AI-assisted base editor development

Leverage Fudan’s CFFF high-performance computing platform to develop next-generation precision base editors with AI large language models and structural prediction.

Leveraging Fudan’s CFFF high-performance computing platform, we develop next-generation precision base editors using AI large language models and structural prediction.

Research aims

  • Build structure-aware datasets of cytosine/adenine deaminases and base editor architectures.
  • Use large language models to predict and optimize deaminase specificity, efficiency, and bystander effects.
  • Validate engineered base editors in cellular and in vivo models.
  • Release open-source models and design pipelines for the gene-editing community.