GoKawiilTiny shifts of less than one ten-billionth of a metre in the gap between an enzyme and its target might seem too subtle to measure directly, or to predict with artificial intelligence. Yet such shifts are exactly what Meng et al.1, writing in Nature, have leveraged to improve the fidelity of ‘base editors’ — a class of genome-rewriting tools that can make single-nucleotide changes at targeted sites, offering potential for treating many genetic diseases. By using AI to predict the structures of thousands of base editors in complex with nucleic acids, and analysing the patterns that emerged, the authors have captured the differences in molecular interactions that correlate with on-target versus off-target editing.
doi: https://doi.org/10.1038/d41586-026-02042-1
References Meng, H. et al. Nature https://doi.org/10.1038/s41586-026-10794-z (2026). Jinek, M. et al. Science 337, 816–821 (2012). Cong, L. et al. Science 339, 819–823 (2013). Komor, A. C., Kim, Y. B., Packer, M. S., Zuris, J. A. & Liu, D. R. Nature 533, 420–424 (2016). Gaudelli, N. M. et al. Nature 551, 464–471 (2017). Vakulskas, C. A. et al. Nature Med. 24, 1216–1224 (2018). Choi, G. C. G. et al. Nature Methods 16, 722–730 (2019). Jumper, J. et al. Nature 596, 583–589 (2021). Abramson, J. et al. Nature 630, 493–500 (2024). Baek, M. et al. Nature Methods 21, 117–121 (2024). Chu, H. Y. et al. Cell Syst. 15, 193–203 (2024). Thean, D. G. L. et al. Nature Commun. 13, 2219 (2022). Silverstein, R. A. et al. Nature 643, 539–550 (2025). Ruffolo, J. A. et al. Nature 645, 518–525 (2025). Peng, J. et al. Mol. Cell 86, 1839–1855 (2026). Download references
Competing Interests The authors declare no competing interests.
Related Articles
Subjects