Ensemble learning from ensemble docking: revisiting the optimum ensemble size problem

Despite considerable advances obtained by applying machine learning approaches in protein–ligand affinity predictions, the incorporation of receptor flexibility has remained an important bottleneck. While ensemble docking has been used widely as a solution to this problem, the optimum choice of receptor conformations is still an open question considering the issues related to the computational cost and false positive pose predictions. Here, a combination of ensemble learning and ensemble docking is suggested to rank different conformations of the target protein in light of their importance for the final accuracy of the model. Available X-ray structures of cyclin-dependent kinase 2 (CDK2) in complex with different ligands are used as an initial receptor ensemble, and its redundancy is removed through a graph-based redundancy removal, which is shown to be more efficient and less subjective than clustering-based representative selection methods. A set of ligands with available experimental affinity are docked to this nonredundant receptor ensemble, and the energetic features of the best scored poses are used in an ensemble learning procedure based on the random forest method. The importance of receptors is obtained through feature selection measures, and it is shown that a few of the most important conformations are sufficient to reach 1 kcal/mol accuracy in affinity prediction with considerable improvement of the early enrichment power of the models compared to the different ensemble docking without learning strategies. A clear strategy has been provided in which machine learning selects the most important experimental conformers of the receptor among a large set of protein–ligand complexes while simultaneously maintaining the final accuracy of affinity predictions at the highest level possible for available data. Our results could be informative for future attempts to design receptor-specific docking-rescoring strategies.

A computational study of cooperative binding to multiple SARS-CoV-2 proteins

Structure-based drug design targeting the SARS-CoV-2 virus has been greatly facilitated by available virus-related protein structures. However, there is an urgent need for effective, safe small-molecule drugs to control the spread of the virus and variants. While many efforts are devoted to searching for compounds that selectively target individual proteins, we investigated the potential interactions between eight proteins related to SARS-CoV-2 and more than 600 compounds from a traditional Chinese medicine which has proven effective at treating the viral infection. Our original ensemble docking and cooperative docking approaches, followed by a total of over 16-micorsecond molecular simulations, have identified at least 9 compounds that may generally bind to key SARS-CoV-2 proteins. Further, we found evidence that some of these compounds can simultaneously bind to the same target, potentially leading to cooperative inhibition to SARS-CoV-2 proteins like the Spike protein and the RNA-dependent RNA polymerase. These results not only present a useful computational methodology to systematically assess the anti-viral potential of small molecules, but also point out a new avenue to seek cooperative compounds toward cocktail therapeutics to target more SARS-CoV-2-related proteins.

DATABASE ENRICHMENTS OF MAO-B THROUGH ENSEMBLE DOCKING

Objective: The recent growth of highly resoluted crystallographic structures, together with the continuous improvements of the computing power, has established molecular docking as a leading drug design technique. However, the problems concerning the receptor flexibility and the lowered ability of docking software to correctly score the occurred interactions in some receptors are still relevant. Methods: Recently, several research groups have reported an enhancement in enrichment values when ensemble docking has been applied. Therefore, we utilized the latest technique for a dataset of Monoamine Oxidase–B (MAO-B) inhibitors. The docking program GOLD 5.3 was used in our study. Several docking parameters (grid space, scoring functions and ligand flexibility) were altered in order to achieve the optimal docking protocol. Results: The results of 200 000+docking simulations are represented in a modest table. The ensembled simulations demonstrated low ability of the docking software to correctly score the actives seeded in the dataset. However, the superimposed complex-1S3B-1OJA-1OJC, achieved a moderate enrichment value equaled to 9. No significant improvements were noted when five complexed receptors were employed. Conclusion: As a conclusion, it should be noted that in some cases the ensemble docking enhanced the database enrichments, however overall the value is not suitable for future virtual screening. Further investigations in that area should be considered.

Novel High Affinity Sigma-1 Receptor Ligands from Minimal Ensemble Docking-Based Virtual Screening

Sigma-1 receptor (S1R) is an intracellular, multi-functional, ligand operated protein that also acts as a chaperone. It is considered as a pluripotent drug target in several pathologies. The publication of agonist and antagonist bound receptor structures has paved the way for receptor-based in silico drug design. However, recent studies on this subject payed no attention to the structural differences of agonist and antagonist binding. In this work, we have developed a new ensemble docking-based virtual screening protocol utilizing both agonist and antagonist bound S1R structures. This protocol was used to screen our in-house compound library. The S1R binding affinities of the 40 highest ranked compounds were measured in competitive radioligand binding assays and the sigma-2 receptor (S2R) affinities of the best S1R binders were also determined. This way three novel high affinity S1R ligands were identified and one of them exhibited a notable S1R/S2R selectivity.