targets, using the results to improve the balance between ligand desolvation and electrostatics in DOCK 3.6. The complete DUD-E benchmarking set is freely available at http://dude.docking.org.