Machine learning (ML) is increasingly valuable for predicting molecular properties and toxicity in drug discovery. However, toxicity-related end points have always been challenging to evaluate experimentally with respect to in vivo translation due to the required resources for human and animal studies; this has impacted data availability in the field. ML can augment or even potentially replace traditional experimental processes depending on the project phase and specific goals of the prediction. For instance, models can be used to select promising compounds for on-target effects or to deselect those with undesirable characteristics (e.g., off-target or ineffective due to unfavorable pharmacokinetics). However, reliance on ML is not without risks, due to biases stemming from nonrepresentative training data, incompatible choice of algorithm to represent the underlying data, or poor model building and validation approaches. This might lead to inaccurate predictions, misinterpretation of the confidence in ML predictions, and ultimately suboptimal decision-making. Hence, understanding the predictive validity of ML models is of utmost importance to enable faster drug development timelines while improving the quality of decisions. This perspective emphasizes the need to enhance the understanding and application of machine learning models in drug discovery, focusing on well-defined data sets for toxicity prediction based on small molecule structures. We focus on five crucial pillars for success with ML-driven molecular property and toxicity prediction: (1) data set selection, (2) structural representations, (3) model algorithm, (4) model validation, and (5) translation of predictions to decision-making. Understanding these key pillars will foster collaboration and coordination between ML researchers and toxicologists, which will help to advance drug discovery and development.

Introduction: Artificial Intelligence (AI) has become a component of our everyday lives, with applications ranging from recommendations on what to buy to the analysis of radiology images. Many of the techniques originally developed for other fields such as language translation and computer vision are now being applied in drug discovery. AI has enabled multiple aspects of drug discovery including the analysis of high content screening data, and the design and synthesis of new molecules. Areas covered: This perspective provides an overview of the application of AI in several areas relevant to drug discovery including property prediction, molecule generation, image analysis, and organic synthesis planning. Expert opinion: While a variety of machine learning methods are now being routinely used to predict biological activity and ADME properties, methods of representing molecules continue to evolve. Molecule generation methods are relatively new and unproven but hold the potential to access new, unexplored areas of chemical space. The application of AI in drug discovery will continue to benefit from dedicated research, as well as AI developments in other fields. With this pairing algorithmic advancements and high-quality data, the impact of AI in drug discovery will continue to grow in the coming years.

PIK3CA (PI3Kα) is a lipid kinase commonly mutated in cancer, including ∼40% of hormone receptor–positive breast cancer. The most frequently observed mutants occur in the kinase and helical domains. Orthosteric PI3Kα inhibitors suffer from poor selectivity leading to undesirable side effects, most prominently hyperglycemia due to inhibition of wild-type (WT) PI3Kα. Here, we used molecular dynamics simulations and cryo-electron microscopy to identify an allosteric network that provides an explanation for how mutations favor PI3Kα activation. A DNA-encoded library screen leveraging electron microscopy-optimized constructs, differential enrichment, and an orthosteric-blocking compound led to the identification of RLY-2608, a first-in-class allosteric mutant-selective inhibitor of PI3Kα. RLY-2608 inhibited tumor growth in PIK3CA-mutant xenograft models with minimal impact on insulin, a marker of dysregulated glucose homeostasis. RLY-2608 elicited objective tumor responses in two patients diagnosed with advanced hormone receptor–positive breast cancer with kinase or helical domain PIK3CA mutations, with no observed WT PI3Kα-related toxicities. Significance: Treatments for PIK3CA-mutant cancers are limited by toxicities associated with the inhibition of WT PI3Kα. Molecular dynamics, cryo-electron microscopy, and DNA-encoded libraries were used to develop RLY-2608, a first-in-class inhibitor that demonstrates mutant selectivity in patients. This marks the advance of clinical mutant-selective inhibition that overcomes limitations of orthosteric PI3Kα inhibitors. See related commentary by Gong and Vanhaesebroeck, p. 204 . See related article by Varkaris et al., p. 227 . This article is featured in Selected Articles from This Issue, p. 201