Combining in silico and in cerebro approaches for virtual screening and pose prediction in SAMPL4

The SAMPL challenges provide an ideal opportunity for unbiased evaluation and comparison of different approaches used in computational drug design. During the fourth round of this SAMPL challenge, we participated in the virtual screening and binding pose prediction on inhibitors targeting the HIV-1 integrase enzyme. For virtual screening, we used well known and widely used in silico methods combined with personal in cerebro insights and experience. Regular docking only performed slightly better than random selection, but the performance was significantly improved upon incorporation of additional filters based on pharmacophore queries and electrostatic similarities. The best performance was achieved when logical selection was added. For the pose prediction, we utilized a similar consensus approach that amalgamated the results of the Glide-XP docking with structural knowledge and rescoring. The pose prediction results revealed that docking displayed reasonable performance in predicting the binding poses. However, prediction performance can be improved utilizing scientific experience and rescoring approaches. In both the virtual screening and pose prediction challenges, the top performance was achieved by our approaches. Here we describe the methods and strategies used in our approaches and discuss the rationale of their performances.     

Using physics-based pose predictions and free energy perturbation calculations to predict binding poses and relative binding affinities for FXR ligands in the D3R Grand Challenge 2

Computer-aided drug design has become an integral part of drug discovery and development in the pharmaceutical and biotechnology industry, and is nowadays extensively used in the lead identification and lead optimization phases. The drug design data resource (D3R) organizes challenges against blinded experimental data to prospectively test computational methodologies as an opportunity for improved methods and algorithms to emerge. We participated in Grand Challenge 2 to predict the crystallographic poses of 36 Farnesoid X Receptor (FXR)-bound ligands and the relative binding affinities for two designated subsets of 18 and 15 FXR-bound ligands. Here, we present our methodology for pose and affinity predictions and its evaluation after the release of the experimental data. For predicting the crystallographic poses, we used docking and physics-based pose prediction methods guided by the binding poses of native ligands. For FXR ligands with known chemotypes in the PDB, we accurately predicted their binding modes, while for those with unknown chemotypes the predictions were more challenging. Our group ranked #1st (based on the median RMSD) out of 46 groups, which submitted complete entries for the binding pose prediction challenge. For the relative binding affinity prediction challenge, we performed free energy perturbation (FEP) calculations coupled with molecular dynamics (MD) simulations. FEP/MD calculations displayed a high success rate in identifying compounds with better or worse binding affinity than the reference (parent) compound. Our studies suggest that when ligands with chemical precedent are available in the literature, binding pose predictions using docking and physics-based methods are reliable; however, predictions are challenging for ligands with completely unknown chemotypes. We also show that FEP/MD calculations hold predictive value and can nowadays be used in a high throughput mode in a lead optimization project provided that crystal structures of sufficiently high quality are available.     

Blind prediction of cyclohexane–water distribution coefficients from the SAMPL5 challenge

In the recent SAMPL5 challenge, participants submitted predictions for cyclohexane/water distribution coefficients for a set of 53 small molecules. Distribution coefficients (log D) replace the hydration free energies that were a central part of the past five SAMPL challenges. A wide variety of computational methods were represented by the 76 submissions from 18 participating groups. Here, we analyze submissions by a variety of error metrics and provide details for a number of reference calculations we performed. As in the SAMPL4 challenge, we assessed the ability of participants to evaluate not just their statistical uncertainty, but their model uncertainty—how well they can predict the magnitude of their model or force field error for specific predictions. Unfortunately, this remains an area where prediction and analysis need improvement. In SAMPL4 the top performing submissions achieved a root-mean-squared error (RMSE) around 1.5 kcal/mol. If we anticipate accuracy in log D predictions to be similar to the hydration free energy predictions in SAMPL4, the expected error here would be around 1.54 log units. Only a few submissions had an RMSE below 2.5 log units in their predicted log D values. However, distribution coefficients introduced complexities not present in past SAMPL challenges, including tautomer enumeration, that are likely to be important in predicting biomolecular properties of interest to drug discovery, therefore some decrease in accuracy would be expected. Overall, the SAMPL5 distribution coefficient challenge provided great insight into the importance of modeling a variety of physical effects. We believe these types of measurements will be a promising source of data for future blind challenges, especially in view of the relatively straightforward nature of the experiments and the level of insight provided.     

Large scale free energy calculations for blind predictions of protein–ligand binding: the D3R Grand Challenge 2015

We describe binding free energy calculations in the D3R Grand Challenge 2015 for blind prediction of the binding affinities of 180 ligands to Hsp90. The present D3R challenge was built around experimental datasets involving Heat shock protein (Hsp) 90, an ATP-dependent molecular chaperone which is an important anticancer drug target. The Hsp90 ATP binding site is known to be a challenging target for accurate calculations of ligand binding affinities because of the ligand-dependent conformational changes in the binding site, the presence of ordered waters and the broad chemical diversity of ligands that can bind at this site. Our primary focus here is to distinguish binders from nonbinders. Large scale absolute binding free energy calculations that cover over 3000 protein–ligand complexes were performed using the BEDAM method starting from docked structures generated by Glide docking. Although the ligand dataset in this study resembles an intermediate to late stage lead optimization project while the BEDAM method is mainly developed for early stage virtual screening of hit molecules, the BEDAM binding free energy scoring has resulted in a moderate enrichment of ligand screening against this challenging drug target. Results show that, using a statistical mechanics based free energy method like BEDAM starting from docked poses offers better enrichment than classical docking scoring functions and rescoring methods like Prime MM-GBSA for the Hsp90 data set in this blind challenge. Importantly, among the three methods tested here, only the mean value of the BEDAM binding free energy scores is able to separate the large group of binders from the small group of nonbinders with a gap of 2.4 kcal/mol. None of the three methods that we have tested provided accurate ranking of the affinities of the 147 active compounds. We discuss the possible sources of errors in the binding free energy calculations. The study suggests that BEDAM can be used strategically to discriminate binders from nonbinders in virtual screening and to more accurately predict the ligand binding modes prior to the more computationally expensive FEP calculations of binding affinity.     

The SAMPL4 host–guest blind prediction challenge: an overview

Prospective validation of methods for computing binding affinities can help assess their predictive power and thus set reasonable expectations for their performance in drug design applications. Supramolecular host–guest systems are excellent model systems for testing such affinity prediction methods, because their small size and limited conformational flexibility, relative to proteins, allows higher throughput and better numerical convergence. The SAMPL4 prediction challenge therefore included a series of host–guest systems, based on two hosts, cucurbit[7]uril and octa-acid. Binding affinities in aqueous solution were measured experimentally for a total of 23 guest molecules. Participants submitted 35 sets of computational predictions for these host–guest systems, based on methods ranging from simple docking, to extensive free energy simulations, to quantum mechanical calculations. Over half of the predictions provided better correlations with experiment than two simple null models, but most methods underperformed the null models in terms of root mean squared error and linear regression slope. Interestingly, the overall performance across all SAMPL4 submissions was similar to that for the prior SAMPL3 host–guest challenge, although the experimentalists took steps to simplify the current challenge. While some methods performed fairly consistently across both hosts, no single approach emerged as consistent top performer, and the nonsystematic nature of the various submissions made it impossible to draw definitive conclusions regarding the best choices of energy models or sampling algorithms. Salt effects emerged as an issue in the calculation of absolute binding affinities of cucurbit[7]uril-guest systems, but were not expected to affect the relative affinities significantly. Useful directions for future rounds of the challenge might involve encouraging participants to carry out some calculations that replicate each others’ studies, and to systematically explore parameter options.     

Molecular mechanics methods for predicting protein-ligand binding

Ligand binding affinity prediction is one of the most important applications of computational chemistry. However, accurately ranking compounds with respect to their estimated binding affinities to a biomolecular target remains highly challenging. We provide an overview of recent work using molecular mechanics energy functions to address this challenge. We briefly review methods that use molecular dynamics and Monte Carlo simulations to predict absolute and relative ligand binding free energies, as well as our own work in which we have developed a physics-based scoring method that can be applied to hundreds of thousands of compounds by invoking a number of simplifying approximations. In our previous studies, we have demonstrated that our scoring method is a promising approach for improving the discrimination between ligands that are known to bind and those that are presumed not to, in virtual screening of large compound databases. In new results presented here, we explore several improvements to our computational method including modifying the dielectric constant used for the protein and ligand interiors, and empirically scaling energy terms to compensate for deficiencies in the energy model. Future directions for further improving our physics-based scoring method are also discussed.     

Virtual screening of integrase inhibitors by large scale binding free energy calculations: the SAMPL4 challenge

As part of the SAMPL4 blind challenge, filtered AutoDock Vina ligand docking predictions and large scale binding energy distribution analysis method binding free energy calculations have been applied to the virtual screening of a focused library of candidate binders to the LEDGF site of the HIV integrase protein. The computational protocol leveraged docking and high level atomistic models to improve enrichment. The enrichment factor of our blind predictions ranked best among all of the computational submissions, and second best overall. This work represents to our knowledge the first example of the application of an all-atom physics-based binding free energy model to large scale virtual screening. A total of 285 parallel Hamiltonian replica exchange molecular dynamics absolute protein-ligand binding free energy simulations were conducted starting from docked poses. The setup of the simulations was fully automated, calculations were distributed on multiple computing resources and were completed in a 6-weeks period. The accuracy of the docked poses and the inclusion of intramolecular strain and entropic losses in the binding free energy estimates were the major factors behind the success of the method. Lack of sufficient time and computing resources to investigate additional protonation states of the ligands was a major cause of mispredictions. The experiment demonstrated the applicability of binding free energy modeling to improve hit rates in challenging virtual screening of focused ligand libraries during lead optimization.     

Validation of the AmpC β-lactamase binding site and identification of inhibitors with novel scaffolds

AmpC β-lactamase confers resistance to β-lactam antibiotics in multiple Gram-negative bacteria. Therefore, identification of non-β-lactam compounds that inhibit the enzyme is considered crucial to the development of novel antibacterial therapies. Given the highly solvent-exposed active site, it is important to study the induced-fit movements and water-mediated interactions to improve docking accuracy and virtual screening enrichments in structure-based design of new AmpC inhibitors. Here, we tested multiple models of the AmpC binding site to investigate the importance of conserved water molecules and binding site plasticity on molecular docking. The results indicate that at least one conserved water molecule greatly improves the binding pose predictions and virtual screening enrichments of known noncovalent AmpC inhibitors. The best model was tested prospectively in the virtual screening of about 6 million commercially available compounds. Sixty-one chemically diverse top-scoring compounds were experimentally tested, which led to the identification of seven previously unknown inhibitors. These findings validate the essential features of the AmpC binding site for molecular recognition and are useful for further optimization of identified inhibitors.     

SAMPL4 & DOCK3.7: lessons for automated docking procedures

The SAMPL4 challenges were used to test current automated methods for solvation energy, virtual screening, pose and affinity prediction of the molecular docking pipeline DOCK 3.7. Additionally, first-order models of binding affinity were proposed as milestones for any method predicting binding affinity. Several important discoveries about the molecular docking software were made during the challenge: (1) Solvation energies of ligands were five-fold worse than any other method used in SAMPL4, including methods that were similarly fast, (2) HIV Integrase is a challenging target, but automated docking on the correct allosteric site performed well in terms of virtual screening and pose prediction (compared to other methods) but affinity prediction, as expected, was very poor, (3) Molecular docking grid sizes can be very important, serious errors were discovered with default settings that have been adjusted for all future work. Overall, lessons from SAMPL4 suggest many changes to molecular docking tools, not just DOCK 3.7, that could improve the state of the art. Future difficulties and projects will be discussed.     

Identification of a minimal subset of receptor conformations for improved multiple conformation docking and two-step scoring

Docking and scoring are critical issues in virtual drug screening methods. Fast and reliable methods are required for the prediction of binding affinity especially when applied to a large library of compounds. The implementation of receptor flexibility and refinement of scoring functions for this purpose are extremely challenging in terms of computational speed. Here we propose a knowledge-based multiple-conformation docking method that efficiently accommodates receptor flexibility thus permitting reliable virtual screening of large compound libraries. Starting with a small number of active compounds, a preliminary docking operation is conducted on a large ensemble of receptor conformations to select the minimal subset of receptor conformations that provides a strong correlation between the experimental binding affinity (e.g., Ki, IC50) and the docking score. Only this subset is used for subsequent multiple-conformation docking of the entire data set of library (test) compounds. In conjunction with the multiple-conformation docking procedure, a two-step scoring scheme is employed by which the optimal scoring geometries obtained from the multiple-conformation docking are re-scored by a molecular mechanics energy function including desolvation terms. To demonstrate the feasibility of this approach, we applied this integrated approach to the estrogen receptor alpha (ERalpha) system for which published binding affinity data were available for a series of structurally diverse chemicals. The statistical correlation between docking scores and experimental values was significantly improved from those of single-conformation dockings. This approach led to substantial enrichment of the virtual screening conducted on mixtures of active and inactive ERalpha compounds.     

Clustering and classifying diverse HIV entry inhibitors using a novel consensus shape-based virtual screening approach: further evidence for multiple binding sites within the CCR5 extracellular pocket

HIV entry inhibitors have emerged as a new generation of antiretroviral drugs that block viral fusion with the CXCR4 and CCR5 membrane coreceptors. Several small molecule antagonists for these coreceptors have been developed, some of which are currently in clinical trials. However, because no crystal structures for the coreceptor proteins are available, the binding modes of the known inhibitors within the coreceptor extracellular pockets need to be analyzed by means of site-directed mutagenesis and computational experiments. Previous studies have indicated that there is more than one binding site within the CCR5 extracellular pocket. This article investigates and develops this hypothesis using a novel spherical harmonic-based consensus shape clustering approach. The consensus shape approach is evaluated using retrospective virtual screening of CXCR4 and CCR5 inhibitors. Multiple combinations of CCR5 ligands in multiple trial superpositions are constructed to find consensus queries that give high virtual screening enrichments. Receiver-operator-characteristic performance analyses for both CXCR4 and CCR5 inhibitors show that the new consensus shape matching approach gives better virtual screening enrichments than existing shape matching and docking virtual screening techniques. The results obtained also provide strong evidence to support the notion that there are three main binding sites within the CCR5 extracellular cavity.     

Computational fragment-based screening using RosettaLigand: the SAMPL3 challenge

SAMPL3 fragment based virtual screening challenge provides a valuable opportunity for researchers to test their programs, methods and screening protocols in a blind testing environment. We participated in SAMPL3 challenge and evaluated our virtual fragment screening protocol, which involves RosettaLigand as the core component by screening a 500 fragments Maybridge library against bovine pancreatic trypsin. Our study reaffirmed that the real test for any virtual screening approach would be in a blind testing environment. The analyses presented in this paper also showed that virtual screening performance can be improved, if a set of known active compounds is available and parameters and methods that yield better enrichment are selected. Our study also highlighted that to achieve accurate orientation and conformation of ligands within a binding site, selecting an appropriate method to calculate partial charges is important. Another finding is that using multiple receptor ensembles in docking does not always yield better enrichment than individual receptors. On the basis of our results and retrospective analyses from SAMPL3 fragment screening challenge we anticipate that chances of success in a fragment screening process could be increased significantly with careful selection of receptor structures, protein flexibility, sufficient conformational sampling within binding pocket and accurate assignment of ligand and protein partial charges.     

Fast and accurate predictions of binding free energies using MM‐PBSA and MM‐GBSA

In the drug discovery process, accurate methods of computing the affinity of small molecules with a biological target are strongly needed. This is particularly true for molecular docking and virtual screening methods, which use approximated scoring functions and struggle in estimating binding energies in correlation with experimental values. Among the various methods, MM‐PBSA and MM‐GBSA are emerging as useful and effective approaches. Although these methods are typically applied to large collections of equilibrated structures of protein‐ligand complexes sampled during molecular dynamics in water, the possibility to reliably estimate ligand affinity using a single energy‐minimized structure and implicit solvation models has not been explored in sufficient detail. Herein, we thoroughly investigate this hypothesis by comparing different methods for the generation of protein‐ligand complexes and diverse methods for free energy prediction for their ability to correlate with experimental values. The methods were tested on a series of structurally diverse inhibitors of Plasmodium falciparum DHFR with known binding mode and measured affinities. The results showed that correlations between MM‐PBSA or MM‐GBSA binding free energies with experimental affinities were in most cases excellent. Importantly, we found that correlations obtained with the use of a single protein‐ligand minimized structure and with implicit solvation models were similar to those obtained after averaging over multiple MD snapshots with explicit water molecules, with consequent save of computing time without loss of accuracy. When applied to a virtual screening experiment, such an approach proved to discriminate between true binders and decoy molecules and yielded significantly better enrichment curves. © 2009 Wiley Periodicals, Inc. J Comput Chem, 2010     

Predicting kinase selectivity profiles using Free-Wilson QSAR analysis

Kinases are involved in a variety of diseases such as cancer, diabetes, and arthritis. In recent years, many kinase small molecule inhibitors have been developed as potential disease treatments. Despite the recent advances, selectivity remains one of the most challenging aspects in kinase inhibitor design. To interrogate kinase selectivity, a panel of 45 kinase assays has been developed in-house at Pfizer. Here we present an application of in silico quantitative structure activity relationship (QSAR) models to extract rules from this experimental screening data and make reliable selectivity profile predictions for all compounds enumerated from virtual libraries. We also propose the construction of R-group selectivity profiles by deriving their activity contribution against each kinase using QSAR models. Such selectivity profiles can be used to provide better understanding of subtle structure selectivity relationships during kinase inhibitor design.     

Exhaustive docking and solvated interaction energy scoring: lessons learned from the SAMPL4 challenge

We continued prospective assessments of the Wilma–solvated interaction energy (SIE) platform for pose prediction, binding affinity prediction, and virtual screening on the challenging SAMPL4 data sets including the HIV-integrase inhibitor and two host–guest systems. New features of the docking algorithm and scoring function are tested here prospectively for the first time. Wilma–SIE provides good correlations with actual binding affinities over a wide range of binding affinities that includes strong binders as in the case of SAMPL4 host–guest systems. Absolute binding affinities are also reproduced with appropriate training of the scoring function on available data sets or from comparative estimation of the change in target’s vibrational entropy. Even when binding modes are known, SIE predictions lack correlation with experimental affinities within dynamic ranges below 2 kcal/mol as in the case of HIV-integrase ligands, but they correctly signaled the narrowness of the dynamic range. Using a common protein structure for all ligands can reduce the noise, while incorporating a more sophisticated solvation treatment improves absolute predictions. The HIV-integrase virtual screening data set consists of promiscuous weak binders with relatively high flexibility and thus it falls outside of the applicability domain of the Wilma–SIE docking platform. Despite these difficulties, unbiased docking around three known binding sites of the enzyme resulted in over a third of ligands being docked within 2 Å from their actual poses and over half of the ligands docked in the correct site, leading to better-than-random virtual screening results.     

Improving molecular docking through eHiTS’ tunable scoring function

We present three complementary approaches for score-tuning that improve docking performance in pose prediction, virtual screening and binding affinity assessment. The methodology utilizes experimental data to customize the scoring function for the system of interest considering the specific docking scenario. The tuning approach, which has been implemented as an automated utility in eHiTS, is introduced as a solution to one of the conundrums of the molecular docking paradigm, namely, the lack of a universally well performing scoring function. The accuracy of scoring functions has been shown to be generally system-dependent, and particularly lacking for binding energy and bio-activity predictions. In the proposed approach, pose and energy predictions are enhanced by adjusting the relative weights of the eHiTS energy terms to improve score-RMSD or score-affinity correlations. In a virtual screening context ligand-based similarity is used to rescale the docking score such that better enrichment factors are achieved. We discuss the algorithmic details of the methods, and demonstrate the effects of score tuning on a variety of targets, including CDK2, BACE1 and neuraminidase, as well as on the popular benchmarks—the Directory of Useful Decoys and the PDBBind database.     

Development of purely structure-based pharmacophores for the topoisomerase I-DNA-ligand binding pocket

Purely structure-based pharmacophores (SBPs) are an alternative method to ligand-based approaches and have the advantage of describing the entire interaction capability of a binding pocket. Here, we present the development of SBPs for topoisomerase I, an anticancer target with an unusual ligand binding pocket consisting of protein and DNA atoms. Different approaches to cluster and select pharmacophore features are investigated, including hierarchical clustering and energy calculations. In addition, the performance of SBPs is evaluated retrospectively and compared to the performance of ligand- and complex-based pharmacophores. SBPs emerge as a valid method in virtual screening and a complementary approach to ligand-focussed methods. The study further reveals that the choice of pharmacophore feature clustering and selection methods has a large impact on the virtual screening hit lists. A prospective application of the SBPs in virtual screening reveals that they can be used successfully to identify novel topoisomerase inhibitors.     

Blind prediction of solvation free energies from the SAMPL4 challenge

Here, we give an overview of the small molecule hydration portion of the SAMPL4 challenge, which focused on predicting hydration free energies for a series of 47 small molecules. These gas-to-water transfer free energies have in the past proven a valuable test of a variety of computational methods and force fields. Here, in contrast to some previous SAMPL challenges, we find a relatively wide range of methods perform quite well on this test set, with RMS errors in the 1.2 kcal/mol range for several of the best performing methods. Top-performers included a quantum mechanical approach with continuum solvent models and functional group corrections, alchemical molecular dynamics simulations with a classical all-atom force field, and a single-conformation Poisson–Boltzmann approach. While 1.2 kcal/mol is still a significant error, experimental hydration free energies covered a range of nearly 20 kcal/mol, so methods typically showed substantial predictive power. Here, a substantial new focus was on evaluation of error estimates, as predicting when a computational prediction is reliable versus unreliable has considerable practical value. We found, however, that in many cases errors are substantially underestimated, and that typically little effort has been invested in estimating likely error. We believe this is an important area for further research.     

Overview of the SAMPL6 host–guest binding affinity prediction challenge

Accurately predicting the binding affinities of small organic molecules to biological macromolecules can greatly accelerate drug discovery by reducing the number of compounds that must be synthesized to realize desired potency and selectivity goals. Unfortunately, the process of assessing the accuracy of current computational approaches to affinity prediction against binding data to biological macromolecules is frustrated by several challenges, such as slow conformational dynamics, multiple titratable groups, and the lack of high-quality blinded datasets. Over the last several SAMPL blind challenge exercises, host–guest systems have emerged as a practical and effective way to circumvent these challenges in assessing the predictive performance of current-generation quantitative modeling tools, while still providing systems capable of possessing tight binding affinities. Here, we present an overview of the SAMPL6 host–guest binding affinity prediction challenge, which featured three supramolecular hosts: octa-acid (OA), the closely related tetra-endo-methyl-octa-acid (TEMOA), and cucurbit[8]uril (CB8), along with 21 small organic guest molecules. A total of 119 entries were received from ten participating groups employing a variety of methods that spanned from electronic structure and movable type calculations in implicit solvent to alchemical and potential of mean force strategies using empirical force fields with explicit solvent models. While empirical models tended to obtain better performance than first-principle methods, it was not possible to identify a single approach that consistently provided superior results across all host–guest systems and statistical metrics. Moreover, the accuracy of the methodologies generally displayed a substantial dependence on the system considered, emphasizing the need for host diversity in blind evaluations. Several entries exploited previous experimental measurements of similar host–guest systems in an effort to improve their physical-based predictions via some manner of rudimentary machine learning; while this strategy succeeded in reducing systematic errors, it did not correspond to an improvement in statistical correlation. Comparison to previous rounds of the host–guest binding free energy challenge highlights an overall improvement in the correlation obtained by the affinity predictions for OA and TEMOA systems, but a surprising lack of improvement regarding root mean square error over the past several challenge rounds. The data suggests that further refinement of force field parameters, as well as improved treatment of chemical effects (e.g., buffer salt conditions, protonation states), may be required to further enhance predictive accuracy.