Accurate prediction of dynamic protein-ligand binding using P-score ranking.

Protein-ligand binding prediction typically relies on docking methodologies and associated scoring functions to propose the binding mode of a ligand in a biological target. Significant challenges are associated with this approach, including the flexibility of the protein-ligand system, solvent-mediated interactions, and associated entropy changes. In addition, scoring functions are only weakly accurate due to the short time required for calculating enthalpic and entropic binding interactions. The workflow described here attempts to address these limitations by combining supervised molecular dynamics with dynamical averaging quantum mechanics fragment molecular orbital. This combination significantly increased the ability to predict the experimental binding structure of protein-ligand complexes independent from the starting position of the ligands or the binding site conformation. We found that the predictive power could be enhanced by combining the residence time and interaction energies as descriptors in a novel scoring function named the P-score. This is illustrated using six different protein-ligand targets as case studies.

How accurately can one predict drug binding modes using AlphaFold models?

Computational prediction of protein structure has been pursued intensely for decades, motivated largely by the goal of using structural models for drug discovery. Recently developed machine-learning methods such as AlphaFold 2 (AF2) have dramatically improved protein structure prediction, with reported accuracy approaching that of experimentally determined structures. To what extent do these advances translate to an ability to predict more accurately how drugs and drug candidates bind to their target proteins? Here, we carefully examine the utility of AF2 protein structure models for predicting binding poses of drug-like molecules at the largest class of drug targets, the G-protein-coupled receptors. We find that AF2 models capture binding pocket structures much more accurately than traditional homology models, with errors nearly as small as differences between structures of the same protein determined experimentally with different ligands bound. Strikingly, however, the accuracy of ligand-binding poses predicted by computational docking to AF2 models is not significantly higher than when docking to traditional homology models and is much lower than when docking to structures determined experimentally without these ligands bound. These results have important implications for all those who might use predicted protein structures for drug discovery.

Human Posture Estimation: A Systematic Review on Force-Based Methods-Analyzing the Differences in Required Expertise and Result Benefits for Their Utilization.

Force-based human posture estimation (FPE) provides a valuable alternative when camera-based human motion capturing is impractical. It offers new opportunities for sensor integration in smart products for patient monitoring, ergonomic optimization and sports science. Due to the interdisciplinary research on the topic, an overview of existing methods and the required expertise for their utilization is lacking. This paper presents a systematic review by the PRISMA 2020 review process. In total, 82 studies are selected (59 machine learning (ML)-based and 23 digital human model (DHM)-based posture estimation methods). The ML-based methods use input data from hardware sensors-mostly pressure mapping sensors-and trained ML models for estimating human posture. The ML-based human posture estimation algorithms mostly reach an accuracy above 90%. DHMs, which represent the structure and kinematics of the human body, adjust posture to minimize physical stress. The required expert knowledge for the utilization of these methods and their resulting benefits are analyzed and discussed. DHM-based methods have shown their general applicability without the need for application-specific training but require expertise in human physiology. ML-based methods can be used with less domain-specific expertise, but an application-specific training of these models is necessary.

Hand Grasp Pose Prediction Based on Motion Prior Field.

Shared control of bionic robot hands has recently attracted much research attention. However, few studies have performed predictive analysis for grasp pose, which is vital for the pre-shape planning of robotic wrists and hands. Aiming at shared control of dexterous hand grasp planning, this paper proposes a framework for grasp pose prediction based on the motion prior field. To map the hand-object pose to the final grasp pose, an object-centered motion prior field is established to learn the prediction model. The results of motion capture reconstruction show that, with the input of a 7-dimensional pose and cluster manifolds of dimension 100, the model performs best in terms of prediction accuracy (90.2%) and error distance (1.27 cm) in the sequence. The model makes correct predictions in the first 50% of the sequence during hand approach to the object. The outcomes of this study enable prediction of the grasp pose in advance as the hand approaches the object, which is very important for enabling the shared control of bionic and prosthetic hands.

Future Pose Prediction from 3D Human Skeleton Sequence with Surrounding Situation.

Human pose prediction is vital for robot applications such as human-robot interaction and autonomous control of robots. Recent prediction methods often use deep learning and are based on a 3D human skeleton sequence to predict future poses. Even if the starting motions of 3D human skeleton sequences are very similar, their future poses will have variety. It makes it difficult to predict future poses only from a given human skeleton sequence. Meanwhile, when carefully observing human motions, we can find that human motions are often affected by objects or other people around the target person. We consider that the presence of surrounding objects is an important clue for the prediction. This paper proposes a method for predicting the future skeleton sequence by incorporating the surrounding situation into the prediction model. The proposed method uses a feature of an image around the target person as the surrounding information. We confirmed the performance improvement of the proposed method through evaluations on publicly available datasets. As a result, the prediction accuracy was improved for object-related and human-related motions.

MDO: A Computational Protocol for Prediction of Flexible Enzyme-Ligand Binding Mode.

AIM: Developing a method for use in computer aided drug design Background: Predicting the structure of enzyme-ligand binding mode is essential for understanding the properties, functions, and mechanisms of the bio-complex, but is rather difficult due to the enormous sampling space involved. OBJECTIVE: Accurate prediction of enzyme-ligand binding mode conformation. METHOD: A new computational protocol, MDO, is proposed for finding the structure of ligand binding pose. MDO consists of sampling enzyme sidechain conformations via molecular dynamics simulation of enzyme-ligand system and clustering of the enzyme configurations, sampling ligand binding poses via molecular docking and clustering of the ligand conformations, and the optimal ligand binding pose prediction via geometry optimization and ranking by the ONIOM method. MDO is tested on 15 enzyme-ligand complexes with known accurate structures. RESULTS: The success rate of MDO predictions, with RMSD < 2 Å, is 67%, substantially higher than the 40% success rate of conventional methods. The MDO success rate can be increased to 83% if the ONIOM calculations are applied only for the starting poses with ligands inside the binding cavities. CONCLUSION: The MDO protocol provides high quality enzyme-ligand binding mode prediction with reasonable computational cost. The MDO protocol is recommended for use in the structure-based drug design.

Modeling receptor flexibility in the structure-based design of KRASG12C inhibitors.

KRAS has long been referred to as an 'undruggable' target due to its high affinity for its cognate ligands (GDP and GTP) and its lack of readily exploited allosteric binding pockets. Recent progress in the development of covalent inhibitors of KRASG12C has revealed that occupancy of an allosteric binding site located between the α3-helix and switch-II loop of KRASG12C-sometimes referred to as the 'switch-II pocket'-holds great potential in the design of direct inhibitors of KRASG12C. In studying diverse switch-II pocket binders during the development of sotorasib (AMG 510), the first FDA-approved inhibitor of KRASG12C, we found the dramatic conformational flexibility of the switch-II pocket posing significant challenges toward the structure-based design of inhibitors. Here, we present our computational approaches for dealing with receptor flexibility in the prediction of ligand binding pose and binding affinity. For binding pose prediction, we modified the covalent docking program CovDock to allow for protein conformational mobility. This new docking approach, termed as FlexCovDock, improves success rates from 55 to 89% for binding pose prediction on a dataset of 10 cross-docking cases and has been prospectively validated across diverse ligand chemotypes. For binding affinity prediction, we found standard free energy perturbation (FEP) methods could not adequately handle the significant conformational change of the switch-II loop. We developed a new computational strategy to accelerate conformational transitions through the use of targeted protein mutations. Using this methodology, the mean unsigned error (MUE) of binding affinity prediction were reduced from 1.44 to 0.89 kcal/mol on a set of 14 compounds. These approaches were of significant use in facilitating the structure-based design of KRASG12C inhibitors and are anticipated to be of further use in the design of covalent (and noncovalent) inhibitors of other conformationally labile protein targets.

LCP: Simple Representation of Docking Poses for Machine Learning: A Case Study on Xanthine Oxidase Inhibitors.

In this paper, we propose a simple descriptor called the ligand coordinate profile (LCP) for describing docking poses. The LCP descriptor is generated from the coordinates of the polar hydrogen and heavy atoms of the docked ligand. We hypothesize that the prediction of binding poses can be enhanced through the combination of machine learning methods with the LCP descriptor. Two docking programs were used to predict ligand docking against xanthine oxidase. Four machine learning methods-k-nearest neighbors, random forest, support vector machine, and LightGBM-were used to determine whether machine learning-based models could be used to accurately identify the correct binding poses. Regardless of the machine learning method employed, the LCP descriptor demonstrated improved performance compared to the existing descriptor. The results of the leave-one-pdb-out approach revealed that the influence of the pose descriptor was also significant, as demonstrated through cross-validation. When evaluated using top-N metrics, the machine learning models were generally more effective than the docking programs. In addition, the LCP-based models outperformed those based on the existing descriptor. The results obtained in this study suggest that our proposed binding pose descriptor is effective for improving the docking accuracy of xanthine oxidase inhibitors.

Evaluating Molecular Docking Software for Small Molecule Binding to G-Quadruplex DNA.

G-quadruplexes are four-stranded nucleic acid secondary structures of biological significance and have emerged as an attractive drug target. The G4 formed in the MYC promoter (MycG4) is one of the most studied small-molecule targets, and a model system for parallel structures that are prevalent in promoter DNA G4s and RNA G4s. Molecular docking has become an essential tool in structure-based drug discovery for protein targets, and is also increasingly applied to G4 DNA. However, DNA, and in particular G4, binding sites differ significantly from protein targets. Here we perform the first systematic evaluation of four commonly used docking programs (AutoDock Vina, DOCK 6, Glide, and RxDock) for G4 DNA-ligand binding pose prediction using four small molecules whose complex structures with the MycG4 have been experimentally determined in solution. The results indicate that there are considerable differences in the performance of the docking programs and that DOCK 6 with GB/SA rescoring performs better than the other programs. We found that docking accuracy is mainly limited by the scoring functions. The study shows that current docking programs should be used with caution to predict G4 DNA-small molecule binding modes.

Using diverse potentials and scoring functions for the development of improved machine-learned models for protein-ligand affinity and docking pose prediction.

The advent of computational drug discovery holds the promise of significantly reducing the effort of experimentalists, along with monetary cost. More generally, predicting the binding of small organic molecules to biological macromolecules has far-reaching implications for a range of problems, including metabolomics. However, problems such as predicting the bound structure of a protein-ligand complex along with its affinity have proven to be an enormous challenge. In recent years, machine learning-based methods have proven to be more accurate than older methods, many based on simple linear regression. Nonetheless, there remains room for improvement, as these methods are often trained on a small set of features, with a single functional form for any given physical effect, and often with little mention of the rationale behind choosing one functional form over another. Moreover, it is not entirely clear why one machine learning method is favored over another. In this work, we endeavor to undertake a comprehensive effort towards developing high-accuracy, machine-learned scoring functions, systematically investigating the effects of machine learning method and choice of features, and, when possible, providing insights into the relevant physics using methods that assess feature importance. Here, we show synergism among disparate features, yielding adjusted R2 with experimental binding affinities of up to 0.871 on an independent test set and enrichment for native bound structures of up to 0.913. When purely physical terms that model enthalpic and entropic effects are used in the training, we use feature importance assessments to probe the relevant physics and hopefully guide future investigators working on this and other computational chemistry problems.

NbX: Machine Learning-Guided Re-Ranking of Nanobody-Antigen Binding Poses.

Modeling the binding pose of an antibody is a prerequisite to structure-based affinity maturation and design. Without knowing a reliable binding pose, the subsequent structural simulation is largely futile. In this study, we have developed a method of machine learning-guided re-ranking of antigen binding poses of nanobodies, the single-domain antibody which has drawn much interest recently in antibody drug development. We performed a large-scale self-docking experiment of nanobody-antigen complexes. By training a decision tree classifier through mapping a feature set consisting of energy, contact and interface property descriptors to a measure of their docking quality of the refined poses, significant improvement in the median ranking of native-like nanobody poses by was achieved eightfold compared with ClusPro and an established deep 3D CNN classifier of native protein-protein interaction. We further interpreted our model by identifying features that showed relatively important contributions to the prediction performance. This study demonstrated a useful method in improving our current ability in pose prediction of nanobodies.

Deep Scoring Neural Network Replacing the Scoring Function Components to Improve the Performance of Structure-Based Molecular Docking.

Accurate prediction of protein-ligand interactions can greatly promote drug development. Recently, a number of deep-learning-based methods have been proposed to predict protein-ligand binding affinities. However, these methods independently extract the feature representations of proteins and ligands but ignore the relative spatial positions and interaction pairs between them. Here, we propose a virtual screening method based on deep learning, called Deep Scoring, which directly extracts the relative position information and atomic attribute information on proteins and ligands from the docking poses. Furthermore, we use two Resnets to extract the features of ligand atoms and protein residues, respectively, and generate an atom-residue interaction matrix to learn the underlying principles of the interactions between proteins and ligands. This is then followed by a dual attention network (DAN) to generate the attention for two related entities (i.e., proteins and ligands) and to weigh the contributions of each atom and residue to binding affinity prediction. As a result, Deep Scoring outperforms other structure-based deep learning methods in terms of screening performance (area under the receiver operating characteristic curve (AUC) of 0.901 for an unbiased DUD-E version), pose prediction (AUC of 0.935 for PDBbind test set), and generalization ability (AUC of 0.803 for the CHEMBL data set). Finally, Deep Scoring was used to select novel ERK2 inhibitor, and two compounds (D264-0698 and D483-1785) were obtained with potential inhibitory activity on ERK2 through the biological experiments.

Cutting Pose Prediction from Point Clouds.

The challenge of getting machines to understand and interact with natural objects is encountered in important areas such as medicine, agriculture, and, in our case, slaughterhouse automation. Recent breakthroughs have enabled the application of Deep Neural Networks (DNN) directly to point clouds, an efficient and natural representation of 3D objects. The potential of these methods has mostly been demonstrated for classification and segmentation tasks involving rigid man-made objects. We present a method, based on the successful PointNet architecture, for learning to regress correct tool placement from human demonstrations, using virtual reality. Our method is applied to a challenging slaughterhouse cutting task, which requires an understanding of the local geometry including the shape, size, and orientation. We propose an intermediate five-Degree of Freedom (DoF) cutting plane representation, a point and a normal vector, which eases the demonstration and learning process. A live experiment is conducted in order to unveil issues and begin to understand the required accuracy. Eleven cuts are rated by an expert, with 8 / 11 being rated as acceptable. The error on the test set is subsequently reduced through the addition of more training data and improvements to the DNN. The result is a reduction in the average translation from 1.5 cm to 0.8 cm and the orientation error from 4 . 59 to 4 . 48 . The method's generalization capacity is assessed on a similar task from the slaughterhouse and on the very different public LINEMOD dataset for object pose estimation across view points. In both cases, the method shows promising results. Code, datasets, and supplementary materials are available at https://github.com/markpp/PoseFromPointClouds.

D3R Grand Challenge 4: ligand similarity and MM-GBSA-based pose prediction and affinity ranking for BACE-1 inhibitors.

The Drug Design Data Resource (D3R) Grand Challenges present an opportunity to assess, in the context of a blind predictive challenge, the accuracy and the limits of tools and methodologies designed to help guide pharmaceutical drug discovery projects. Here, we report the results of our participation in the D3R Grand Challenge 4 (GC4), which focused on predicting the binding poses and affinity ranking for compounds targeting the [Formula: see text]-amyloid precursor protein (BACE-1). Our ligand similarity-based protocol using HYBRID (OpenEye Scientific Software) successfully identified poses close to the native binding mode for most of the ligands with less than 2 Å RMSD accuracy. Furthermore, we compared the performance of our HYBRID-based approach to that of AutoDock Vina and DOCK 6 and found that using a reference ligand to guide the docking process is a better strategy for pose prediction and helped HYBRID to perform better here. We also conducted end-point free energy estimates on molecules dynamics based ensembles of protein-ligand complexes using molecular mechanics combined with generalized Born surface area method (MM-GBSA). We found that the binding affinity ranking based on MM-GBSA scores have poor correlation with the experimental values. Finally, the main lessons from our participation in D3R GC4 are: (i) the generation of the macrocyclic conformers is a key step for successful pose prediction, (ii) the protonation states of the BACE-1 binding site should be treated carefully, (iii) the MM-GBSA method could not discriminate well between different predicted binding poses, and (iv) the MM-GBSA method does not perform well at predicting protein-ligand binding affinities here.

MathDL: mathematical deep learning for D3R Grand Challenge 4.

We present the performances of our mathematical deep learning (MathDL) models for D3R Grand Challenge 4 (GC4). This challenge involves pose prediction, affinity ranking, and free energy estimation for beta secretase 1 (BACE) as well as affinity ranking and free energy estimation for Cathepsin S (CatS). We have developed advanced mathematics, namely differential geometry, algebraic graph, and/or algebraic topology, to accurately and efficiently encode high dimensional physical/chemical interactions into scalable low-dimensional rotational and translational invariant representations. These representations are integrated with deep learning models, such as generative adversarial networks (GAN) and convolutional neural networks (CNN) for pose prediction and energy evaluation, respectively. Overall, our MathDL models achieved the top place in pose prediction for BACE ligands in Stage 1a. Moreover, our submissions obtained the highest Spearman correlation coefficient on the affinity ranking of 460 CatS compounds, and the smallest centered root mean square error on the free energy set of 39 CatS molecules. It is worthy to mention that our method on docking pose predictions has significantly improved from our previous ones.

Cross-docking benchmark for automated pose and ranking prediction of ligand binding.

Significant efforts have been devoted in the last decade to improving molecular docking techniques to predict both accurate binding poses and ranking affinities. Some shortcomings in the field are the limited number of standard methods for measuring docking success and the availability of widely accepted standard data sets for use as benchmarks in comparing different docking algorithms throughout the field. In order to address these issues, we have created a Cross-Docking Benchmark server. The server is a versatile cross-docking data set containing 4,399 protein-ligand complexes across 95 protein targets intended to serve as benchmark set and gold standard for state-of-the-art pose and ranking prediction in easy, medium, hard, or very hard docking targets. The benchmark along with a customizable cross-docking data set generation tool is available at http://disco.csb.pitt.edu. We further demonstrate the potential uses of the server in questions outside of basic benchmarking such as the selection of the ideal docking reference structure.

Assessing and improving the performance of consensus docking strategies using the DockBox package.

Molecular docking is a well-established computational technique that aims to predict how a ligand binds to a specific protein target, as well as to assess the strength of the binding. Although docking programs are used worldwide for drug discovery, it is not always simple to identify which program or combination of programs provides the best results for a target of interest. Here we present DockBox, a computational package designed to facilitate the use of multiple docking and scoring programs allowing to combine them using different consensus strategies. As part of the DockBox package, a new consensus docking method called score-based consensus docking (SBCD) is introduced. SBCD was found to significantly improve the pose prediction success rates of single docking programs. When applied to virtual screening, SBCD enhanced enrichment factors while producing higher hit rates than standard consensus docking (CD). SBCD can be run with almost no additional computational cost and time compared to CD, if the same docking programs are used for pose generation. Furthermore, SBCD allows the use of many scoring functions to assess consensus without significant overhead, making it a promising new approach for the screening of large chemical libraries. DockBox is an open-source package publicly available at https://pypi.org/project/dockbox .

Electrostatic-field and surface-shape similarity for virtual screening and pose prediction.

We introduce a new method for rapid computation of 3D molecular similarity that combines electrostatic field comparison with comparison of molecular surface-shape and directional hydrogen-bonding preferences (called "eSim"). Rather than employing heuristic "colors" or user-defined molecular feature types to represent conformation-dependent molecular electrostatics, eSim calculates the similarity of the electrostatic fields of two molecules (in addition to shape and hydrogen-bonding). We present detailed virtual screening performance data on the standard 102 target DUD-E set. In its moderately fast screening mode, eSim running on a single computing core is capable of processing over 60 molecules per second. In this mode, eSim performed significantly better than all alternate methods for which full DUD-E data were available (mean ROC area of 0.74, p [Formula: see text], by paired t-test, compared with the best performing alternate method). In addition, for 92 targets of the DUD-E set where multiple ligand-bound crystal structures were available, screening performance was assessed using alternate ligands or sets thereof (in their bound poses) as similarity targets. Using the joint alignment of five ligands for each protein target, mean ROC area exceeded 0.82 for the 92 targets. Design-focused application of ligand similarity methods depends on accurate predictions of geometric molecular relationships. We comprehensively assessed pose prediction accuracy by curating nearly 400,000 bound ligand pose pairs across the DUD-E targets. Overall, beginning from agnostic initial poses, we observed an 80% success rate for RMSD [Formula: see text] Å  among the top 20 predicted eSim poses. These examples were split roughly 50/50 into cases with high direct atomic overlap (where a shared scaffold exists between a pair) and low direct atomic overlap (where where a ligand pair has dissimilar scaffolds but largely occupies the same space). Within the high direct atomic overlap subset, the pose prediction success rate was 93%. For the more challenging subset (where dissimilar scaffolds are to be aligned), the success rate was 70%. The eSim approach enables both large-scale screening and rational design of ligands and is rooted in physically meaningful, non-heuristic, molecular comparisons.

Improving ligand 3D shape similarity-based pose prediction with a continuum solvent model.

In order to improve the pose prediction performance of docking methods, we have previously developed the pose prediction using shape similarity (PoPSS) method. It identifies a ligand conformation of the highest shape similarity with target protein crystal ligands. The identified ligand conformation is then placed into the target protein binding pocket and refined using side-chain repacking and Monte Carlo energy minimization. Subsequently, we have reported a modification to PoPSS, named as PoPSS-Lite, using a simple grid-based energy minimization for side-chain repacking and Tversky correlation coefficient as the similarity metric. This modification has improved the pose prediction performance and PoPSS-Lite was one of the top performers in D3R GC3. Here we report a further modification to PoPSS that utilizes a continuum solvent model to account for water mediated protein ligand interactions. In this approach, named as PoPSS-PB, the ligand conformation of the highest shape similarity with crystal ligands is refined along with the target protein binding site by incorporating the Poisson-Boltzmann electrostatics. The performance of PoPSS-PB along with PoPSS and PoPSS-Lite was prospectively evaluated in D3R GC4. PoPSS-PB not only demonstrated excellent performance with mean and median RMSDs of 1.20 and 1.13 Å but also achieved improved performance over PoPSS and PoPSS-Lite. Furthermore, the comparison with other D3R GC4 pose prediction submissions revealed admirable performance. Our results showed that the binding poses of ligands with unknown binding modes can be successfully predicted by utilizing ligand 3D shape similarity with known crystallographic ligands and that taking the solvation into consideration improves pose prediction.

Continuous Evaluation of Ligand Protein Predictions: A Weekly Community Challenge for Drug Docking.

Docking calculations can accelerate drug discovery by predicting the bound poses of ligands for a targeted protein. However, it is not clear which docking methods work best. Furthermore, predicting poses requires steps outside the docking algorithm itself, such as preparation of the protein and ligand, and it is not known which components are most in need of improvement. The Continuous Evaluation of Ligand Protein Predictions (CELPP) is a blinded prediction challenge designed to address these issues. Participants create a workflow to predict protein-ligand binding poses, which is then tasked with predicting 10-100 new protein-ligand crystal structures each week. CELPP evaluates the accuracy of each workflow's predictions and posts the scores online. The results can be used to identify the strengths and weaknesses of current approaches, help map docking problems to the algorithms most likely to overcome them, and illuminate areas of unmet need in structure-guided drug design.