Abstract
Automating structural optimization of drug molecules for on-target potency by machine learning is an open challenge in chemistry. Here, we capitalize on the ability of chemical language models (CLMs) to learn from sequential data and design new molecules with desired properties. We establish a training strategy mimicking the learning trajectory of a drug discovery program. Incremental CLM fine-tuning with increasingly potent template molecules from a given structure-activity relationship (SAR) series successfully biases the model to design highly active analogues. Prospective application of this technique to ligand development enables the data-driven design of molecules exceeding known representatives of given bioactive chemotypes in potency without external scoring. Our results reveal an ability of CLMs to capture SAR patterns and long-range dependencies, and to exploit SAR knowledge in designing analogues with improved on-target activity de novo corroborating their applicability to structural optimization of drug molecules.