Explore the vast range of titles available.

To be effective as a drug, a potent molecule must reach its target in the body in sufficient concentration, and stay there in a bioactive form long enough for the expected biologic events to occur. Drug development involves assessment of absorption, distribution, metabolism and excretion (ADME) increasingly earlier in the discovery process, at a stage when considered compounds are numerous but access to the physical samples is limited. In that context, computer models constitute valid alternatives to experiments. Here, we present the new SwissADME web tool that gives free access to a pool of fast yet robust predictive models for physicochemical properties, pharmacokinetics, drug-likeness and medicinal chemistry friendliness, among which in-house proficient methods such as the BOILED-Egg, iLOGP and Bioavailability Radar. Easy efficient input and interpretation are ensured thanks to a user-friendly interface through the login-free website http://www.swissadme.ch . Specialists, but also nonexpert in cheminformatics or computational chemistry can predict rapidly key parameters for a collection of molecules to support their drug discovery endeavours.

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. AI has become a crucial tool for drug discovery, but how to properly represent molecules for data-driven property prediction is still an open question. Here the authors evaluate 62,820 models to highlight existing challenges, the impact of activity cliffs, and the crucial role of dataset size.

Finding new molecules with a desired biological activity is an extremely difficult task. In this context, artificial intelligence and generative models have been used for molecular de novo design and compound optimization. Herein, we report a generative model that bridges systems biology and molecular design, conditioning a generative adversarial network with transcriptomic data. By doing so, we can automatically design molecules that have a high probability to induce a desired transcriptomic profile. As long as the gene expression signature of the desired state is provided, this model is able to design active-like molecules for desired targets without any previous target annotation of the training compounds. Molecules designed by this model are more similar to active compounds than the ones identified by similarity of gene expression signatures. Overall, this method represents an alternative approach to bridge chemistry and biology in the long and difficult road of drug discovery. High quality hit identification remains a considerable challenge in de novo drug design. Here, the authors train a generative adversarial network with transcriptome profiles induced by a large set of compounds, enabling it to design molecules that are likely to induce desired expression profiles.

Deep learning bears promise for drug discovery, including advanced image analysis, prediction of molecular structure and function, and automated generation of innovative chemical entities with bespoke properties. Despite the growing number of successful prospective applications, the underlying mathematical models often remain elusive to interpretation by the human mind. There is a demand for ‘explainable’ deep learning methods to address the need for a new narrative of the machine language of the molecular sciences. This Review summarizes the most prominent algorithmic concepts of explainable artificial intelligence, and forecasts future opportunities, potential applications as well as several remaining challenges. We also hope it encourages additional efforts towards the development and acceptance of explainable artificial intelligence techniques. Drug discovery has recently profited greatly from the use of deep learning models. However, these models can be notoriously hard to interpret. In this Review, Jiménez-Luna and colleagues summarize recent approaches to use explainable artificial intelligence techniques in drug discovery.

Streamlined data-driven drug discovery remains challenging, especially in resource-limited settings. We present ZairaChem, an artificial intelligence (AI)- and machine learning (ML)-based tool for quantitative structure-activity/property relationship (QSAR/QSPR) modelling. ZairaChem is fully automated, requires low computational resources and works across a broad spectrum of datasets. We describe an end-to-end implementation at the H3D Centre, the leading integrated drug discovery unit in Africa, at which no prior AI/ML capabilities were available. By leveraging in-house data collected over a decade, we have developed a virtual screening cascade for malaria and tuberculosis drug discovery comprising 15 models for key decision-making assays ranging from whole-cell phenotypic screening and cytotoxicity to aqueous solubility, permeability, microsomal metabolic stability, cytochrome inhibition, and cardiotoxicity. We show how computational profiling of compounds, prior to synthesis and testing, can inform progression of frontrunner compounds at H3D. This project is a first-of-its-kind deployment at scale of AI/ML tools in a research centre operating in a low-resource setting. Streamlined data-driven drug discovery remains challenging, especially in resource-limited settings. Here, the authors present ZairaChem, an AI/ML tool that streamlines QSAR/QSPR modelling, implemented for the first time at the H3D Centre in South Africa.

Key Points Ligand efficiency measures quantify the molecular properties, particularly size and lipophilicity, of small molecules that are required to gain binding affinity to a drug target. There are additional efficiency measures for groups in a molecule, and for combinations of size and lipophilicity. The application of ligand efficiency metrics has been widely reported in the selection and optimization of fragments, hits and leads. In particular, optimization of lipophilic ligand efficiency shows that it is possible to increase affinity and reduce lipophilicity at the same time, even with challenging 'lipophile-preferring' targets. Mean ligand efficiency measures of molecules acting at a specific target, when combined with their drug-like physicochemical properties, are a practical means of estimating target 'druggability'. This is exemplified with 480 target–assay pairs from the primary literature. Across these targets, correlations between biological activity in vitro and physicochemical properties are generally weak, which shows that increasing activity by increasing physicochemical properties is not always necessary. An analysis of 46 recently marketed oral drugs shows that they frequently have highly optimized ligand efficiency values and lipophilic ligand efficiency values for their target. Compared with 'only-in-class' oral drugs, only 1.5% of all molecules per target — on average — possess superior combined ligand efficiency and lipophilic ligand efficiency values. Optimizing ligand efficiencies based on both molecular size and lipophilicity, when set in the context of the specific target, has the potential to ameliorate the molecular inflation that pervades current practice in medicinal chemistry, and to increase the ability to develop drug candidates. Ligand efficiency metrics quantify the molecular properties required to gain binding affinity for a drug target. This article discusses the application of such metrics in the selection and optimization of fragments, hits and leads, highlighting how optimizing ligand efficiency metrics based on both molecular mass and lipophilicity, when set in the context of the specific target, has the potential to increase the quality of drug candidates. The judicious application of ligand or binding efficiency metrics, which quantify the molecular properties required to obtain binding affinity for a drug target, is gaining traction in the selection and optimization of fragments, hits and leads. Retrospective analysis of recently marketed oral drugs shows that they frequently have highly optimized ligand efficiency values for their targets. Optimizing ligand efficiency metrics based on both molecular mass and lipophilicity, when set in the context of the specific target, has the potential to ameliorate the inflation of these properties that has been observed in current medicinal chemistry practice, and to increase the quality of drug candidates.

Generative drug design facilitates the creation of compounds effective against pathogenic target proteins. This opens up the potential to discover novel compounds within the vast chemical space and fosters the development of innovative therapeutic strategies. However, the practicality of generated molecules is often limited, as many designs focus on a narrow set of drug-related properties, failing to improve the success rate of subsequent drug discovery process. To overcome these challenges, we develop TamGen, a method that employs a GPT-like chemical language model and enables target-aware molecule generation and compound refinement. We demonstrate that the compounds generated by TamGen have improved molecular quality and viability. Additionally, we have integrated TamGen into a drug discovery pipeline and identified 14 compounds showing compelling inhibitory activity against the Tuberculosis ClpP protease, with the most effective compound exhibiting a half maximal inhibitory concentration (IC 50 ) of 1.9 μM. Our findings underscore the practical potential and real-world applicability of generative drug design approaches, paving the way for future advancements in the field. Generative AI holds promise for creating novel compounds. Here, authors introduce TamGen, a GPT-like model designed to generate molecules tailored to specific target proteins. TamGen identified 14 potent compounds against the Tuberculosis ClpP protease, showing its potential for drug discovery.

The growing prevalence of resistance to antibiotics motivates the search for new antibacterial agents. Antimicrobial peptides are a diverse class of well-studied membrane-active peptides which function as part of the innate host defence system, and form a promising avenue in antibiotic drug research. Some antimicrobial peptides exhibit toxicity against eukaryotic membranes, typically characterised by hemolytic activity assays, but currently, the understanding of what differentiates hemolytic and non-hemolytic peptides is limited. This study leverages advances in machine learning research to produce a novel artificial neural network classifier for the prediction of hemolytic activity from a peptide’s primary sequence. The classifier achieves best-in-class performance, with cross-validated accuracy of 85.7 % and Matthews correlation coefficient of 0.71. This innovative classifier is available as a web server at https://research.timmons.eu/happenn , allowing the research community to utilise it for in silico screening of peptide drug candidates for high therapeutic efficacies.

Key Points Chemical predictive modelling encompasses empirical computational methods based on observed patterns in data that guide the design of future compounds. Simple physicochemical property-based guidelines and structure-based chemical filters, such as AstraZeneca's AZFilters, are used to identify poor-quality compounds in screening set selection and compound design. Despite their limitations, quantitative structure–activity relationship (QSAR) models of ADMET (absorption, distribution, metabolism, excretion and toxicity) properties are widely used in compound design; advice and guidance on the judicious use of QSAR methods has been published. A key problem with QSAR methods is estimating confidence in the predictions, which is linked to the definition of the model's domain of applicability. Project- or chemical series-specific QSAR models are one approach to solve the 'domain of applicability' problem but this approach requires automated model building to be practical for a large organization with multiple projects. Interpretable models and inverse QSAR methods provide additional information to inform the design of compounds with improved properties. Matched molecular pair analysis is complementary to standard QSAR, is interpretable and can be used to propose new compounds. Despite the progress in chemical predictive modelling techniques, their impact on improving compound quality is difficult to assess and is limited by cultural factors. These include continued debate over the application of compound quality guidelines and the diversity of opinions among medicinal chemists on attractive versus unattractive structures. Current techniques are most successful in modelling ADMET properties, whereas prediction of potency or efficacy is more challenging. Areas of active research include descriptors to incorporate chirality, multi-objective optimization and expert systems for compound optimization. The 'quality' of small-molecule drug candidates — encompassing aspects including their potency, selectivity and pharmacokinetic characteristics — is a key factor influencing the chances of success in clinical trials. Cumming and colleagues discuss the application of computational methods, particularly quantitative structure–activity relationships, in guiding the selection of higher-quality drug candidates, as well as cultural factors that may have affected their impact. The 'quality' of small-molecule drug candidates, encompassing aspects including their potency, selectivity and ADMET (absorption, distribution, metabolism, excretion and toxicity) characteristics, is a key factor influencing the chances of success in clinical trials. Importantly, such characteristics are under the control of chemists during the identification and optimization of lead compounds. Here, we discuss the application of computational methods, particularly quantitative structure–activity relationships (QSARs), in guiding the selection of higher-quality drug candidates, as well as cultural factors that may have affected their use and impact.