Application of Mixed-Solvent Molecular Dynamics Simulations for Prediction of Allosteric Sites on G Protein-Coupled Receptors.

The development of small molecule allosteric modulators acting at G protein-coupled receptors (GPCRs) is becoming increasingly attractive. Such compounds have advantages over traditional drugs acting at orthosteric sites on these receptors, in particular target specificity. However, the number and locations of druggable allosteric sites within most clinically relevant GPCRs are unknown. In the present study, we describe the development and application of a mixed-solvent molecular dynamics (MixMD)-based method for the identification of allosteric sites on GPCRs. The method employs small organic probes with druglike qualities to identify druggable hotspots in multiple replicate short-timescale simulations. As proof of principle, we first applied the method retrospectively to a test set of five GPCRs (cannabinoid receptor type 1, C-C chemokine receptor type 2, M2 muscarinic receptor, P2Y purinoceptor 1, and protease-activated receptor 2) with known allosteric sites in diverse locations. This resulted in the identification of the known allosteric sites on these receptors. We then applied the method to the µ-opioid receptor. Several allosteric modulators for this receptor are known, although the binding sites for these modulators are not known. The MixMD-based method revealed several potential allosteric sites on the mu-opioid receptor. Implementation of the MixMD-based method should aid future efforts in the structure-based drug design of drugs targeting allosteric sites on GPCRs. SIGNIFICANCE STATEMENT: Allosteric modulation of G protein-coupled receptors (GPCRs) has the potential to provide more selective drugs. However, there are limited structures of GPCRs bound to allosteric modulators, and obtaining such structures is problematic. Current computational methods utilize static structures and therefore may not identify hidden or cryptic sites. Here we describe the use of small organic probes and molecular dynamics to identify druggable allosteric hotspots on GPCRs. The results reinforce the importance of protein dynamics in allosteric site identification.

The structure and characterization of human cytochrome P450 8B1 supports future drug design for nonalcoholic fatty liver disease and diabetes.

Human cytochrome P450 8B1 (CYP8B1) is involved in conversion of cholesterol to bile acids. It hydroxylates the steroid ring at C12 to ultimately produce the bile acid cholic acid. Studies implicated this enzyme as a good drug target for nonalcoholic fatty liver disease and type 2 diabetes, but there are no selective inhibitors known for this enzyme and no structures to guide inhibitor development. Herein, the human CYP8B1 protein was generated and used to identify and characterize interactions with a series of azole inhibitors, which tend to be poorly selective P450 inhibitors. Structurally related miconazole, econazole, and tioconazole bound with submicromolar dissociation constants and were effective inhibitors of the native reaction. CYP8B was cocrystallized with S-tioconazole to yield the first X-ray structure. This inhibitor bound in the active site with its azole nitrogen coordinating the heme iron, consistent with inhibitor binding and inhibition assay data. Additionally, the CYP8B1 active site was compared with similar P450 enzymes to identify features that may facilitate the design of more selective inhibitors. Selective inhibitors should promote a better understanding of the role of CYP8B1 inhibition in normal physiology and disease states and provide a possible treatment for nonalcoholic fatty liver disease and type 2 diabetes.

Randomized Trial of a Palliative Care Intervention to Improve End-of-Life Care Discussions in Patients With Metastatic Breast Cancer.

BACKGROUND: Studies show that early, integrated palliative care (PC) improves quality of life (QoL) and end-of-life (EoL) care for patients with poor-prognosis cancers. However, the optimal strategy for delivering PC for those with advanced cancers who have longer disease trajectories, such as metastatic breast cancer (MBC), remains unknown. We tested the effect of a PC intervention on the documentation of EoL care discussions, patient-reported outcomes, and hospice utilization in this population. PATIENTS AND METHODS: Patients with MBC and clinical indicators of poor prognosis (n=120) were randomly assigned to receive an outpatient PC intervention (n=61) or usual care (n=59) between May 2, 2016, and December 26, 2018, at an academic cancer center. The intervention entailed 5 structured PC visits focusing on symptom management, coping, prognostic awareness, decision-making, and EoL planning. The primary outcome was documentation of EoL care discussions in the electronic health record (EHR). Secondary outcomes included patient-report of discussions with clinicians about EoL care, QoL, and mood symptoms at 6, 12, 18, and 24 weeks after baseline and hospice utilization. RESULTS: The rate of EoL care discussions documented in the EHR was higher among intervention patients versus those receiving usual care (67.2% vs 40.7%; P=.006), including a higher completion rate of a Medical Orders for Life-Sustaining Treatment form (39.3% vs 13.6%; P=.002). Intervention patients were also more likely to report discussing their EoL care wishes with their doctor (odds ratio [OR], 3.10; 95% CI, 1.21-7.94; P=.019) and to receive hospice services (OR, 4.03; 95% CI, 1.10-14.73; P=.035) compared with usual care patients. Study groups did not differ in patient-reported QoL or mood symptoms. CONCLUSIONS: This PC intervention significantly improved rates of discussion and documentation regarding EoL care and delivery of hospice services among patients with MBC, demonstrating that PC can be tailored to address the supportive care needs of patients with longer disease trajectories. ClinicalTrials.gov identifier: NCT02730858.

Crowdsourced identification of multi-target kinase inhibitors for RET- and TAU- based disease: The Multi-Targeting Drug DREAM Challenge.

A continuing challenge in modern medicine is the identification of safer and more efficacious drugs. Precision therapeutics, which have one molecular target, have been long promised to be safer and more effective than traditional therapies. This approach has proven to be challenging for multiple reasons including lack of efficacy, rapidly acquired drug resistance, and narrow patient eligibility criteria. An alternative approach is the development of drugs that address the overall disease network by targeting multiple biological targets ('polypharmacology'). Rational development of these molecules will require improved methods for predicting single chemical structures that target multiple drug targets. To address this need, we developed the Multi-Targeting Drug DREAM Challenge, in which we challenged participants to predict single chemical entities that target pro-targets but avoid anti-targets for two unrelated diseases: RET-based tumors and a common form of inherited Tauopathy. Here, we report the results of this DREAM Challenge and the development of two neural network-based machine learning approaches that were applied to the challenge of rational polypharmacology. Together, these platforms provide a potentially useful first step towards developing lead therapeutic compounds that address disease complexity through rational polypharmacology.

Computational Identification of Possible Allosteric Sites and Modulators of the SARS-CoV-2 Main Protease.

In this study, we target the main protease (Mpro) of the SARS-CoV-2 virus as it is a crucial enzyme for viral replication. Herein, we report three plausible allosteric sites on Mpro that can expand structure-based drug discovery efforts for new Mpro inhibitors. To find these sites, we used mixed-solvent molecular dynamics (MixMD) simulations, an efficient computational protocol that finds binding hotspots through mapping the surface of unbound proteins with 5% cosolvents in water. We have used normal mode analysis to support our claim of allosteric control for these sites. Further, we have performed virtual screening against the sites with 361 hits from Mpro screenings available through the National Center for Advancing Translational Sciences (NCATS). We have identified the NCATS inhibitors that bind to the remote sites better than the active site of Mpro, and we propose these molecules may be allosteric regulators of the system. After identifying our sites, new X-ray crystal structures were released that show fragment molecules in the sites we found, supporting the notion that these sites are accurate and druggable.

Mixed-solvent molecular dynamics simulation-based discovery of a putative allosteric site on regulator of G protein signaling 4.

Regulator of G protein signaling 4 (RGS4) is an intracellular protein that binds to the Gα subunit ofheterotrimeric G proteins and aids in terminating G protein coupled receptor signaling. RGS4 has been implicated in pain, schizophrenia, and the control of cardiac contractility. Inhibitors of RGS4 have been developed but bind covalently to cysteine residues on the protein. Therefore, we sought to identify alternative druggable sites on RGS4 using mixed-solvent molecular dynamics simulations, which employ low concentrations of organic probes to identify druggable hotspots on the protein. Pseudo-ligands were placed in consensus hotspots, and perturbation with normal mode analysis led to the identification and characterization of a putative allosteric site, which would be invaluable for structure-based drug design of non-covalent, small molecule inhibitors. Future studies on the mechanism of this allostery will aid in the development of novel therapeutics targeting RGS4.

Identification of Cryptic Binding Sites Using MixMD with Standard and Accelerated Molecular Dynamics.

Protein dynamics play an important role in small molecule binding and can pose a significant challenge in the identification of potential binding sites. Cryptic binding sites have been defined as sites which require significant rearrangement of the protein structure to become physically accessible to a ligand. Mixed-solvent MD (MixMD) is a computational protocol which maps the surface of the protein using molecular dynamics (MD) of the unbound protein solvated in a 5% box of probe molecules with explicit water. This method has successfully identified known active and allosteric sites which did not require reorganization. In this study, we apply the MixMD protocol to identify known cryptic sites of 12 proteins characterized by a wide range of conformational changes. Of these 12 proteins, three require reorganization of side chains, five require loop movements, and four require movement of more significant structures such as whole helices. In five cases, we find that standard MixMD simulations are able to map the cryptic binding sites with at least one probe type. In two cases (guanylate kinase and TIE-2), accelerated MD, which increases sampling of torsional angles, was necessary to achieve mapping of portions of the cryptic binding site missed by standard MixMD. For more complex systems where movement of a helix or domain is necessary, MixMD was unable to map the binding site even with accelerated dynamics, possibly due to the limited timescale (100 ns for individual simulations). In general, similar conformational dynamics are observed in water-only simulations and those with probe molecules. This could imply that the probes are not driving opening events but rather take advantage of mapping sites that spontaneously open as part of their inherent conformational behavior. Finally, we show that docking to an ensemble of conformations from the standard MixMD simulations performs better than docking the apo crystal structure in nine cases and even better than half of the bound crystal structures. Poorer performance was seen in docking to ensembles of conformations from the accelerated MixMD simulations.

Chemical validation of a druggable site on Hsp27/HSPB1 using in silico solvent mapping and biophysical methods.

Destabilizing mutations in small heat shock proteins (sHsps) are linked to multiple diseases; however, sHsps are conformationally dynamic, lack enzymatic function and have no endogenous chemical ligands. These factors render sHsps as classically "undruggable" targets and make it particularly challenging to identify molecules that might bind and stabilize them. To explore potential solutions, we designed a multi-pronged screening workflow involving a combination of computational and biophysical ligand-discovery platforms. Using the core domain of the sHsp family member Hsp27/HSPB1 (Hsp27c) as a target, we applied mixed solvent molecular dynamics (MixMD) to predict three possible binding sites, which we confirmed using NMR-based solvent mapping. Using this knowledge, we then used NMR spectroscopy to carry out a fragment-based drug discovery (FBDD) screen, ultimately identifying two fragments that bind to one of these sites. A medicinal chemistry effort improved the affinity of one fragment by ~50-fold (16 µM), while maintaining good ligand efficiency (~0.32 kcal/mol/non-hydrogen atom). Finally, we found that binding to this site partially restored the stability of disease-associated Hsp27 variants, in a redox-dependent manner. Together, these experiments suggest a new and unexpected binding site on Hsp27, which might be exploited to build chemical probes.

Inherent versus induced protein flexibility: Comparisons within and between apo and holo structures.

Understanding how ligand binding influences protein flexibility is important, especially in rational drug design. Protein flexibility upon ligand binding is analyzed herein using 305 proteins with 2369 crystal structures with ligands (holo) and 1679 without (apo). Each protein has at least two apo and two holo structures for analysis. The inherent variation in structures with and without ligands is first established as a baseline. This baseline is then compared to the change in conformation in going from the apo to holo states to probe induced flexibility. The inherent backbone flexibility across the apo structures is roughly the same as the variation across holo structures. The induced backbone flexibility across apo-holo pairs is larger than that of the apo or holo states, but the increase in RMSD is less than 0.5 Å. Analysis of χ1 angles revealed a distinctly different pattern with significant influences seen for ligand binding on side-chain conformations in the binding site. Within the apo and holo states themselves, the variation of the χ1 angles is the same. However, the data combining both apo and holo states show significant displacements. Upon ligand binding, χ1 angles are frequently pushed to new orientations outside the range seen in the apo states. Influences on binding-site variation could not be easily attributed to features such as ligand size or x-ray structure resolution. By combining these findings, we find that most binding site flexibility is compatible with the common practice in flexible docking, where backbones are kept rigid and side chains are allowed some degree of flexibility.

Free Energies and Entropies of Binding Sites Identified by MixMD Cosolvent Simulations.

In our recent efforts to map protein surfaces using mixed-solvent molecular dynamics (MixMD) (Ghanakota, P.; Carlson, H. A. Moving Beyond Active-Site Detection: MixMD Applied to Allosteric Systems. J. Phys. Chem. B 2016, 120, 8685-8695), we were able to successfully capture active sites and allosteric sites within the top-four most occupied hotspots. In this study, we describe our approach for estimating the thermodynamic profile of the binding sites identified by MixMD. First, we establish a framework for calculating free energies from MixMD simulations, and we compare our approach to alternative methods. Second, we present a means to obtain a relative ranking of the binding sites by their configurational entropy. The theoretical maximum and minimum free energy and entropy values achievable under such a framework along with the limitations of the techniques are discussed. Using this approach, the free energy and relative entropy ranking of the top-four MixMD binding sites were computed and analyzed across our allosteric protein targets: Abl Kinase, Androgen Receptor, Pdk1 Kinase, Farnesyl Pyrophosphate Synthase, Chk1 Kinase, Glucokinase, and Protein Tyrosine Phosphatase 1B.

Are there physicochemical differences between allosteric and competitive ligands?

Previous studies have compared the physicochemical properties of allosteric compounds to non-allosteric compounds. Those studies have found that allosteric compounds tend to be smaller, more rigid, more hydrophobic, and more drug-like than non-allosteric compounds. However, previous studies have not properly corrected for the fact that some protein targets have much more data than other systems. This generates concern regarding the possible skew that can be introduced by the inherent bias in the available data. Hence, this study aims to determine how robust the previous findings are to the addition of newer data. This study utilizes the Allosteric Database (ASD v3.0) and ChEMBL v20 to systematically obtain large datasets of both allosteric and competitive ligands. This dataset contains 70,219 and 9,511 unique ligands for the allosteric and competitive sets, respectively. Physically relevant compound descriptors were computed to examine the differences in their chemical properties. Particular attention was given to removing redundancy in the data and normalizing across ligand diversity and varied protein targets. The resulting distributions only show that allosteric ligands tend to be more aromatic and rigid and do not confirm the increase in hydrophobicity or difference in drug-likeness. These results are robust across different normalization schemes.

Predicting Displaceable Water Sites Using Mixed-Solvent Molecular Dynamics.

Water molecules are an important factor in protein-ligand binding. Upon binding of a ligand with a protein's surface, waters can either be displaced by the ligand or may be conserved and possibly bridge interactions between the protein and ligand. Depending on the specific interactions made by the ligand, displacing waters can yield a gain in binding affinity. The extent to which binding affinity may increase is difficult to predict, as the favorable displacement of a water molecule is dependent on the site-specific interactions made by the water and the potential ligand. Several methods have been developed to predict the location of water sites on a protein's surface, but the majority of methods are not able to take into account both protein dynamics and the interactions made by specific functional groups. Mixed-solvent molecular dynamics (MixMD) is a cosolvent simulation technique that explicitly accounts for the interaction of both water and small molecule probes with a protein's surface, allowing for their direct competition. This method has previously been shown to identify both active and allosteric sites on a protein's surface. Using a test set of eight systems, we have developed a method using MixMD to identify conserved and displaceable water sites. Conserved sites can be determined by an occupancy-based metric to identify sites which are consistently occupied by water even in the presence of probe molecules. Conversely, displaceable water sites can be found by considering the sites which preferentially bind probe molecules. Furthermore, the inclusion of six probe types allows the MixMD method to predict which functional groups are capable of displacing which water sites. The MixMD method consistently identifies sites which are likely to be nondisplaceable and predicts the favorable displacement of water sites that are known to be displaced upon ligand binding.

MixMD Probeview: Robust Binding Site Prediction from Cosolvent Simulations.

Mixed-solvent molecular dynamics (MixMD) is a cosolvent simulation technique for identifying binding hotspots and specific favorable interactions on a protein's surface. MixMD studies have the ability to identify these biologically relevant sites by examining the occupancy of the cosolvent over the course of the simulation. However, previous MixMD analysis required a great deal of manual inspection to identify relevant sites. To address this limitation, we have developed MixMD Probeview as a plugin for the freely available, open-source version of the molecular visualization program PyMOL. MixMD Probeview incorporates two analysis procedures: (1) identifying and ranking whole binding sites and (2) identifying and ranking local maxima for each probe type. These functionalities were validated using four common benchmark proteins, including two with both active and allosteric sites. In addition, three different cosolvent procedures were compared to examine the impact of including more than one cosolvent in the simulations. For all systems tested, MixMD Probeview successfully identified known active and allosteric sites based on the total occupancy of neutral probe molecules. As an easy-to-use PyMOL plugin, we expect that MixMD Probeview will facilitate identification and analysis of binding sites from cosolvent simulations performed on a wide range of systems.

ChemTreeMap: an interactive map of biochemical similarity in molecular datasets.

MOTIVATION: What if you could explain complex chemistry in a simple tree and share that data online with your collaborators? Computational biology often incorporates diverse chemical data to probe a biological question, but the existing tools for chemical data are ill-suited for the very large datasets inherent to bioinformatics. Furthermore, existing visualization methods often require an expert chemist to interpret the patterns. Biologists need an interactive tool for visualizing chemical information in an intuitive, accessible way that facilitates its integration into today's team-based biological research. RESULTS: ChemTreeMap is an interactive, bioinformatics tool designed to explore chemical space and mine the relationships between chemical structure, molecular properties, and biological activity. ChemTreeMap synergistically combines extended connectivity fingerprints and a neighbor-joining algorithm to produce a hierarchical tree with branch lengths proportional to molecular similarity. Compound properties are shown by leaf color, size and outline to yield a user-defined visualization of the tree. Two representative analyses are included to demonstrate ChemTreeMap's capabilities and utility: assessing dataset overlap and mining structure-activity relationships. AVAILABILITY AND IMPLEMENTATION: The examples from this paper may be accessed at http://ajing.github.io/ChemTreeMap/ Code for the server and client are available in the Supplementary Information, at the aforementioned github site, and on Docker Hub (https://hub.docker.com) with the nametag ajing/chemtreemap. CONTACT: carlsonh@umich.eduSupplementary information: Supplementary data are available at Bioinformatics online.

Comparing pharmacophore models derived from crystallography and NMR ensembles.

NMR and X-ray crystallography are the two most widely used methods for determining protein structures. Our previous study examining NMR versus X-Ray sources of protein conformations showed improved performance with NMR structures when used in our Multiple Protein Structures (MPS) method for receptor-based pharmacophores (Damm, Carlson, J Am Chem Soc 129:8225-8235, 2007). However, that work was based on a single test case, HIV-1 protease, because of the rich data available for that system. New data for more systems are available now, which calls for further examination of the effect of different sources of protein conformations. The MPS technique was applied to Growth factor receptor bound protein 2 (Grb2), Src SH2 homology domain (Src-SH2), FK506-binding protein 1A (FKBP12), and Peroxisome proliferator-activated receptor-γ (PPAR-γ). Pharmacophore models from both crystal and NMR ensembles were able to discriminate between high-affinity, low-affinity, and decoy molecules. As we found in our original study, NMR models showed optimal performance when all elements were used. The crystal models had more pharmacophore elements compared to their NMR counterparts. The crystal-based models exhibited optimum performance only when pharmacophore elements were dropped. This supports our assertion that the higher flexibility in NMR ensembles helps focus the models on the most essential interactions with the protein. Our studies suggest that the "extra" pharmacophore elements seen at the periphery in X-ray models arise as a result of decreased protein flexibility and make very little contribution to model performance.

Recent improvements to Binding MOAD: a resource for protein-ligand binding affinities and structures.

For over 10 years, Binding MOAD (Mother of All Databases; http://www.BindingMOAD.org) has been one of the largest resources for high-quality protein-ligand complexes and associated binding affinity data. Binding MOAD has grown at the rate of 1994 complexes per year, on average. Currently, it contains 23,269 complexes and 8156 binding affinities. Our annual updates curate the data using a semi-automated literature search of the references cited within the PDB file, and we have recently upgraded our website and added new features and functionalities to better serve Binding MOAD users. In order to eliminate the legacy application server of the old platform and to accommodate new changes, the website has been completely rewritten in the LAMP (Linux, Apache, MySQL and PHP) environment. The improved user interface incorporates current third-party plugins for better visualization of protein and ligand molecules, and it provides features like sorting, filtering and filtered downloads. In addition to the field-based searching, Binding MOAD now can be searched by structural queries based on the ligand. In order to remove redundancy, Binding MOAD records are clustered in different families based on 90% sequence identity. The new Binding MOAD, with the upgraded platform, features and functionalities, is now equipped to better serve its users.

Identifying binding hot spots on protein surfaces by mixed-solvent molecular dynamics: HIV-1 protease as a test case.

Mixed-solvent molecular dynamics (MixMD) simulations use full protein flexibility and competition between water and small organic probes to achieve accurate hot-spot mapping on protein surfaces. In this study, we improved MixMD using human immunodeficiency virus type-1 protease (HIVp) as the test case. We used three probe-water solutions (acetonitrile-water, isopropanol-water, and pyrimidine-water), first at 50% w/w concentration and later at 5% v/v. Paradoxically, better mapping was achieved by using fewer probes; 5% simulations gave a superior signal-to-noise ratio and far fewer spurious hot spots than 50% MixMD. Furthermore, very intense and well-defined probe occupancies were observed in the catalytic site and potential allosteric sites that have been confirmed experimentally. The Eye site, an allosteric site underneath the flap of HIVp, has been confirmed by the presence of a 5-nitroindole fragment in a crystal structure. MixMD also mapped two additional hot spots: the Exo site (between the Gly16-Gly17 and Cys67-Gly68 loops) and the Face site (between Glu21-Ala22 and Val84-Ile85 loops). The Exo site was observed to overlap with crystallographic additives such as acetate and dimethyl sulfoxide that are present in different crystal forms of the protein. Analysis of crystal structures of HIVp in different symmetry groups has shown that some surface sites are common interfaces for crystal contacts, which means that they are surfaces that are relatively easy to desolvate and complement with organic molecules. MixMD should identify these sites; in fact, their occupancy values help establish a solid cut-off where "druggable" sites are required to have higher occupancies than the crystal-packing faces.

CSAR Benchmark Exercise 2013: Evaluation of Results from a Combined Computational Protein Design, Docking, and Scoring/Ranking Challenge.

Community Structure-Activity Resource (CSAR) conducted a benchmark exercise to evaluate the current computational methods for protein design, ligand docking, and scoring/ranking. The exercise consisted of three phases. The first phase required the participants to identify and rank order which designed sequences were able to bind the small molecule digoxigenin. The second phase challenged the community to select a near-native pose of digoxigenin from a set of decoy poses for two of the designed proteins. The third phase investigated the ability of current methods to rank/score the binding affinity of 10 related steroids to one of the designed proteins (pKd = 4.1 to 6.7). We found that 11 of 13 groups were able to correctly select the sequence that bound digoxigenin, with most groups providing the correct three-dimensional structure for the backbone of the protein as well as all atoms of the active-site residues. Eleven of the 14 groups were able to select the appropriate pose from a set of plausible decoy poses. The ability to predict absolute binding affinities is still a difficult task, as 8 of 14 groups were able to correlate scores to affinity (Pearson-r > 0.7) of the designed protein for congeneric steroids and only 5 of 14 groups were able to correlate the ranks of the 10 related ligands (Spearman-ρ > 0.7).

CSAR 2014: A Benchmark Exercise Using Unpublished Data from Pharma.

The 2014 CSAR Benchmark Exercise was the last community-wide exercise that was conducted by the group at the University of Michigan, Ann Arbor. For this event, GlaxoSmithKline (GSK) donated unpublished crystal structures and affinity data from in-house projects. Three targets were used: tRNA (m1G37) methyltransferase (TrmD), Spleen Tyrosine Kinase (SYK), and Factor Xa (FXa). A particularly strong feature of the GSK data is its large size, which lends greater statistical significance to comparisons between different methods. In Phase 1 of the CSAR 2014 Exercise, participants were given several protein-ligand complexes and asked to identify the one near-native pose from among 200 decoys provided by CSAR. Though decoys were requested by the community, we found that they complicated our analysis. We could not discern whether poor predictions were failures of the chosen method or an incompatibility between the participant's method and the setup protocol we used. This problem is inherent to decoys, and we strongly advise against their use. In Phase 2, participants had to dock and rank/score a set of small molecules given only the SMILES strings of the ligands and a protein structure with a different ligand bound. Overall, docking was a success for most participants, much better in Phase 2 than in Phase 1. However, scoring was a greater challenge. No particular approach to docking and scoring had an edge, and successful methods included empirical, knowledge-based, machine-learning, shape-fitting, and even those with solvation and entropy terms. Several groups were successful in ranking TrmD and/or SYK, but ranking FXa ligands was intractable for all participants. Methods that were able to dock well across all submitted systems include MDock,1 Glide-XP,2 PLANTS,3 Wilma,4 Gold,5 SMINA,6 Glide-XP2/PELE,7 FlexX,8 and MedusaDock.9 In fact, the submission based on Glide-XP2/PELE7 cross-docked all ligands to many crystal structures, and it was particularly impressive to see success across an ensemble of protein structures for multiple targets. For scoring/ranking, submissions that showed statistically significant achievement include MDock1 using ITScore1,10 with a flexible-ligand term,11 SMINA6 using Autodock-Vina,12,13 FlexX8 using HYDE,14 and Glide-XP2 using XP DockScore2 with and without ROCS15 shape similarity.16 Of course, these results are for only three protein targets, and many more systems need to be investigated to truly identify which approaches are more successful than others. Furthermore, our exercise is not a competition.

D3R grand challenge 2015: Evaluation of protein-ligand pose and affinity predictions.

The Drug Design Data Resource (D3R) ran Grand Challenge 2015 between September 2015 and February 2016. Two targets served as the framework to test community docking and scoring methods: (1) HSP90, donated by AbbVie and the Community Structure Activity Resource (CSAR), and (2) MAP4K4, donated by Genentech. The challenges for both target datasets were conducted in two stages, with the first stage testing pose predictions and the capacity to rank compounds by affinity with minimal structural data; and the second stage testing methods for ranking compounds with knowledge of at least a subset of the ligand-protein poses. An additional sub-challenge provided small groups of chemically similar HSP90 compounds amenable to alchemical calculations of relative binding free energy. Unlike previous blinded Challenges, we did not provide cognate receptors or receptors prepared with hydrogens and likewise did not require a specified crystal structure to be used for pose or affinity prediction in Stage 1. Given the freedom to select from over 200 crystal structures of HSP90 in the PDB, participants employed workflows that tested not only core docking and scoring technologies, but also methods for addressing water-mediated ligand-protein interactions, binding pocket flexibility, and the optimal selection of protein structures for use in docking calculations. Nearly 40 participating groups submitted over 350 prediction sets for Grand Challenge 2015. This overview describes the datasets and the organization of the challenge components, summarizes the results across all submitted predictions, and considers broad conclusions that may be drawn from this collaborative community endeavor.