RNA-targeting small molecules have the potential to unlock an entirely new therapeutic route for the treatment of disease: the next frontier in drug development. While there is much excitement in the targeting of RNAs, our understanding of the chemical interactions between RNAs and small molecules is still relatively poor. This becomes particularly stark when looking at the very limited availability of data compared to the well-explored area of protein binding small molecules - and resources such as ChEMBL or the PDB. This is coupled with low hit rates in biochemical high throughput screening assays, and it has long been anticipated that identifying a molecule capable of binding to a specific RNA structure element and eliciting a functional outcome, rather than pan-RNA binders, will be an extremely difficult task. Data availability that explores chemical space relative to RNA structures, is crucial to help address these challenges. In this work, we examined publicly available RNA binding data for small molecules and generated additional biochemical data. We have then used chemical informatics approaches to explore the relationship between molecules in these datasets to determine whether FDA-approved protein-targeting drugs and RNA binding compounds occupy a different chemical space, and to identify chemical characteristics likely driving RNA binding. To supplement these efforts we have deployed machine learning algorithms in an attempt to use chemical features to differentiate between RNA binders and non-binders, and RNA-specific binders and RNA pan-binders. Using external validation we show these algorithms are able to improve experimental hit rates in high throughput screens, increasing the hit rate compared to a diversity screen by 4X, and inform chemical understanding through SHAP analysis. By informing on what chemical features drive preferential RNA binding, the approach described here represents an important step forward in RNA-targeted drug discovery.

Combined chemoinformatics-machine learning framework to identify RNA binding properties of small molecules

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.