Large-scale comparison of machine learning methods for drug target prediction on ChEMBL

Efficient Deep Model Ensemble Framework for Drug-Target Interaction Prediction

Accurate prediction of Drug-Target Interactions (DTI) is crucial for drug development. Current state-of-the-art deep learning methods have significantly advanced the field; however, these methods exhibit limitations in predictive performance and the propensity for false negatives. Therefore, we propose EADTN, a simple and efficient ensemble model. We have designed an innovative feature adaptation technique to automatically extract local weights of drugs and targets, and we utilize clustering-enhanced parameter fine-tuning to overcome the issue of false negatives, thereby enhancing its reliability in drug discovery. Based on EADTN, we also propose a Shapley value-based method for identifying key drug substructures, effectively enhancing the model's interpretability. Additionally, we utilized EADTN to reveal potential interactions between NQO1 targets and the drugs SIRT-IN-1 and LY2183240, which were subsequently validated through wet-lab experiments. Experimental evidence demonstrates that EADTN consistently outperforms existing best-performing models across various data sets, promising significant benefits in fields such as drug repositioning.

The Role of AI in Drug Discovery

The emergence of Artificial Intelligence (AI) in drug discovery marks a pivotal shift in pharmaceutical research, blending sophisticated computational techniques with conventional scientific exploration to break through enduring obstacles. This review paper elucidates the multifaceted applications of AI across various stages of drug development, highlighting significant advancements and methodologies. It delves into AI's instrumental role in drug design, polypharmacology, chemical synthesis, drug repurposing, and the prediction of drug properties such as toxicity, bioactivity, and physicochemical characteristics. Despite AI's promising advancements, the paper also addresses the challenges and limitations encountered in the field, including data quality, generalizability, computational demands, and ethical considerations. By offering a comprehensive overview of AI's role in drug discovery, this paper underscores the technology's potential to significantly enhance drug development, while also acknowledging the hurdles that must be overcome to fully realize its benefits.

Polypharmacology prediction: the long road toward comprehensively anticipating small-molecule selectivity to de-risk drug discovery

INTRODUCTION

Small molecules often bind to multiple targets, a behavior termed polypharmacology. Anticipating polypharmacology is essential for drug discovery since unknown off-targets can modulate safety and efficacy - profoundly affecting drug discovery success. Unfortunately, experimental methods to assess selectivity present significant limitations and drugs still fail in the clinic due to unanticipated off-targets. Computational methods are a cost-effective, complementary approach to predict polypharmacology.

AREAS COVERED

This review aims to provide a comprehensive overview of the state of polypharmacology prediction and discuss its strengths and limitations, covering both classical cheminformatics methods and bioinformatic approaches. The authors review available data sources, paying close attention to their different coverage. The authors then discuss major algorithms grouped by the types of data that they exploit using selected examples.

EXPERT OPINION

Polypharmacology prediction has made impressive progress over the last decades and contributed to identify many off-targets. However, data incompleteness currently limits most approaches to comprehensively predict selectivity. Moreover, our limited agreement on model assessment challenges the identification of the best algorithms - which at present show modest performance in prospective real-world applications. Despite these limitations, the exponential increase of multidisciplinary Big Data and AI hold much potential to better polypharmacology prediction and de-risk drug discovery.

Multi-Task ADME/PK prediction at industrial scale: leveraging large and diverse experimentaldatasets

ADME (Absorption, Distribution, Metabolism, Excretion) properties are key parameters to judge whether a drug candidate exhibits a desired pharmacokinetic (PK) profile. In this study, we tested multi-task machine learning (ML) models to predict ADME and animal PK endpoints trained on in-house data generated at Boehringer Ingelheim. Models were evaluated both at the design stage of a compound (i. e., no experimental data of test compounds available) and at testing stage when a particular assay would be conducted (i. e., experimental data of earlier conducted assays may be available). Using realistic time-splits, we found a clear benefit in performance of multi-task graph-based neural network models over single-task model, which was even stronger when experimental data of earlier assays is available. In an attempt to explain the success of multi-task models, we found that especially endpoints with the largest numbers of data points (physicochemical endpoints, clearance in microsomes) are responsible for increased predictivity in more complex ADME and PK endpoints. In summary, our study provides insight into how data for multiple ADME/PK endpoints in a pharmaceutical company can be best leveraged to optimize predictivity of ML models.

In silico off-target profiling for enhanced drug safety assessment

Ensuring drug safety in the early stages of drug development is crucial to avoid costly failures in subsequent phases. However, the economic burden associated with detecting drug off-targets and potential side effects through in vitro safety screening and animal testing is substantial. Drug off-target interactions, along with the adverse drug reactions they induce, are significant factors affecting drug safety. To assess the liability of candidate drugs, we developed an artificial intelligence model for the precise prediction of compound off-target interactions, leveraging multi-task graph neural networks. The outcomes of off-target predictions can serve as representations for compounds, enabling the differentiation of drugs under various ATC codes and the classification of compound toxicity. Furthermore, the predicted off-target profiles are employed in adverse drug reaction (ADR) enrichment analysis, facilitating the inference of potential ADRs for a drug. Using the withdrawn drug Pergolide as an example, we elucidate the mechanisms underlying ADRs at the target level, contributing to the exploration of the potential clinical relevance of newly predicted off-target interactions. Overall, our work facilitates the early assessment of compound safety/toxicity based on off-target identification, deduces potential ADRs of drugs, and ultimately promotes the secure development of drugs.

Prediction and Interpretation Microglia Cytotoxicity by Machine Learning

Ameliorating microglia-mediated neuroinflammation is a crucial strategy in developing new drugs for neurodegenerative diseases. Plant compounds are an important screening target for the discovery of drugs for the treatment of neurodegenerative diseases. However, due to the spatial complexity of phytochemicals, it becomes particularly important to evaluate the effectiveness of compounds while avoiding the mixing of cytotoxic substances in the early stages of compound screening. Traditional high-throughput screening methods suffer from high cost and low efficiency. A computational model based on machine learning provides a novel avenue for cytotoxicity determination. In this study, a microglia cytotoxicity classifier was developed using a machine learning approach. First, we proposed a data splitting strategy based on the molecule murcko generic scaffold, under this condition, three machine learning approaches were coupled with three kinds of molecular representation methods to construct microglia cytotoxicity classifier, which were then compared and assessed by the predictive accuracy, balanced accuracy, F1-score, and Matthews Correlation Coefficient. Then, the recursive feature elimination integrated with support vector machine (RFE-SVC) dimension reduction method was introduced to molecular fingerprints with high dimensions to further improve the model performance. Among all the microglial cytotoxicity classifiers, the SVM coupled with ECFP4 fingerprint after feature selection (ECFP4-RFE-SVM) obtained the most accurate classification for the test set (ACC of 0.99, BA of 0.99, F1-score of 0.99, MCC of 0.97). Finally, the Shapley additive explanations (SHAP) method was used in interpreting the microglia cytotoxicity classifier and key substructure smart identified as structural alerts. Experimental results show that ECFP4-RFE-SVM have reliable classification capability for microglia cytotoxicity, and SHAP can not only provide a rational explanation for microglia cytotoxicity predictions, but also offer a guideline for subsequent molecular cytotoxicity modifications.

Leveraging binding pose metadynamics to optimise target fishing predictions for three diverse ligands and their true targets

Computational target fishing plays an important role in target identification, particularly in drug discovery campaigns utilizing phenotypic screening. Numerous approaches exist to predict potential targets for a given ligand, but true targets may be inconsistently ranked. More advanced simulation methods may provide benefit in such cases by reranking these initial predictions. We evaluated the ability of binding pose metadynamics to improve the predicted rankings for three diverse ligands and their six true targets. Initial predictions using pharmacophore mapping showed no true targets ranked in the top 50 and two targets each ranked within the 50-100, 100-150, and 250-300 ranges respectively. Following binding pose metadynamics, ranking of true targets improved for four out of the six targets and included the highest ranked predictions overall, while rankings deteriorated for two targets. The revised rankings predicted two true targets ranked within the top 50, and one target each within the 50-100, 100-150, 150-200, and 200-250 ranges respectively. The findings of this study demonstrate that binding pose metadynamics may be of benefit in refining initial predictions from structure-based target fishing algorithms, thereby improving the efficiency of the target identification process in drug discovery efforts.

The Goldilocks paradigm: comparing classical machine learning, large language models, and few-shot learning for drug discovery applications

Recent advances in machine learning (ML) have led to newer model architectures including transformers (large language models, LLMs) showing state of the art results in text generation and image analysis as well as few-shot learning (FSLC) models which offer predictive power with extremely small datasets. These new architectures may offer promise, yet the 'no-free lunch' theorem suggests that no single model algorithm can outperform at all possible tasks. Here, we explore the capabilities of classical (SVR), FSLC, and transformer models (MolBART) over a range of dataset tasks and show a 'goldilocks zone' for each model type, in which dataset size and feature distribution (i.e. dataset "diversity") determines the optimal algorithm strategy. When datasets are small ( < 50 molecules), FSLC tend to outperform both classical ML and transformers. When datasets are small-to-medium sized (50-240 molecules) and diverse, transformers outperform both classical models and few-shot learning. Finally, when datasets are of larger and of sufficient size, classical models then perform the best, suggesting that the optimal model to choose likely depends on the dataset available, its size and diversity. These findings may help to answer the perennial question of which ML algorithm is to be used when faced with a new dataset.

One chiral fingerprint to find them all

Molecular fingerprints are indispensable tools in cheminformatics. However, stereochemistry is generally not considered, which is problematic for large molecules which are almost all chiral. Herein we report MAP4C, a chiral version of our previously reported fingerprint MAP4, which lists MinHashes computed from character strings containing the SMILES of all pairs of circular substructures up to a diameter of four bonds and the shortest topological distance between their central atoms. MAP4C includes the Cahn-Ingold-Prelog (CIP) annotation (R, S, r or s) whenever the chiral atom is the center of a circular substructure, a question mark for undefined stereocenters, and double bond cis-trans information if specified. MAP4C performs slightly better than the achiral MAP4, ECFP and AP fingerprints in non-stereoselective virtual screening benchmarks. Furthermore, MAP4C distinguishes between stereoisomers in chiral molecules from small molecule drugs to large natural products and peptides comprising thousands of diastereomers, with a degree of distinction smaller than between structural isomers and proportional to the number of chirality changes. Due to its excellent performance across diverse molecular classes and its ability to handle stereochemistry, MAP4C is recommended as a generally applicable chiral molecular fingerprint. SCIENTIFIC CONTRIBUTION: The ability of our chiral fingerprint MAP4C to handle stereoisomers from small molecules to large natural products and peptides is unprecedented and opens the way for cheminformatics to include stereochemistry as an important molecular parameter across all fields of molecular design.

MELLODDY: Cross-pharma Federated Learning at Unprecedented Scale Unlocks Benefits in QSAR without Compromising Proprietary Information

Federated multipartner machine learning has been touted as an appealing and efficient method to increase the effective training data volume and thereby the predictivity of models, particularly when the generation of training data is resource-intensive. In the landmark MELLODDY project, indeed, each of ten pharmaceutical companies realized aggregated improvements on its own classification or regression models through federated learning. To this end, they leveraged a novel implementation extending multitask learning across partners, on a platform audited for privacy and security. The experiments involved an unprecedented cross-pharma data set of 2.6+ billion confidential experimental activity data points, documenting 21+ million physical small molecules and 40+ thousand assays in on-target and secondary pharmacodynamics and pharmacokinetics. Appropriate complementary metrics were developed to evaluate the predictive performance in the federated setting. In addition to predictive performance increases in labeled space, the results point toward an extended applicability domain in federated learning. Increases in collective training data volume, including by means of auxiliary data resulting from single concentration high-throughput and imaging assays, continued to boost predictive performance, albeit with a saturating return. Markedly higher improvements were observed for the pharmacokinetics and safety panel assay-based task subsets.

HyperPCM: Robust Task-Conditioned Modeling of Drug-Target Interactions

A central problem in drug discovery is to identify the interactions between drug-like compounds and protein targets. Over the past few decades, various quantitative structure-activity relationship (QSAR) and proteo-chemometric (PCM) approaches have been developed to model and predict these interactions. While QSAR approaches solely utilize representations of the drug compound, PCM methods incorporate both representations of the protein target and the drug compound, enabling them to achieve above-chance predictive accuracy on previously unseen protein targets. Both QSAR and PCM approaches have recently been improved by machine learning and deep neural networks, that allow the development of drug-target interaction prediction models from measurement data. However, deep neural networks typically require large amounts of training data and cannot robustly adapt to new tasks, such as predicting interaction for unseen protein targets at inference time. In this work, we propose to use HyperNetworks to efficiently transfer information between tasks during inference and thus to accurately predict drug-target interactions on unseen protein targets. Our HyperPCM method reaches state-of-the-art performance compared to previous methods on multiple well-known benchmarks, including Davis, DUD-E, and a ChEMBL derived data set, and particularly excels at zero-shot inference involving unseen protein targets. Our method, as well as reproducible data preparation, is available at https://github.com/ml-jku/hyper-dti.

Compound Activity Prediction with Dose-Dependent Transcriptomic Profiles and Deep Learning

Predicting compound activity in assays is a long-standing challenge in drug discovery. Computational models based on compound-induced gene expression signatures from a single profiling assay have shown promise toward predicting compound activity in other, seemingly unrelated, assays. Applications of such models include predicting mechanisms-of-action (MoA) for phenotypic hits, identifying off-target activities, and identifying polypharmacologies. Here, we introduce transcriptomics-to-activity transformer (TAT) models that leverage gene expression profiles observed over compound treatment at multiple concentrations to predict the compound activity in other biochemical or cellular assays. We built TAT models based on gene expression data from a RASL-seq assay to predict the activity of 2692 compounds in 262 dose-response assays. We obtained useful models for 51% of the assays, as determined through a realistic held-out set. Prospectively, we experimentally validated the activity predictions of a TAT model in a malaria inhibition assay. With a 63% hit rate, TAT successfully identified several submicromolar malaria inhibitors. Our results thus demonstrate the potential of transcriptomic responses over compound concentration and the TAT modeling framework as a cost-efficient way to identify the bioactivities of promising compounds across many assays.

A bidirectional interpretable compound-protein interaction prediction framework based on cross attention

The identification of compound-protein interactions (CPIs) plays a vital role in drug discovery. However, the huge cost and labor-intensive nature in vitro and vivo experiments make it urgent for researchers to develop novel CPI prediction methods. Despite emerging deep learning methods have achieved promising performance in CPI prediction, they also face ongoing challenges: (i) providing bidirectional interpretability from both the chemical and biological perspective for the prediction results; (ii) comprehensively evaluating model generalization performance; (iii) demonstrating the practical applicability of these models. To overcome the challenges posed by current deep learning methods, we propose a cross multi-head attention oriented bidirectional interpretable CPI prediction model (CmhAttCPI). First, CmhAttCPI takes molecular graphs and protein sequences as inputs, utilizing the GCW module to learn atom features and the CNN module to learn residue features, respectively. Second, the model applies cross multi-head attention module to compute attention weights for atoms and residues. Finally, CmhAttCPI employs a fully connected neural network to predict scores for CPIs. We evaluated the performance of CmhAttCPI on balanced datasets and imbalanced datasets. The results consistently show that CmhAttCPI outperforms multiple state-of-the-art methods. We constructed three scenarios based on compound and protein clustering and comprehensively evaluated the model generalization ability within these scenarios. The results demonstrate that the generalization ability of CmhAttCPI surpasses that of other models. Besides, the visualizations of attention weights reveal that CmhAttCPI provides chemical and biological interpretation for CPI prediction. Moreover, case studies confirm the practical applicability of CmhAttCPI in discovering anticancer candidates.

Pre-training molecular representation model with spatial geometry for property prediction

AI-enhanced bioinformatics and cheminformatics pivots on generating increasingly descriptive and generalized molecular representation. Accurate prediction of molecular properties needs a comprehensive description of molecular geometry. We design a novel Graph Isomorphic Network (GIN) based model integrating a three-level network structure with a dual-level pre-training approach that aligns the characteristics of molecules. In our Spatial Molecular Pre-training (SMPT) Model, the network can learn implicit geometric information in layers from lower to higher according to the dimension. Extensive evaluations against established baseline models validate the enhanced efficacy of SMPT, with notable accomplishments in classification tasks. These results emphasize the importance of spatial geometric information in molecular representation modeling and demonstrate the potential of SMPT as a valuable tool for property prediction.

Towards safer and efficient formulations: Machine learning approaches to predict drug-excipient compatibility

Predicting drug-excipient compatibility is a critical aspect of pharmaceutical formulation design. In this study, we introduced an innovative approach that leverages machine learning techniques to improve the accuracy of drug-excipient compatibility predictions. Mol2vec and 2D molecular descriptors combined with the stacking technique were used to improve the performance of the model. This approach achieved a significant advancement in the predictive capacity as demonstrated by the accuracy, precision, recall, AUC, and MCC of 0.98, 0.87, 0.88, 0.93 and 0.86, respectively. Using the DE-INTERACT model as the benchmark, our stacking model could remarkably detect drug-excipient incompatibility in 10/12 tested cases, while DE-INTERACT managed to recognize only 3 out of 12 incompatibility cases in the validation experiments. To ensure user accessibility, the trained model was deployed to a user-friendly web platform (URL: https://decompatibility.streamlit.app/). This interactive interface accommodated inputs through various types, including names, PubChem CID, or SMILES strings. It promptly generated compatibility predictions alongside corresponding probability scores. However, the continual refinement of model performance is crucial before applying this model in practice.

Toward Unified AI Drug Discovery with Multimodal Knowledge

Background: In real-world drug discovery, human experts typically grasp molecular knowledge of drugs and proteins from multimodal sources including molecular structures, structured knowledge from knowledge bases, and unstructured knowledge from biomedical literature. Existing multimodal approaches in AI drug discovery integrate either structured or unstructured knowledge independently, which compromises the holistic understanding of biomolecules. Besides, they fail to address the missing modality problem, where multimodal information is missing for novel drugs and proteins. Methods: In this work, we present KEDD, a unified, end-to-end deep learning framework that jointly incorporates both structured and unstructured knowledge for vast AI drug discovery tasks. The framework first incorporates independent representation learning models to extract the underlying characteristics from each modality. Then, it applies a feature fusion technique to calculate the prediction results. To mitigate the missing modality problem, we leverage sparse attention and a modality masking technique to reconstruct the missing features based on top relevant molecules. Results: Benefiting from structured and unstructured knowledge, our framework achieves a deeper understanding of biomolecules. KEDD outperforms state-of-the-art models by an average of 5.2% on drug-target interaction prediction, 2.6% on drug property prediction, 1.2% on drug-drug interaction prediction, and 4.1% on protein-protein interaction prediction. Through qualitative analysis, we reveal KEDD's promising potential in assisting real-world applications. Conclusions: By incorporating biomolecular expertise from multimodal knowledge, KEDD bears promise in accelerating drug discovery.

Learning long- and short-term dependencies for improving drug-target binding affinity prediction using transformer and edge contraction pooling

The accurate identification of drug-target affinity (DTA) is crucial for advancements in drug discovery and development. Many deep learning-based approaches have been devised to predict drug-target binding affinity accurately, exhibiting notable improvements in performance. However, the existing prediction methods often fall short of capturing the global features of proteins. In this study, we proposed a novel model called ETransDTA, specifically designed for predicting drug-target binding affinity. ETransDTA combines convolutional layers and transformer, allowing for the simultaneous extraction of both global and local features of target proteins. Additionally, we have integrated a new graph pooling mechanism into the topology adaptive graph convolutional network (TAGCN) to enhance its capacity for learning feature representations of chemical compounds. The proposed ETransDTA model has been evaluated using the Davis and Kinase Inhibitor BioActivity (KIBA) datasets, consistently outperforming other baseline methods. The evaluation results on the KIBA dataset reveal that our model achieves the lowest mean square error (MSE) of 0.125, representing a 0.6% reduction compared to the lowest-performing baseline method. Furthermore, the incorporation of queries, keys and values produced by the stacked convolutional neural network (CNN) enables our model to better integrate the local and global context of protein representation, leading to further improvements in the accuracy of DTA prediction.

Characterizing emerging companies in computational drug development

Computation promises to accelerate, de-risk and optimize drug research and development. An increasing number of companies have entered this space, specializing in the design of new algorithms, computing on proprietary data, and/or development of hardware to improve distinct drug pipeline stages. The large number of such companies and their unique strategies and deals have created a highly complex and competitive industry. We comprehensively analyze the companies in this space to highlight trends and opportunities, identifying highly occupied areas of risk and currently underrepresented niches of high value.

The ChEMBL Database in 2023: a drug discovery platform spanning multiple bioactivity data types and time periods

ChEMBL (https://www.ebi.ac.uk/chembl/) is a manually curated, high-quality, large-scale, open, FAIR and Global Core Biodata Resource of bioactive molecules with drug-like properties, previously described in the 2012, 2014, 2017 and 2019 Nucleic Acids Research Database Issues. Since its introduction in 2009, ChEMBL's content has changed dramatically in size and diversity of data types. Through incorporation of multiple new datasets from depositors since the 2019 update, ChEMBL now contains slightly more bioactivity data from deposited data vs data extracted from literature. In collaboration with the EUbOPEN consortium, chemical probe data is now regularly deposited into ChEMBL. Release 27 made curated data available for compounds screened for potential anti-SARS-CoV-2 activity from several large-scale drug repurposing screens. In addition, new patent bioactivity data have been added to the latest ChEMBL releases, and various new features have been incorporated, including a Natural Product likeness score, updated flags for Natural Products, a new flag for Chemical Probes, and the initial annotation of the action type for ∼270 000 bioactivity measurements.

Relative molecule self-attention transformer

The prediction of molecular properties is a crucial aspect in drug discovery that can save a lot of money and time during the drug design process. The use of machine learning methods to predict molecular properties has become increasingly popular in recent years. Despite advancements in the field, several challenges remain that need to be addressed, like finding an optimal pre-training procedure to improve performance on small datasets, which are common in drug discovery. In our paper, we tackle these problems by introducing Relative Molecule Self-Attention Transformer for molecular representation learning. It is a novel architecture that uses relative self-attention and 3D molecular representation to capture the interactions between atoms and bonds that enrich the backbone model with domain-specific inductive biases. Furthermore, our two-step pretraining procedure allows us to tune only a few hyperparameter values to achieve good performance comparable with state-of-the-art models on a wide selection of downstream tasks.

1H-NMR metabolomics analysis identifies hypoxanthine as a novel metastasis-associated metabolite in breast cancer

Breast cancer is one of the leading causes of death in females, mainly because of metastasis. Oncometabolites, produced via metabolic reprogramming, can influence metastatic signaling cascades. Accordingly, and based on our previous results, we propose that metabolites from highly metastatic breast cancer cells behave differently from less-metastatic cells and may play a significant role in metastasis. For instance, we aim to identify these metabolites and their role in breast cancer metastasis. Less metastatic cells (MCF-7) were treated with metabolites secreted from highly metastatic cells (MDA-MB-231) and the gene expression of three epithelial-to-mesenchymal transition (EMT) markers including E-cadherin, N-cadherin and vimentin were examined. Some metabolites secreted from MDA-MB-231 cells significantly induced EMT activity. Specifically, hypoxanthine demonstrated a significant EMT effect and increased the migration and invasion effects of MCF-7 cells through a hypoxia-associated mechanism. Hypoxanthine exhibited pro-angiogenic effects via increasing the VEGF and PDGF gene expression and affected lipid metabolism by increasing the gene expression of PCSK-9. Notably, knockdown of purine nucleoside phosphorylase, a gene encoding for an important enzyme in the biosynthesis of hypoxanthine, and inhibition of hypoxanthine uptake caused a significant decrease in hypoxanthine-associated EMT effects. Collectively for the first time, hypoxanthine was identified as a novel metastasis-associated metabolite in breast cancer cells and represents a promising target for diagnosis and therapy.

Multimodal multi-task deep neural network framework for kinase-target prediction

Kinase plays a significant role in various disease signaling pathways. Due to the highly conserved sequence of kinase family members, understanding the selectivity profile of kinase inhibitors remains a priority for drug discovery. Previous methods for kinase selectivity identification use biochemical assays, which are very useful but limited by the protein available. The lack of kinase selectivity can exert benefits but also can cause adverse effects. With the explosion of the dataset for kinase activities, current computational methods can achieve accuracy for large-scale selectivity predictions. Here, we present a multimodal multi-task deep neural network model for kinase selectivity prediction by calculating the fingerprint and physiochemical descriptors. With the multimodal inputs of structure and physiochemical properties information, the multi-task framework could accurately predict the kinome map for selectivity analysis. The proposed model displays better performance for kinase-target prediction based on system evaluations.

Drug-target affinity prediction method based on multi-scale information interaction and graph optimization

Drug-target affinity (DTA) prediction as an emerging and effective method is widely applied to explore the strength of drug-target interactions in drug development research. By predicting these interactions, researchers can assess the potential efficacy and safety of candidate drugs at an early stage, narrowing down the search space for therapeutic targets and accelerating the discovery and development of new drugs. However, existing DTA prediction models mainly use graphical representations of drug molecules, which lack information on interactions between individual substructures, thus affecting prediction accuracy and model interpretability. Therefore, transformer and diffusion on drug graphs in DTA prediction (TDGraphDTA) are introduced to predict drug-target interactions using multi-scale information interaction and graph optimization. An interactive module is integrated into feature extraction of drug and target features at different granularity levels. A diffusion model-based graph optimization module is proposed to improve the representation of molecular graph structures and enhance the interpretability of graph representations while obtaining optimal feature representations. In addition, TDGraphDTA improves the accuracy and reliability of predictions by capturing relationships and contextual information between molecular substructures. The performance of the proposed TDGraphDTA in DTA prediction was verified on three publicly available benchmark datasets (Davis, Metz, and KIBA). Compared with state-of-the-art baseline models, it achieved better results in terms of consistency index, R-squared, etc. Furthermore, compared with some existing methods, the proposed TDGraphDTA is demonstrated to have better structure capturing capabilities by visualizing the feature capturing capabilities of the model using Grad-AAM toxicity labels in the ToxCast dataset. The corresponding source codes are available at https://github.com/Lamouryz/TDGraph.

MulinforCPI: enhancing precision of compound-protein interaction prediction through novel perspectives on multi-level information integration

Forecasting the interaction between compounds and proteins is crucial for discovering new drugs. However, previous sequence-based studies have not utilized three-dimensional (3D) information on compounds and proteins, such as atom coordinates and distance matrices, to predict binding affinity. Furthermore, numerous widely adopted computational techniques have relied on sequences of amino acid characters for protein representations. This approach may constrain the model's ability to capture meaningful biochemical features, impeding a more comprehensive understanding of the underlying proteins. Here, we propose a two-step deep learning strategy named MulinforCPI that incorporates transfer learning techniques with multi-level resolution features to overcome these limitations. Our approach leverages 3D information from both proteins and compounds and acquires a profound understanding of the atomic-level features of proteins. Besides, our research highlights the divide between first-principle and data-driven methods, offering new research prospects for compound-protein interaction tasks. We applied the proposed method to six datasets: Davis, Metz, KIBA, CASF-2016, DUD-E and BindingDB, to evaluate the effectiveness of our approach.

Advancing Computational Toxicology by Interpretable Machine Learning

Chemical toxicity evaluations for drugs, consumer products, and environmental chemicals have a critical impact on human health. Traditional animal models to evaluate chemical toxicity are expensive, time-consuming, and often fail to detect toxicants in humans. Computational toxicology is a promising alternative approach that utilizes machine learning (ML) and deep learning (DL) techniques to predict the toxicity potentials of chemicals. Although the applications of ML- and DL-based computational models in chemical toxicity predictions are attractive, many toxicity models are "black boxes" in nature and difficult to interpret by toxicologists, which hampers the chemical risk assessments using these models. The recent progress of interpretable ML (IML) in the computer science field meets this urgent need to unveil the underlying toxicity mechanisms and elucidate the domain knowledge of toxicity models. In this review, we focused on the applications of IML in computational toxicology, including toxicity feature data, model interpretation methods, use of knowledge base frameworks in IML development, and recent applications. The challenges and future directions of IML modeling in toxicology are also discussed. We hope this review can encourage efforts in developing interpretable models with new IML algorithms that can assist new chemical assessments by illustrating toxicity mechanisms in humans.

A knowledge-guided pre-training framework for improving molecular representation learning

Learning effective molecular feature representation to facilitate molecular property prediction is of great significance for drug discovery. Recently, there has been a surge of interest in pre-training graph neural networks (GNNs) via self-supervised learning techniques to overcome the challenge of data scarcity in molecular property prediction. However, current self-supervised learning-based methods suffer from two main obstacles: the lack of a well-defined self-supervised learning strategy and the limited capacity of GNNs. Here, we propose Knowledge-guided Pre-training of Graph Transformer (KPGT), a self-supervised learning framework to alleviate the aforementioned issues and provide generalizable and robust molecular representations. The KPGT framework integrates a graph transformer specifically designed for molecular graphs and a knowledge-guided pre-training strategy, to fully capture both structural and semantic knowledge of molecules. Through extensive computational tests on 63 datasets, KPGT exhibits superior performance in predicting molecular properties across various domains. Moreover, the practical applicability of KPGT in drug discovery has been validated by identifying potential inhibitors of two antitumor targets: hematopoietic progenitor kinase 1 (HPK1) and fibroblast growth factor receptor 1 (FGFR1). Overall, KPGT can provide a powerful and useful tool for advancing the artificial intelligence (AI)-aided drug discovery process.

Deep Learning with Geometry-Enhanced Molecular Representation for Augmentation of Large-Scale Docking-Based Virtual Screening

Structure-based virtual screening has been a crucial tool in drug discovery for decades. However, as the chemical space expands, the existing structure-based virtual screening techniques based on molecular docking and scoring struggle to handle billion-entry ultralarge libraries due to the high computational cost. To address this challenge, people have resorted to machine learning techniques to enhance structure-based virtual screening for efficiently exploring the vast chemical space. In those cases, compounds are usually treated as sequential strings or two-dimensional topology graphs, limiting their ability to incorporate three-dimensional structural information for downstream tasks. We herein propose a novel deep learning protocol, GEM-Screen, which utilizes the geometry-enhanced molecular representation of the compounds docking to a specific target and is trained on docking scores of a small fraction of a library through an active learning strategy to approximate the docking outcome for yet nontraining entries. This protocol is applied to virtual screening campaigns against the AmpC and D4 targets, demonstrating that GEM-Screen enriches more than 90% of the hit scaffolds for AmpC in the top 4% of model predictions and more than 80% of the hit scaffolds for D4 in the same top-ranking size of library. GEM-Screen can be used in conjunction with traditional docking programs for docking of only the top-ranked compounds to avoid the exhaustive docking of the whole library, thus allowing for discovering top-scoring compounds from billion-entry libraries in a rapid yet accurate fashion.

CLOOME: contrastive learning unlocks bioimaging databases for queries with chemical structures

The field of bioimage analysis is currently impacted by a profound transformation, driven by the advancements in imaging technologies and artificial intelligence. The emergence of multi-modal AI systems could allow extracting and utilizing knowledge from bioimaging databases based on information from other data modalities. We leverage the multi-modal contrastive learning paradigm, which enables the embedding of both bioimages and chemical structures into a unified space by means of bioimage and molecular structure encoders. This common embedding space unlocks the possibility of querying bioimaging databases with chemical structures that induce different phenotypic effects. Concretely, in this work we show that a retrieval system based on multi-modal contrastive learning is capable of identifying the correct bioimage corresponding to a given chemical structure from a database of ~2000 candidate images with a top-1 accuracy >70 times higher than a random baseline. Additionally, the bioimage encoder demonstrates remarkable transferability to various further prediction tasks within the domain of drug discovery, such as activity prediction, molecule classification, and mechanism of action identification. Thus, our approach not only addresses the current limitations of bioimaging databases but also paves the way towards foundation models for microscopy images.

Evaluating uncertainty-based active learning for accelerating the generalization of molecular property prediction

Deep learning models have proven to be a powerful tool for the prediction of molecular properties for applications including drug design and the development of energy storage materials. However, in order to learn accurate and robust structure-property mappings, these models require large amounts of data which can be a challenge to collect given the time and resource-intensive nature of experimental material characterization efforts. Additionally, such models fail to generalize to new types of molecular structures that were not included in the model training data. The acceleration of material development through uncertainty-guided experimental design has the promise to significantly reduce the data requirements and enable faster generalization to new types of materials. To evaluate the potential of such approaches for electrolyte design applications, we perform comprehensive evaluation of existing uncertainty quantification methods on the prediction of two relevant molecular properties - aqueous solubility and redox potential. We develop novel evaluation methods to probe the utility of the uncertainty estimates for both in-domain and out-of-domain data sets. Finally, we leverage selected uncertainty estimation methods for active learning to evaluate their capacity to support experimental design.

Identification of Drug Compounds for Capsular Contracture Based on Text Mining and Deep Learning

BACKGROUND

Capsular contracture is a common and unpredictable complication after breast implant placement. Currently, the pathogenesis of capsular contracture is unclear, and the effectiveness of nonsurgical treatment is still doubtful. The authors' study aimed to investigate new drug therapies for capsular contracture by using computational methods.

METHODS

Genes related to capsular contracture were identified by text mining and GeneCodis. Then, the candidate key genes were selected through protein-protein interaction analysis in Search Tool for the Retrieval of Interacting Genes/Proteins and Cytoscape. Drugs targeting the candidate genes with relation to capsular contracture were screened out in Pharmaprojects. Based on the drug-target interaction analysis by DeepPurpose, candidate drugs with highest predicted binding affinity were obtained eventually.

RESULTS

The authors' study identified 55 genes related to capsular contracture. Gene set enrichment analysis and protein-protein interaction analysis generated eight candidate genes. One hundred drugs targeting the candidate genes were selected. The seven candidate drugs with the highest predicted binding affinity were determined by DeepPurpose, including tumor necrosis factor alpha antagonist, estrogen receptor agonist, insulin-like growth factor 1 receptor, tyrosine kinase inhibitor, and matrix metallopeptidase 1 inhibitor.

CONCLUSION

Text mining and DeepPurpose can be used as a promising tool for drug discovery in exploring nonsurgical treatment to capsular contracture.

CLINICAL QUESTION/LEVEL OF EVIDENCE

Therapeutic, V.

A Computational Workflow to Predict Biological Target Mutations: The Spike Glycoprotein Case Study

The biological target identification process, a pivotal phase in the drug discovery workflow, becomes particularly challenging when mutations affect proteins' mechanisms of action. COVID-19 Spike glycoprotein mutations are known to modify the affinity toward the human angiotensin-converting enzyme ACE2 and several antibodies, compromising their neutralizing effect. Predicting new possible mutations would be an efficient way to develop specific and efficacious drugs, vaccines, and antibodies. In this work, we developed and applied a computational procedure, combining constrained logic programming and careful structural analysis based on the Structural Activity Relationship (SAR) approach, to predict and determine the structure and behavior of new future mutants. "Mutations rules" that would track statistical and functional types of substitutions for each residue or combination of residues were extracted from the GISAID database and used to define constraints for our software, having control of the process step by step. A careful molecular dynamics analysis of the predicted mutated structures was carried out after an energy evaluation of the intermolecular and intramolecular interactions using the HINT (Hydrophatic INTeraction) force field. Our approach successfully predicted, among others, known Spike mutants.

Analysis of metabolites in human gut: illuminating the design of gut-targeted drugs

Gut-targeted drugs provide a new drug modality besides that of oral, systemic molecules, that could tap into the growing knowledge of gut metabolites of bacterial or host origin and their involvement in biological processes and health through their interaction with gut targets (bacterial or host, too). Understanding the properties of gut metabolites can provide guidance for the design of gut-targeted drugs. In the present work we analyze a large set of gut metabolites, both shared with serum or present only in gut, and compare them with oral systemic drugs. We find patterns specific for these two subsets of metabolites that could be used to design drugs targeting the gut. In addition, we develop and openly share a Super Learner model to predict gut permanence, in order to aid in the design of molecules with appropriate profiles to remain in the gut, resulting in molecules with putatively reduced secondary effects and better pharmacokinetics.

A systematic study of key elements underlying molecular property prediction

Artificial intelligence (AI) has been widely applied in drug discovery with a major task as molecular property prediction. Despite booming techniques in molecular representation learning, key elements underlying molecular property prediction remain largely unexplored, which impedes further advancements in this field. Herein, we conduct an extensive evaluation of representative models using various representations on the MoleculeNet datasets, a suite of opioids-related datasets and two additional activity datasets from the literature. To investigate the predictive power in low-data and high-data space, a series of descriptors datasets of varying sizes are also assembled to evaluate the models. In total, we have trained 62,820 models, including 50,220 models on fixed representations, 4200 models on SMILES sequences and 8400 models on molecular graphs. Based on extensive experimentation and rigorous comparison, we show that representation learning models exhibit limited performance in molecular property prediction in most datasets. Besides, multiple key elements underlying molecular property prediction can affect the evaluation results. Furthermore, we show that activity cliffs can significantly impact model prediction. Finally, we explore into potential causes why representation learning models can fail and show that dataset size is essential for representation learning models to excel.

Recent Applications of Machine Learning in Molecular Property and Chemical Reaction Outcome Predictions

Burgeoning developments in machine learning (ML) and its rapidly growing adaptations in chemistry are noteworthy. Motivated by the successful deployments of ML in the realm of molecular property prediction (MPP) and chemical reaction prediction (CRP), herein we highlight some of its most recent applications in predictive chemistry. We present a nonmathematical and concise overview of the progression of ML implementations, ranging from an ensemble-based random forest model to advanced graph neural network algorithms. Similarly, the prospects of various feature engineering and feature learning approaches that work in conjunction with ML models are described. Highly accurate predictions reported in MPP tasks (e.g., lipophilicity, solubility, distribution coefficient), using methods such as D-MPNN, MolCLR, SMILES-BERT, and MolBERT, offer promising avenues in molecular design and drug discovery. Whereas MPP pertains to a given molecule, ML applications in chemical reactions present a different level of challenge, primarily arising from the simultaneous involvement of multiple molecules and their diverse roles in a reaction setting. The reported RMSEs in MPP tasks range from 0.287 to 2.20, while those for yield predictions are well over 4.9 in the lower end, reaching thresholds of >10.0 in several examples. Our Review concludes with a set of persisting challenges in dealing with reaction data sets and an overall optimistic outlook on benefits of ML-driven workflows for various MPP as well as CRP tasks.

Retrosynthesis prediction with an interpretable deep-learning framework based on molecular assembly tasks

Automating retrosynthesis with artificial intelligence expedites organic chemistry research in digital laboratories. However, most existing deep-learning approaches are hard to explain, like a "black box" with few insights. Here, we propose RetroExplainer, formulizing the retrosynthesis task into a molecular assembly process, containing several retrosynthetic actions guided by deep learning. To guarantee a robust performance of our model, we propose three units: a multi-sense and multi-scale Graph Transformer, structure-aware contrastive learning, and dynamic adaptive multi-task learning. The results on 12 large-scale benchmark datasets demonstrate the effectiveness of RetroExplainer, which outperforms the state-of-the-art single-step retrosynthesis approaches. In addition, the molecular assembly process renders our model with good interpretability, allowing for transparent decision-making and quantitative attribution. When extended to multi-step retrosynthesis planning, RetroExplainer has identified 101 pathways, in which 86.9% of the single reactions correspond to those already reported in the literature. As a result, RetroExplainer is expected to offer valuable insights for reliable, high-throughput, and high-quality organic synthesis in drug development.

Predicting adsorption of organic compounds onto graphene and black phosphorus by molecular dynamics and machine learning

With an increase in production and application of various engineering nanomaterials (ENMs), they will inevitably be released into the environment. Adsorption of various organic chemicals onto ENMs will impact on their environmental behavior and toxicology. It is unrealistic to experimentally determine adsorption equilibrium constants (K) for the vast number of organics and ENMs due to high cost in expenditure and time. Herein, appropriate molecular dynamics (MD) methods were evaluated and selected by comparing experimental K values of seven organics adsorbed onto graphene with the MD-calculated ones. Machine learning (ML) models on K of organics adsorption onto graphene and black phosphorus nanomaterials were constructed based on a benchmark data set from the MD simulations. Lasso models based on Mordred descriptors outperformed ML models built by support vector machine, random forest, k-nearest neighbor, and gradient boosting decision tree, in terms of cross-validation coefficients (Q2 > 0.90). The Lasso models also outperformed conventional poly-parameter linear free energy relationship models for predicting logK. Compared with previous models, the Lasso models considered more compounds with different functional groups and thus have broader applicability domains. This study provides a promising way to fill the data gap in logK for chemicals adsorbed onto the ENMs.

Extended study on atomic featurization in graph neural networks for molecular property prediction

Graph neural networks have recently become a standard method for analyzing chemical compounds. In the field of molecular property prediction, the emphasis is now on designing new model architectures, and the importance of atom featurization is oftentimes belittled. When contrasting two graph neural networks, the use of different representations possibly leads to incorrect attribution of the results solely to the network architecture. To better understand this issue, we compare multiple atom representations by evaluating them on the prediction of free energy, solubility, and metabolic stability using graph convolutional networks. We discover that the choice of atom representation has a significant impact on model performance and that the optimal subset of features is task-specific. Additional experiments involving more sophisticated architectures, including graph transformers, support these findings. Moreover, we demonstrate that some commonly used atom features, such as the number of neighbors or the number of hydrogens, can be easily predicted using only information about bonds and atom type, yet their explicit inclusion in the representation has a positive impact on model performance. Finally, we explain the predictions of the best-performing models to better understand how they utilize the available atomic features.

MAVEN: compound mechanism of action analysis and visualisation using transcriptomics and compound structure data in R/Shiny

BACKGROUND

Understanding the Mechanism of Action (MoA) of a compound is an often challenging but equally crucial aspect of drug discovery that can help improve both its efficacy and safety. Computational methods to aid MoA elucidation usually either aim to predict direct drug targets, or attempt to understand modulated downstream pathways or signalling proteins. Such methods usually require extensive coding experience and results are often optimised for further computational processing, making them difficult for wet-lab scientists to perform, interpret and draw hypotheses from.

RESULTS

To address this issue, we in this work present MAVEN (Mechanism of Action Visualisation and Enrichment), an R/Shiny app which allows for GUI-based prediction of drug targets based on chemical structure, combined with causal reasoning based on causal protein-protein interactions and transcriptomic perturbation signatures. The app computes a systems-level view of the mechanism of action of the input compound. This is visualised as a sub-network linking predicted or known targets to modulated transcription factors via inferred signalling proteins. The tool includes a selection of MSigDB gene set collections to perform pathway enrichment on the resulting network, and also allows for custom gene sets to be uploaded by the researcher. MAVEN is hence a user-friendly, flexible tool for researchers without extensive bioinformatics or cheminformatics knowledge to generate interpretable hypotheses of compound Mechanism of Action.

CONCLUSIONS

MAVEN is available as a fully open-source tool at https://github.com/laylagerami/MAVEN with options to install in a Docker or Singularity container. Full documentation, including a tutorial on example data, is available at https://laylagerami.github.io/MAVEN .

Overfit deep neural network for predicting drug-target interactions

Drug-target interactions (DTIs) prediction is an important step in drug discovery. As traditional biological experiments or high-throughput screening are high cost and time-consuming, many deep learning models have been developed. Overfitting must be avoided when training deep learning models. We propose a simple framework, called OverfitDTI, for DTI prediction. In OverfitDTI, a deep neural network (DNN) model is overfit to sufficiently learn the features of the chemical space of drugs and the biological space of targets. The weights of trained DNN model form an implicit representation of the nonlinear relationship between drugs and targets. Performance of OverfitDTI on three public datasets showed that the overfit DNN models fit the nonlinear relationship with high accuracy. We identified fifteen compounds that interacted with TEK, a receptor tyrosine kinase contributing to vascular homeostasis, and the predicted AT9283 and dorsomorphin were experimentally demonstrated as inhibitors of TEK in human umbilical vein endothelial cells (HUVECs).

Evaluating Scalable Supervised Learning for Synthesize-on-Demand Chemical Libraries

Traditional small-molecule drug discovery is a time-consuming and costly endeavor. High-throughput chemical screening can only assess a tiny fraction of drug-like chemical space. The strong predictive power of modern machine-learning methods for virtual chemical screening enables training models on known active and inactive compounds and extrapolating to much larger chemical libraries. However, there has been limited experimental validation of these methods in practical applications on large commercially available or synthesize-on-demand chemical libraries. Through a prospective evaluation with the bacterial protein-protein interaction PriA-SSB, we demonstrate that ligand-based virtual screening can identify many active compounds in large commercial libraries. We use cross-validation to compare different types of supervised learning models and select a random forest (RF) classifier as the best model for this target. When predicting the activity of more than 8 million compounds from Aldrich Market Select, the RF substantially outperforms a naïve baseline based on chemical structure similarity. 48% of the RF's 701 selected compounds are active. The RF model easily scales to score one billion compounds from the synthesize-on-demand Enamine REAL database. We tested 68 chemically diverse top predictions from Enamine REAL and observed 31 hits (46%), including one with an IC50 value of 1.3 μM.

Poor Generalization by Current Deep Learning Models for Predicting Binding Affinities of Kinase Inhibitors

The extreme surge of interest over the past decade surrounding the use of neural networks has inspired many groups to deploy them for predicting binding affinities of drug-like molecules to their receptors. A model that can accurately make such predictions has the potential to screen large chemical libraries and help streamline the drug discovery process. However, despite reports of models that accurately predict quantitative inhibition using protein kinase sequences and inhibitors' SMILES strings, it is still unclear whether these models can generalize to previously unseen data. Here, we build a Convolutional Neural Network (CNN) analogous to those previously reported and evaluate the model over four datasets commonly used for inhibitor/kinase predictions. We find that the model performs comparably to those previously reported, provided that the individual data points are randomly split between the training set and the test set. However, model performance is dramatically deteriorated when all data for a given inhibitor is placed together in the same training/testing fold, implying that information leakage underlies the models' performance. Through comparison to simple models in which the SMILES strings are tokenized, or in which test set predictions are simply copied from the closest training set data points, we demonstrate that there is essentially no generalization whatsoever in this model. In other words, the model has not learned anything about molecular interactions, and does not provide any benefit over much simpler and more transparent models. These observations strongly point to the need for richer structure-based encodings, to obtain useful prospective predictions of not-yet-synthesized candidate inhibitors.

Practical guidelines for the use of gradient boosting for molecular property prediction

Decision tree ensembles are among the most robust, high-performing and computationally efficient machine learning approaches for quantitative structure-activity relationship (QSAR) modeling. Among them, gradient boosting has recently garnered particular attention, for its performance in data science competitions, virtual screening campaigns, and bioactivity prediction. However, different variants of gradient boosting exist, the most popular being XGBoost, LightGBM and CatBoost. Our study provides the first comprehensive comparison of these approaches for QSAR. To this end, we trained 157,590 gradient boosting models, which were evaluated on 16 datasets and 94 endpoints, comprising 1.4 million compounds in total. Our results show that XGBoost generally achieves the best predictive performance, while LightGBM requires the least training time, especially for larger datasets. In terms of feature importance, the models surprisingly rank molecular features differently, reflecting differences in regularization techniques and decision tree structures. Thus, expert knowledge must always be employed when evaluating data-driven explanations of bioactivity. Furthermore, our results show that the relevance of each hyperparameter varies greatly across datasets and that it is crucial to optimize as many hyperparameters as possible to maximize the predictive performance. In conclusion, our study provides the first set of guidelines for cheminformatics practitioners to effectively train, optimize and evaluate gradient boosting models for virtual screening and QSAR applications.

FP-MAP: an extensive library of fingerprint-based molecular activity prediction tools

Discovering new drugs for disease treatment is challenging, requiring a multidisciplinary effort as well as time, and resources. With a view to improving hit discovery and lead compound identification, machine learning (ML) approaches are being increasingly used in the decision-making process. Although a number of ML-based studies have been published, most studies only report fragments of the wider range of bioactivities wherein each model typically focuses on a particular disease. This study introduces FP-MAP, an extensive atlas of fingerprint-based prediction models that covers a diverse range of activities including neglected tropical diseases (caused by viral, bacterial and parasitic pathogens) as well as other targets implicated in diseases such as Alzheimer's. To arrive at the best predictive models, performance of ≈4,000 classification/regression models were evaluated on different bioactivity data sets using 12 different molecular fingerprints. The best performing models that achieved test set AUC values of 0.62-0.99 have been integrated into an easy-to-use graphical user interface that can be downloaded from https://gitlab.com/vishsoft/fpmap.

3D graph neural network with few-shot learning for predicting drug-drug interactions in scaffold-based cold start scenario

Understanding drug-drug interactions (DDI) of new drugs is critical for minimizing unexpected adverse drug reactions. The modeling of new drugs is called a cold start scenario. In this scenario, Only a few structural information or physicochemical information about new drug is available. The 3D conformation of drug molecules usually plays a crucial role in chemical properties compared to the 2D structure. 3D graph network with few-shot learning is a promising solution. However, the 3D heterogeneity of drug molecules and the discretization of atomic distributions lead to spatial confusion in few-shot learning. Here, we propose a 3D graph neural network with few-shot learning, Meta3D-DDI, to predict DDI events in cold start scenario. The 3DGNN ensures rotation and translation invariance by calculating atomic pairwise distances, and incorporates 3D structure and distance information in the information aggregation stage. The continuous filter interaction module can continuously simulate the filter to obtain the interaction between the target atom and other atoms. Meta3D-DDI further develops a FSL strategy based on bilevel optimization to transfer meta-knowledge for DDI prediction tasks from existing drugs to new drugs. In addition, the existing cold start setting may cause the scaffold structure information in the training set to leak into the test set. We design scaffold-based cold start scenario to ensure that the drug scaffolds in the training set and test set do not overlap. The extensive experiments demonstrate that our architecture achieves the SOTA performance for DDI prediction under scaffold-based cold start scenario on two real-world datasets. The visual experiment shows that Meta3D-DDI significantly improves the learning for DDI prediction of new drugs. We also demonstrate how Meta3D-DDI can reduce the amount of data required to make meaningful DDI predictions.

Application of message passing neural networks for molecular property prediction

Accurate molecular property prediction, as one of the classical cheminformatics topics, plays a prominent role in the fields of computer-aided drug design. For instance, property prediction models can be used to quickly screen large molecular libraries to find lead compounds. Message-passing neural networks (MPNNs), a sub-class of Graph neural networks (GNNs), have recently been demonstrated to outperform other deep learning methods on a variety of tasks, including the prediction of molecular characteristics. In this survey, we provide a brief review of the MPNN models and their applications on molecular property prediction.

Discovering NDM-1 inhibitors using molecular substructure embeddings representations

NDM-1 (New-Delhi-Metallo-β-lactamase-1) is an enzyme developed by bacteria that is implicated in bacteria resistance to almost all known antibiotics. In this study, we deliver a new, curated NDM-1 bioactivities database, along with a set of unifying rules for managing different activity properties and inconsistencies. We define the activity classification problem in terms of Multiple Instance Learning, employing embeddings corresponding to molecular substructures and present an ensemble ranking and classification framework, relaying on a k-fold Cross Validation method employing a per fold hyper-parameter optimization procedure, showing promising generalization ability. The MIL paradigm displayed an improvement up to 45.7 %, in terms of Balanced Accuracy, in comparison to the classical Machine Learning paradigm. Moreover, we investigate different compact molecular representations, based on atomic or bi-atomic substructures. Finally, we scanned the Drugbank for strongly active compounds and we present the top-15 ranked compounds.

Discovery of senolytics using machine learning

Cellular senescence is a stress response involved in ageing and diverse disease processes including cancer, type-2 diabetes, osteoarthritis and viral infection. Despite growing interest in targeted elimination of senescent cells, only few senolytics are known due to the lack of well-characterised molecular targets. Here, we report the discovery of three senolytics using cost-effective machine learning algorithms trained solely on published data. We computationally screened various chemical libraries and validated the senolytic action of ginkgetin, periplocin and oleandrin in human cell lines under various modalities of senescence. The compounds have potency comparable to known senolytics, and we show that oleandrin has improved potency over its target as compared to best-in-class alternatives. Our approach led to several hundred-fold reduction in drug screening costs and demonstrates that artificial intelligence can take maximum advantage of small and heterogeneous drug screening data, paving the way for new open science approaches to early-stage drug discovery.

ProfhEX: AI-based platform for small molecules liability profiling

Off-target drug interactions are a major reason for candidate failure in the drug discovery process. Anticipating potential drug's adverse effects in the early stages is necessary to minimize health risks to patients, animal testing, and economical costs. With the constantly increasing size of virtual screening libraries, AI-driven methods can be exploited as first-tier screening tools to provide liability estimation for drug candidates. In this work we present ProfhEX, an AI-driven suite of 46 OECD-compliant machine learning models that can profile small molecules on 7 relevant liability groups: cardiovascular, central nervous system, gastrointestinal, endocrine, renal, pulmonary and immune system toxicities. Experimental affinity data was collected from public and commercial data sources. The entire chemical space comprised 289'202 activity data for a total of 210'116 unique compounds, spanning over 46 targets with dataset sizes ranging from 819 to 18896. Gradient boosting and random forest algorithms were initially employed and ensembled for the selection of a champion model. Models were validated according to the OECD principles, including robust internal (cross validation, bootstrap, y-scrambling) and external validation. Champion models achieved an average Pearson correlation coefficient of 0.84 (SD of 0.05), an R² determination coefficient of 0.68 (SD = 0.1) and a root mean squared error of 0.69 (SD of 0.08). All liability groups showed good hit-detection power with an average enrichment factor at 5% of 13.1 (SD of 4.5) and AUC of 0.92 (SD of 0.05). Benchmarking against already existing tools demonstrated the predictive power of ProfhEX models for large-scale liability profiling. This platform will be further expanded with the inclusion of new targets and through complementary modelling approaches, such as structure and pharmacophore-based models. ProfhEX is freely accessible at the following address: https://profhex.exscalate.eu/ .

Next-Generation Adjuvants: Applying Engineering Methods to Create and Evaluate Novel Immunological Responses

Adjuvants are a critical component of vaccines. Adjuvants typically target receptors that activate innate immune signaling pathways. Historically, adjuvant development has been laborious and slow, but has begun to accelerate over the past decade. Current adjuvant development consists of screening for an activating molecule, formulating lead molecules with an antigen, and testing this combination in an animal model. There are very few adjuvants approved for use in vaccines, however, as new candidates often fail due to poor clinical efficacy, intolerable side effects, or formulation limitations. Here, we consider new approaches using tools from engineering to improve next-generation adjuvant discovery and development. These approaches will create new immunological outcomes that will be evaluated with novel diagnostic tools. Potential improved immunological outcomes include reduced vaccine reactogenicity, tunable adaptive responses, and enhanced adjuvant delivery. Evaluations of these outcomes can leverage computational approaches to interpret "big data" obtained from experimentation. Applying engineering concepts and solutions will provide alternative perspectives, further accelerating the field of adjuvant discovery.

NHGNN-DTA: a node-adaptive hybrid graph neural network for interpretable drug-target binding affinity prediction

MOTIVATION

Large-scale prediction of drug-target affinity (DTA) plays an important role in drug discovery. In recent years, machine learning algorithms have made great progress in DTA prediction by utilizing sequence or structural information of both drugs and proteins. However, sequence-based algorithms ignore the structural information of molecules and proteins, while graph-based algorithms are insufficient in feature extraction and information interaction.

RESULTS

In this article, we propose NHGNN-DTA, a node-adaptive hybrid neural network for interpretable DTA prediction. It can adaptively acquire feature representations of drugs and proteins and allow information to interact at the graph level, effectively combining the advantages of both sequence-based and graph-based approaches. Experimental results have shown that NHGNN-DTA achieved new state-of-the-art performance. It achieved the mean squared error (MSE) of 0.196 on the Davis dataset (below 0.2 for the first time) and 0.124 on the KIBA dataset (3% improvement). Meanwhile, in the case of cold start scenario, NHGNN-DTA proved to be more robust and more effective with unseen inputs than baseline methods. Furthermore, the multi-head self-attention mechanism endows the model with interpretability, providing new exploratory insights for drug discovery. The case study on Omicron variants of SARS-CoV-2 illustrates the efficient utilization of drug repurposing in COVID-19.

AVAILABILITY AND IMPLEMENTATION

The source code and data are available at https://github.com/hehh77/NHGNN-DTA.