RNA-Binding Chemical Space library

The role of RNA molecules beyond genetic information transfer has been noticed previously [1], however recently a new impetus is focusing on RNA’s as drug targets [2, 3]. RNA’s as targets are generally tougher than proteins, but the expected payoff is also significant. In particular, messenger and non-coding RNAs contain highly structured elements, and evidence suggests that many of these structures are important for function. Targeting these RNAs with small molecules offers opportunities to therapeutically modulate numerous cellular processes, including those linked to ‘undruggable’ protein targets [4].

            High quality focused libraries are crucial to success at early stages of drug discovery. In order to quantify the properties qualify small molecules as being potential RNA binders a current research reports a neural network (NN) machine learning model built on a large (ca 1.6M data points) dataset of experimental RNA interaction measurements [5]. The study shows that the RNA binders are basically very similar to “drug-like” (Rule-of-five, Ro5) molecules, and the difference does not lie in any simple structural and physico-chemical descriptor combination. Instead a sophisticated (NN) ML model could well discriminate the chemical space of “regular” drug like molecules (basically implying protein binding) and RNA binding chemical space.

            Our focused library of potential RNA binders is composed of the ChemDiv’s drug-like molecules, which are predicted to have high probability of RNA binding according to the ML model [5]. We believe that the selected 24K molecules can serve as a judicious starting point for early stages of drug discovery targeting RNA.

References

1. Thomas J. R. & Hergenrother P. J. (2008) // Chem. Rev., 108(4), 1171-1224. 10.1021/cr0681546

2. Fullenkamp C. R. & Schneekloth J. S. (2023) // Nature Chemistry, 15(10), 1329-1331. 10.1038/s41557-023-01330-x

3. Hargrove A. E. (2020) //  Chem. Communications, 56(94), 14744-14756. 10.1039/D0CC06796B

4. Warner K. D. et al. (2018) // Nature Rev. Drug Disc., 17(8), 547-558. 10.1038/nrd.2018.93 

5. Yazdani K. et al. (2023) // Angewandte Chemie, 135(11), e202211358. 10.1002/ange.202211358