An efficient computational chemistry approach to generating negative data for drug discovery pipeline validation

Targeting OPA1 protein for therapeutic intervention in autosomal dominant optic atrophy: In silico drug discovery.

In a Rational Drug Design (RDD) one important step is the receptor-ligand interaction evaluation through molecular docking simulations. How it is a way impossible to test all available compounds for a target receptor, there is a need to select the most promising. One possible approach for such selection is to consider characteristics like a set of molecular properties called molecular descriptors. Aiming at describing these characteristics, we introduce a Data Warehouse (DW) model that integrates molecular descriptors from different public databases of compounds, as well as relates them with Virtual Screening (VS) experiments data. With the proposed DW we are able to produce proper data sets for classification mining experiments. We performed a case study with a VS considering as receptor the HIV-1 Protease receptor and 76 compounds. The data sets produced from our DW are composed by 7 molecular descriptors as the predictive attributes, and as a target attribute the discretized Free Energy of Binding (FEB) value between the ligands and the target receptor. By performing C4.5 algorithm over the generated data sets, we got decision-trees models that indicates which molecular descriptors and their respective values are relevant to influence on good FEB results.

Quantitative Structure Activity Relationship (QSAR) models can inform on the correlation between activities and structure-based molecular descriptors. This information is important for the understanding of the factors that govern molecular properties and for designing new compounds with favorable properties. Due to the large number of calculate-able descriptors and consequently, the much larger number of descriptors combinations, the derivation of QSAR models could be treated as an optimization problem. For continuous responses, metrics which are typically being optimized in this process are related to model performances on the training set, for example, and . Similar metrics, calculated on an external set of data (e.g., ), are used to evaluate the performances of the final models. A common theme of these metrics is that they are context -” ignorant”. In this work we propose that QSAR models should be evaluated based on their intended usage. More specifically, we argue that QSAR models developed for Virtual Screening (VS) should be derived and evaluated using a virtual screening-aware metric, e.g., an enrichment-based metric. To demonstrate this point, we have developed 21 Multiple Linear Regression (MLR) models for seven targets (three models per target), evaluated them first on validation sets and subsequently tested their performances on two additional test sets constructed to mimic small-scale virtual screening campaigns. As expected, we found no correlation between model performances evaluated by “classical” metrics, e.g., and and the number of active compounds picked by the models from within a pool of random compounds. In particular, in some cases models with favorable and/or values were unable to pick a single active compound from within the pool whereas in other cases, models with poor and/or values performed well in the context of virtual screening. We also found no significant correlation between the number of active compounds correctly identified by the models in the training, validation and test sets. Next, we have developed a new algorithm for the derivation of MLR models by optimizing an enrichment-based metric and tested its performances on the same datasets. We found that the best models derived in this manner showed, in most cases, much more consistent results across the training, validation and test sets and outperformed the corresponding MLR models in most virtual screening tests. Finally, we demonstrated that when tested as binary classifiers, models derived for the same targets by the new algorithm outperformed Random Forest (RF) and Support Vector Machine (SVM)-based models across training/validation/test sets, in most cases. We attribute the better performances of the Enrichment Optimizer Algorithm (EOA) models in VS to better handling of inactive random compounds. Optimizing an enrichment-based metric is therefore a promising strategy for the derivation of QSAR models for classification and virtual screening.

Docking is one of the most important steps in virtual screening pipelines, and it is an established method for examining potential interactions between ligands and receptors. However, this method is computationally expensive, and it is often among the last steps of the process of compound libraries evaluation. In this work, we investigate the feasibility of learning a deep neural network to predict the docking output directly from a two-dimensional compound structure. The developed protocol is orders of magnitude faster than typical docking software, and it returns ligand-receptor complexes encoded in the form of the interaction fingerprint. Its speed and efficiency unlock the application possibilities, such as screening compound libraries of vast size on the basis of contact patterns or docking score (derived on the basis of predicted interaction schemes). We tested our approach on several G protein-coupled receptor targets and 4 CYP enzymes in retrospective virtual screening experiments, and a variant of graph convolutional network appeared to be most effective in emulating docking results. The method can be easily used by the community based on the code available in the Supporting Information.

Rising CO2 levels, largely from flue gas emissions, are a significant contributor to global climate change. Adsorption using Metal-Organic Frameworks (MOFs) offers a promising solution for CO2 capture due to their high surface area, tunable porosity, and selectivity. To streamline the discovery of efficient MOFs, we developed a high-throughput virtual screening (HTVS) pipeline by integrating Grand-Canonical Monte Carlo (GCMC) simulations, molecular modeling and machine learning. We screened a filtered subset of the CoREMOF database and a user-defined hypothetical MOF bank to identify candidates with high CO2 adsorption capacity and selectivity. This approach yielded several high-performing MOFs, including five from the CoREMOF dataset and a new structure from the hypothetical bank exceeding the performance of existing MOFs. Our findings highlight the complex relationship between MOF geometries and CO2 capture performance, emphasizing the importance of features like open metal sites and pore geometry. This computational framework accelerates MOF discovery and provides valuable insights for experimental synthesis. Future work will focus on expanding the hypothetical MOF dataset, improving simulation accuracy, and employing advanced optimization techniques to enhance the screening process.

Virtual screening (VS) is a well-established method in the initial stages of many drug and material design projects. VS is typically performed using structure-based approaches such as molecular docking, or various ligand-based approaches. Most docking tools were designed to be as global as possible, and consequently only require knowledge on the 3D structure of the biotarget. In contrast, many ligand-based approaches (e.g., 3D-QSAR and pharmacophore) require prior development of project-specific predictive models. Depending on the type of model (e.g., classification or regression), predictive ability is typically evaluated using metrics of performance on either the training set () or the test set (e.g., specificity, selectivity or ). However, none of these metrics were developed with VS in mind, and consequently, their ability to reliably assess the performances of a model in the context of VS is at best limited. With this in mind we have recently reported the development of the enrichment optimization algorithm (EOA). EOA derives QSAR models in the form of multiple linear regression (MLR) equations for VS by optimizing an enrichment-based metric in the space of the descriptors. Here we present an improved version of the algorithm which better handles active compounds and which also takes into account information on inactive (either known inactive or decoy) compounds. We compared the improved EOA in small-scale VS experiments with three common docking tools, namely, Glide-SP, GOLD and AutoDock Vina, employing five molecular targets (acetylcholinesterase, human immunodeficiency virus type 1 protease, MAP kinase p38 alpha, urokinase-type plasminogen activator, and trypsin I). We found that EOA consistently outperformed all docking tools in terms of the area under the ROC curve (AUC) and EF1% metrics that measured the overall and initial success of the VS process, respectively. This was the case when the docking metrics were calculated based on a consensus approach and when they were calculated based on two different sets of single crystal structures. Finally, we propose that EOA could be combined with molecular docking to derive target-specific scoring functions.

BioProtIS: Streamlining protein-ligand interaction pipeline for analysis in genomic and transcriptomic exploration

Protein-protein interactions (PPIs) are attractive targets for drug design because of their essential role in numerous cellular processes and disease pathways. However, in general, PPIs display exposed binding pockets at the interface, and as such, have been largely unexploited for therapeutic interventions with low-molecular weight compounds. Here, we used docking and various rescoring strategies in an attempt to recover PPI inhibitors from a set of active and inactive molecules for 11 targets collected in ChEMBL and PubChem. Our focus is on the screening power of the various developed protocols and on using fast approaches so as to be able to apply such a strategy to the screening of ultralarge libraries in the future. First, we docked compounds into each target using the fast "pscreen" mode of the structure-based virtual screening (VS) package Surflex. Subsequently, the docking poses were postprocessed to derive a set of 3D topological descriptors: (i) shape similarity and (ii) interaction fingerprint similarity with a co-crystallized inhibitor, (iii) solvent-accessible surface area, and (iv) extent of deviation from the geometric center of a reference inhibitor. The derivatized descriptors, together with descriptor-scaled scoring functions, were utilized to investigate possible impacts on VS performance metrics. Moreover, four standalone scoring functions, RF-Score-VS (machine-learning), DLIGAND2 (knowledge-based), Vinardo (empirical), and X-SCORE (empirical), were employed to rescore the PPI compounds. Collectively, the results indicate that the topological scoring algorithms could be valuable both at a global level, with up to 79% increase in areas under the receiver operating characteristic curve for some targets, and in early stages, with up to a 4-fold increase in enrichment factors at 1% of the screened collections. Outstandingly, DLIGAND2 emerged as the best scoring function on this data set, outperforming all rescoring techniques in terms of VS metrics. The described methodology could help in the rational design of small-molecule PPI inhibitors and has direct applications in many therapeutic areas, including cancer, CNS, and infectious diseases such as COVID-19.

RNAs are a vast reservoir of untapped drug targets. Structure-based virtual screening (VS) identifies candidate molecules by leveraging binding site information, traditionally using molecular docking simulations. However, docking struggles to scale with large compound libraries and RNA targets. Machine learning offers a solution but remains underdeveloped for RNA due to limited data and practical evaluations. We introduce a data-driven VS pipeline tailored for RNA, utilizing coarse-grained 3D modeling, synthetic data augmentation, and RNA-specific self-supervision. Our model achieves a 10,000x speedup over docking while ranking active compounds in the top 2.8% on structurally distinct test sets. It is robust to binding site variations and successfully screens unseen RNA riboswitches in a 20,000-compound in-vitro microarray, with a mean enrichment factor of 2.93 at 1%. This marks the first experimentally validated success of structure-based deep learning for RNA VS.

SummaryAccurate prediction of the solubility of chemical substances in solvents remains a challenge. The sparsity of high-quality solubility data is recognized as the biggest hurdle in the development of robust data-driven methods for practical use. Nonetheless, the effects of the quality and quantity of data on aqueous solubility predictions have not yet been scrutinized. In this study, the roles of the size and the quality of data sets on the performances of the solubility prediction models are unraveled, and the concepts of actual and observed performances are introduced. In an effort to curtail the gap between actual and observed performances, a quality-oriented data selection method, which evaluates the quality of data and extracts the most accurate part of it through statistical validation, is designed. Applying this method on the largest publicly available solubility database and using a consensus machine learning approach, a top-performing solubility prediction model is achieved.

In this report are used two data sets involving the main antidiabetic enzyme targets α-amylase and α-glucosidase. The prediction of α-amylase and α-glucosidase inhibitory activity as antidiabetic is carried out using LDA and classification trees (CT). A large data set of 640 compounds for α-amylase and 1546 compounds in the case of α-glucosidase are selected to develop the tree model. In the case of CT-J48 have the better classification model performances for both targets with values above 80%-90% for the training and prediction sets, correspondingly. The best model shows an accuracy higher than 95% for training set; the model was also validated using 10-fold cross-validation procedure and through a test set achieving accuracy values of 85.32% and 86.80%, correspondingly. Additionally, the obtained model is compared with other approaches previously published in the international literature showing better results. Finally, we can say that the present results provided a double-target approach for increasing the estimation of antidiabetic chemicals identification aimed by double-way workflow in virtual screening pipelines.

We have previously examined the binding patterns of various substrates to human cytochrome P450 2D6 (CYP2D6) using a series of molecular modeling methods. In this study, we further explored the binding modes of various types of inhibitors to CYP2D6 using a combination of ligand- and protein-based modeling approaches. Firstly, we developed and validated a pharmacophore model for CYP2D6 inhibitors, which consisted of two hydrophobic features and one hydrogen bond acceptor feature. Secondly, we constructed and validated a quantitative structure-activity relationship (QSAR) model for CYP2D6 inhibitors which gave a poor to moderate prediction accuracy. Thirdly, a panel of CYP2D6 inhibitors were subject to molecular docking into the active site of wild-type and mutated CYP2D6 enzyme. We demonstrated that 8 residues in the active site (Leu213, Glu216, Ser217, Gln244, Asp301, Ser304, Ala305, and Phe483) played an important role in the binding to the inhibitors via hydrogen bond formation and/or π-π stacking interaction. Apparent changes in the binding modes of the inhibitors have been observed with Phe120Ile, Glu216Asp, Asp301Glu mutations in CYP2D6. Finally, we screened for potential binders/inhibitors from the Chinese herbal medicine Scutellaria baicalensis (Huangqin, Baikal Skullcap) using the established pharmacophore model for CYP2D6 inhibitors and molecular docking approach. Overall, 18 out of 40 compounds from S. baicalensis were mapped to the pharmacophore model of CYP2D6 inhibitors and most herbal compounds from S. baicalensis could be docked into the active site of CYP2D6. Our study has provided insights into the molecular mechanisms of interaction of synthetic and herbal compounds with human CYP2D6 and further benchmarking studies are needed to validate our modeling and virtual screening results.

Several in-house developed descriptors and our in-house docking tool ActDock were compared with virtual screening on the data set of useful decoys (DUD). The results were compared with the chemical fingerprint descriptor from ChemAxon and with the docking results of the original DUD publication. The DUD is the first published data set providing active molecules, decoys, and references for crystal structures of ligand-target complexes. The DUD was designed for the purpose of evaluating docking programs. It contains 2950 active compounds against a total of 40 target proteins. Furthermore, for every ligand the data set contains 36 structurally dissimilar decoy compounds with similar physicochemical properties. We extracted the ligands from the target proteins to extend the applicability of the data set to include ligand based virtual screening. From the 40 target proteins, 37 contained ligands that we used as query molecules for virtual screening evaluation. With this data set a large comparison was done between four different chemical fingerprints, a topological pharmacophore descriptor, the Flexophore descriptor, and ActDock. The Actelion docking tool relies on a MM2 forcefield and a pharmacophore point interaction statistic for scoring; the details are described in this publication. In terms of enrichment rates the chemical fingerprint descriptors performed better than the Flexophore and the docking tool. After removing molecules chemically similar to the query molecules the Flexophore descriptor outperformed the chemical descriptors and the topological pharmacophore descriptors. With the similarity matrix calculations used in this study it was shown that the Flexophore is well suited to find new chemical entities via "scaffold hopping". The Flexophore descriptor can be explored with a Java applet at http://www.cheminformatics.ch in the submenu Tools-->Flexophore. Its usage is free of charge and does not require registration.

<h3>Abstract</h3> Computational methods for virtual screening can dramatically accelerate early-stage drug discovery by identifying potential hits for a specified target. Docking algorithms traditionally use physics-based simulations to address this challenge...

In this work we announce and evaluate a high throughput virtual screening pipeline for in-silico screening of virtual compound databases using high performance computing (HPC). Notable features of this pipeline are an automated receptor preparation scheme with unsupervised binding site identification. The pipeline includes receptor/target preparation, ligand preparation, VinaLC docking calculation, and molecular mechanics/generalized Born surface area (MM/GBSA) rescoring using the GB model by Onufriev and co-workers [J. Chem. Theory Comput. 2007, 3, 156-169]. Furthermore, we leverage HPC resources to perform an unprecedented, comprehensive evaluation of MM/GBSA rescoring when applied to the DUD-E data set (Directory of Useful Decoys: Enhanced), in which we selected 38 protein targets and a total of ∼0.7 million actives and decoys. The computer wall time for virtual screening has been reduced drastically on HPC machines, which increases the feasibility of extremely large ligand database screening with more accurate methods. HPC resources allowed us to rescore 20 poses per compound and evaluate the optimal number of poses to rescore. We find that keeping 5-10 poses is a good compromise between accuracy and computational expense. Overall the results demonstrate that MM/GBSA rescoring has higher average receiver operating characteristic (ROC) area under curve (AUC) values and consistently better early recovery of actives than Vina docking alone. Specifically, the enrichment performance is target-dependent. MM/GBSA rescoring significantly out performs Vina docking for the folate enzymes, kinases, and several other enzymes. The more accurate energy function and solvation terms of the MM/GBSA method allow MM/GBSA to achieve better enrichment, but the rescoring is still limited by the docking method to generate the poses with the correct binding modes.