Yan Leyfman, Medical Oncologist, Co-Founder and Executive Director of MedNews Week, shared a post on LinkedIn:
“AI is now helping redesign the molecular machinery behind genome editing itself.
Prime editing has been one of the most promising advances in CRISPR technology: the ability to precisely rewrite DNA without creating double-strand breaks.
But there’s been a persistent problem: Many engineering changes that improve editing efficiency also destabilize the protein, limiting expression and performance.
In this study, researchers turned to AI-guided protein design.
Using structure-informed models including ProteinMPNN, they redesigned the reverse transcriptase component of prime editors while preserving the catalytic regions required for editing.
The result:
- Prime editors carrying up to 163 amino acid substitutions
- Improved folding stability
- Higher soluble expression
- Up to 2-fold higher intracellular protein levels after mRNA delivery
And the functional impact was substantial.
Across human primary cells and multiple delivery platforms, the redesigned PE8 editors achieved markedly improved editing efficiency.
In mice: Editing efficiency reached up to 2.9-fold higher than current state-of-the-art prime editors including PE6, PE7, and PEmax.
The bigger story: AI in biology is moving beyond prediction.
It’s increasingly being used to design entirely improved molecular systems – from proteins and antibodies to now the core enzymes powering genome editing.
This is a glimpse of a future where therapeutic platforms themselves become AI-optimized.”
Title: AI-guided redesign of laboratory-evolved reverse transcriptases enhances prime editing
Authors: Y. Allen Tao, Holt A. Sakai, Allen Y. Jiang, Nicholas A. Krasnow, Vasilii S. Vaganov, Brian Shim, Zachary Barsdale, Smriti Pandey, Nouraiz Ahmed, Man Na, Ting-Wei Liao, Keyede Oye, Ana Cristian, Emily Zhang, Joy A. Xu, Mattijs Bulcaen, David R. Liu
Other articles featuring Yan Leyfman on OncoDaily.