Rapid customization of base editors via machine learning-powered combinatorial mutagenesis
Clinical Snapshot
PICO Framework
| P — Population | Base editor proteins (evoAPOBEC1 and TadA variants) tested in human cells |
| I — Intervention | Machine learning-guided combinatorial mutagenesis for base editor optimization |
| C — Comparator | Previous base editor versions and unoptimized variants |
| O — Outcomes | Motif-specific editing activity, reduction in bystander edits, success rate in correcting disease-associated mutations |
Bottom Line
This study represents a significant advance in precision gene editing technology, combining machine learning with large-scale experimental screening to optimize base editors. The achievement of undetectable bystander effects in 50% of cases and successful prediction of functional variants addresses key safety concerns that have limited clinical translation of base editing. While the work is primarily laboratory-based, the demonstrated improvement in precision across >800 disease-associated mutations suggests strong potential for therapeutic applications. The integration of computational prediction with experimental validation offers a scalable approach for developing patient-specific editing tools. However, clinical translation will require extensive safety validation and regulatory approval. Australian researchers and clinicians should monitor developments in this field as it may inform future gene therapy protocols for inherited diseases.
Key Findings
P Value: Not reported
Effect Size: 50% of cases achieved undetectable bystander edits
Primary Outcome: Enhanced base editor precision with reduced bystander effects
Nnt Or Sensitivity: 63% prediction success rate for functional TadA variants across 2026 variants
Confidence Interval: Not reported
Clinical Application
Requires specialized laboratory infrastructure and expertise in gene editing technologies Relevant to Australian gene therapy research programs and potential future TGA-regulated clinical trials for genetic diseases Patients with genetic diseases amenable to base editing correction
Abstract
Being able to control the complementarity and hindrance between target DNA and base editor proteins enables precise, bystander-free editing. Here, we combined combinatorial mutagenesis with machine learning to analyze and engineer these interactions at scale. By profiling DNA motif preferences across 160,000 evoAPOBEC1 and 64 million TadA variants in human cells, we used as little as 0.004% of the mutational landscape to make predictions. This identified variants with motif-specific activity and eliminated residual adenine editing in cytosine base editors. In correcting >800 disease-associated mutations, our variants outperformed previous versions in precluding unintended edits at purine motifs, achieving undetectable bystander edits in 50% of cases. Additionally, a pre-trained, structure-based deep learning model predicted functional TadA variants with 63% success across 2026 variants spanning 26 amino acid sites, without experimental data and in a single prediction round. These approaches streamline the re-engineering of base editors for enhanced precision tailored to specific targets.
References
- 1.Peng, J., Chan, D. C. T., Chu, H. Y., Fong, J. H. C., Lam, Y. K., Cheung, M. S. H., Leung, W., Choi, G. C. G., & Wong, A. S. L. (2026). Rapid customization of base editors via machine learning-powered combinatorial mutagenesis. Molecular Cell. https://doi.org/10.1016/j.molcel.2026.03.030
Related Research
PloS one
Fundus image analysis of retinitis pigmentosa using artificial intelligence
25 July 2026
Journal of medical Internet research
Initial-Visit Specialty Triage in Rare Diseases Using Large Language Models: Retrospective Benchmarking Study
24 July 2026
Nucleic acids research
SNPnexus: an enhanced web platform for large-scale and multi-sample variant analysis (2025 update)
13 July 2026
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service