A comprehensive comparative assessment of 3D molecular similarity tools in ligand-based virtual screening

3MTox: A motif-level graph-based multi-view chemical language model for toxicity identification with deep interpretation

Toxicity identification plays a key role in maintaining human health, as it can alert humans to the potential hazards caused by long-term exposure to a wide variety of chemical compounds. Experimental methods for determining toxicity are time-consuming, and costly, while computational methods offer an alternative for the early identification of toxicity. For example, some classical ML and DL methods, which demonstrate excellent performance in toxicity prediction. However, these methods also have some defects, such as over-reliance on artificial features and easy overfitting, etc. Proposing novel models with superior prediction performance is still an urgent task. In this study, we propose a motifs-level graph-based multi-view pretraining language model, called 3MTox, for toxicity identification. The 3MTox model uses Bidirectional Encoder Representations from Transformers (BERT) as the backbone framework, and a motif graph as input. The results of extensive experiments showed that our 3MTox model achieved state-of-the-art performance on toxicity benchmark datasets and outperformed the baseline models considered. In addition, the interpretability of the model ensures that the it can quickly and accurately identify toxicity sites in a given molecule, thereby contributing to the determination of the status of toxicity and associated analyses. We think that the 3MTox model is among the most promising tools that are currently available for toxicity identification.

On the relevance of query definition in the performance of 3D ligand-based virtual screening

Ligand-based virtual screening (LBVS) methods are widely used to explore the vast chemical space in the search of novel compounds resorting to a variety of properties encoded in 1D, 2D or 3D descriptors. The success of 3D-LBVS is affected by the overlay of molecular pairs, thus making selection of the template compound, search of accessible conformational space and choice of the query conformation to be potential factors that modulate the successful retrieval of actives. This study examines the impact of adopting different choices for the query conformation of the template, paying also attention to the influence exerted by the structural similarity between templates and actives. The analysis is performed using PharmScreen, a 3D LBVS tool that relies on similarity measurements of the hydrophobic/philic pattern of molecules, and Phase Shape, which is based on the alignment of atom triplets followed by refinement of the volume overlap. The study is performed for the original DUD-E+ database and a Morgan Fingerprint filtered version (denoted DUD-E+-Diverse; available in https://github.com/Pharmacelera/Query-models-to-3DLBVS ), which was prepared to minimize the 2D resemblance between template and actives. Although in most cases the query conformation exhibits a mild influence on the overall performance, a critical analysis is made to disclose factors, such as the content of structural features between template and actives and the induction of conformational strain in the template, that underlie the drastic impact of the query definition in the recovery of actives for certain targets. The findings of this research also provide valuable guidance for assisting the selection of the query definition in 3D LBVS campaigns.

SENSAAS-Flex: a joint optimization approach for aligning 3D shapes and exploring the molecular conformation space

MOTIVATION

Popular shape-based alignment methods handle molecular flexibility by utilizing conformational ensembles to select the most fitted conformer. However, the initial conformer library generation step is computationally intensive and limiting to the overall alignment process. In this work, we describe a method to perform flexible alignment of two molecular shapes by optimizing the 3D conformation. SENSAAS-Flex, an add-on to the SENSAAS tool, is able to proceed from a limited set of initial conformers through an iterative process where additional conformational optimizations are made at the substructure level and constrained by the target shape.

RESULTS

In self- and cross-alignment experiments, SENSAAS-Flex is able to reproduce the crystal structure geometry of ligands of the AstraZeneca Molecule Overlay Test set and PDBbind refined dataset. Our study shows that the point-based representation of molecular surfaces is appropriate in terms of shape constraint to sample the conformational space and perform flexible molecular alignments.

AVAILABILITY AND IMPLEMENTATION

The documentation and source code are available at https://chemoinfo.ipmc.cnrs.fr/Sensaas-flex/sensaas-flex-main.tar.gz.

AODB: A comprehensive database for antioxidants including small molecules, peptides and proteins

Antioxidants are widely used in the fields of food, medicine, nutraceuticals, and cosmetics. Given their important roles in promoting and maintaining human health, a large number of antioxidants have been reported. Some antioxidant-related databases have been developed; however, the annotation of antioxidants and related information stored in existing databases is incomplete and requires more efficient retrieval methods. This study aimed to develop a manually curated comprehensive antioxidant database (AODB). Currently, it stores 56,666 small molecules tested for antioxidant activity, 1480 antioxidant peptides, and 998 antioxidant proteins, including their structures, names, antioxidant assay records, computable physicochemical and ADMET properties, and sources. AODB supports text search and mining, 2D and 3D chemical structure search, and BLAST-based protein sequence search, enabling users to retrieve antioxidant data quickly and easily. AODB, as a one-stop antioxidant database, can facilitate the exploration of antioxidants and potential applications. AODB is publicly available and updated annually at https://aodb.idruglab.cn/.

HiGNN: A Hierarchical Informative Graph Neural Network for Molecular Property Prediction Equipped with Feature-Wise Attention

Elucidating and accurately predicting the druggability and bioactivities of molecules plays a pivotal role in drug design and discovery and remains an open challenge. Recently, graph neural networks (GNNs) have made remarkable advancements in graph-based molecular property prediction. However, current graph-based deep learning methods neglect the hierarchical information of molecules and the relationships between feature channels. In this study, we propose a well-designed hierarchical informative graph neural network (termed HiGNN) framework for predicting molecular property by utilizing a corepresentation learning of molecular graphs and chemically synthesizable breaking of retrosynthetically interesting chemical substructure (BRICS) fragments. Furthermore, a plug-and-play feature-wise attention block is first designed in HiGNN architecture to adaptively recalibrate atomic features after the message passing phase. Extensive experiments demonstrate that HiGNN achieves state-of-the-art predictive performance on many challenging drug discovery-associated benchmark data sets. In addition, we devise a molecule-fragment similarity mechanism to comprehensively investigate the interpretability of the HiGNN model at the subgraph level, indicating that HiGNN as a powerful deep learning tool can help chemists and pharmacists identify the key components of molecules for designing better molecules with desired properties or functions. The source code is publicly available at https://github.com/idruglab/hignn.

CSAlign and CSAlign-Dock: Structure alignment of ligands considering full flexibility and application to protein-ligand docking

Structure prediction of protein-ligand complexes, called protein-ligand docking, is a critical computational technique that can be used to understand the underlying principle behind the protein functions at the atomic level and to design new molecules regulating the functions. Protein-ligand docking methods have been employed in structure-based drug discovery for hit discovery and lead optimization. One of the important technical challenges in protein-ligand docking is to account for protein conformational changes induced by ligand binding. A small change such as a single side-chain rotation upon ligand binding can hinder accurate docking. Here we report an increase in docking performance achieved by structure alignment to known complex structures. First, a fully flexible compound-to-compound alignment method CSAlign is developed by global optimization of a shape score. Next, the alignment method is combined with a docking algorithm to dock a new ligand to a target protein when a reference protein-ligand complex structure is available. This alignment-based docking method, called CSAlign-Dock, showed superior performance to ab initio docking methods in cross-docking benchmark tests. Both CSAlign and CSAlign-Dock are freely available as a web server at https://galaxy.seoklab.org/csalign.

FP-GNN: a versatile deep learning architecture for enhanced molecular property prediction

Accurate prediction of molecular properties, such as physicochemical and bioactive properties, as well as ADME/T (absorption, distribution, metabolism, excretion and toxicity) properties, remains a fundamental challenge for molecular design, especially for drug design and discovery. In this study, we advanced a novel deep learning architecture, termed FP-GNN (fingerprints and graph neural networks), which combined and simultaneously learned information from molecular graphs and fingerprints for molecular property prediction. To evaluate the FP-GNN model, we conducted experiments on 13 public datasets, an unbiased LIT-PCBA dataset and 14 phenotypic screening datasets for breast cell lines. Extensive evaluation results showed that compared to advanced deep learning and conventional machine learning algorithms, the FP-GNN algorithm achieved state-of-the-art performance on these datasets. In addition, we analyzed the influence of different molecular fingerprints, and the effects of molecular graphs and molecular fingerprints on the performance of the FP-GNN model. Analysis of the anti-noise ability and interpretation ability also indicated that FP-GNN was competitive in real-world situations. Collectively, FP-GNN algorithm can assist chemists, biologists and pharmacists in predicting and discovering better molecules with desired functions or properties.

Ligand- and structure-based identification of novel CDK9 inhibitors for the potential treatment of leukemia

Cyclin-dependent kinase 9 (CDK9) plays a vital role in controlling cell transcription and has been an attractive target for cancer treatment. Herein, ten predictive models derived from 1330 unique molecules against CDK9 were constructed based on molecular fingerprints and graphs using two conventional machine learning and four deep learning methods. The evaluation results showed that FP-GNN deep learning architecture performed best for CDK9 inhibitors prediction with the highest BA and F1 values of 0.681 and 0.912 for testing set. We then performed virtual screening to identify new CDK9 inhibitors by incorporating the optimal established predictive model and molecular docking. Five compounds were identified to show broad anticancer activity against various cancer cell lines through bioassays. For example, C9 exhibited antiproliferative activities against HeLa, MOLM-13 and MDA-MB-231 with IC₅₀ values of 2.53, 3.92 and 11.65 μM. Kinase inhibition assay results demonstrated that these compounds displayed submicromolar (214 ∼ 504 nM) inhibitory activities against CDK9. Further cellular mechanism evaluation revealed that C9 suppressed the activity of CDK9 and interfered with the expression of Mcl-1 and cleaved PARP in MOLM-13 cells, resulting in the induction of cellular apoptosis. In addition, C9 displayed a good stability in rat liver microsomes, artificial gastrointestinal fluid and plasm. An online platform (called DEEPCDK9Pred) was developed based on the FP-GNN models to predict or design new CDK9 inhibitors. Collectively, our findings demonstrated that FP-GNN algorithm can achieve accurate prediction of CDK9 inhibitors and the subsequent discovery of C9 as a new potential CDK9 inhibitor deserves further structural modification for the treatment of leukemia.

Simulation and Machine Learning Methods for Ion-Channel Structure Determination, Mechanistic Studies and Drug Design

Ion channels are expressed in almost all living cells, controlling the in-and-out communications, making them ideal drug targets, especially for central nervous system diseases. However, owing to their dynamic nature and the presence of a membrane environment, ion channels remain difficult targets for the past decades. Recent advancement in cryo-electron microscopy and computational methods has shed light on this issue. An explosion in high-resolution ion channel structures paved way for structure-based rational drug design and the state-of-the-art simulation and machine learning techniques dramatically improved the efficiency and effectiveness of computer-aided drug design. Here we present an overview of how simulation and machine learning-based methods fundamentally changed the ion channel-related drug design at different levels, as well as the emerging trends in the field.

In Silico Drug Repositioning to Target the SARS-CoV-2 Main Protease as Covalent Inhibitors Employing a Combined Structure-Based Virtual Screening Strategy of Pharmacophore Models and Covalent Docking

The epidemic caused by the SARS-CoV-2 coronavirus, which has spread rapidly throughout the world, requires urgent and effective treatments considering that the appearance of viral variants limits the efficacy of vaccines. The main protease of SARS-CoV-2 (M^pro) is a highly conserved cysteine proteinase, fundamental for the replication of the coronavirus and with a specific cleavage mechanism that positions it as an attractive therapeutic target for the proposal of irreversible inhibitors. A structure-based strategy combining 3D pharmacophoric modeling, virtual screening, and covalent docking was employed to identify the interactions required for molecular recognition, as well as the spatial orientation of the electrophilic warhead, of various drugs, to achieve a covalent interaction with Cys145 of M^pro. The virtual screening on the structure-based pharmacophoric map of the SARS-CoV-2 M^pro in complex with an inhibitor N3 (reference compound) provided high efficiency by identifying 53 drugs (FDA and DrugBank databases) with probabilities of covalent binding, including N3 (Michael acceptor) and others with a variety of electrophilic warheads. Adding the energy contributions of affinity for non-covalent and covalent docking, 16 promising drugs were obtained. Our findings suggest that the FDA-approved drugs Vaborbactam, Cimetidine, Ixazomib, Scopolamine, and Bicalutamide, as well as the other investigational peptide-like drugs (DB04234, DB03456, DB07224, DB7252, and CMX-2043) are potential covalent inhibitors of SARS-CoV-2 M^pro.

Machine Learning Enables Accurate and Rapid Prediction of Active Molecules Against Breast Cancer Cells

Breast cancer (BC) has surpassed lung cancer as the most frequently occurring cancer, and it is the leading cause of cancer-related death in women. Therefore, there is an urgent need to discover or design new drug candidates for BC treatment. In this study, we first collected a series of structurally diverse datasets consisting of 33,757 active and 21,152 inactive compounds for 13 breast cancer cell lines and one normal breast cell line commonly used in in vitro antiproliferative assays. Predictive models were then developed using five conventional machine learning algorithms, including naïve Bayesian, support vector machine, k-Nearest Neighbors, random forest, and extreme gradient boosting, as well as five deep learning algorithms, including deep neural networks, graph convolutional networks, graph attention network, message passing neural networks, and Attentive FP. A total of 476 single models and 112 fusion models were constructed based on three types of molecular representations including molecular descriptors, fingerprints, and graphs. The evaluation results demonstrate that the best model for each BC cell subtype can achieve high predictive accuracy for the test sets with AUC values of 0.689–0.993. Moreover, important structural fragments related to BC cell inhibition were identified and interpreted. To facilitate the use of the model, an online webserver called ChemBC (http://chembc.idruglab.cn/) and its local version software (https://github.com/idruglab/ChemBC) were developed to predict whether compounds have potential inhibitory activity against BC cells.