Abstract:Summary Dual base editors (DBEs) enable simultaneous A‐to‐G and C‐to‐T conversions, expanding mutation types. However, low editing efficiency and narrow targeting range limit the widespread use of DBEs in plants. The single‐strand DNA binding domain of RAD51 DBD can be fused to base editors to improve their editing efficiency. However, it remains unclear how the DBD affects dual base editing performance in plants. In this study, we generated a series of novel plant DBE‐SpGn tools consisting of nine constructs using the high‐activity cytidine deaminase evoFERNY, adenosine deaminase TadA8e and DBD in various fusion modes with the PAM‐flexible Streptococcus pyogenes Cas9 (SpCas9) nickase variant SpGn (with NG‐PAM). By analysing their editing performance on 48 targets in rice, we found that DBE‐SpGn constructs containing a single DBD and deaminases located at the N‐terminus of SpGn exhibited the highest editing efficiencies. Meanwhile, constructs with deaminases located at the C‐terminus and/or multiple DBDs failed to function normally and exhibited inhibited editing activity. We identified three particularly high‐efficiency dual base editors (C‐A‐SpGn, C‐A‐D‐SpGn and A‐C‐D‐SpGn), named PhieDBEs (Plant high‐efficiency dual base editors), capable of producing efficient dual base conversions within a narrow editing window (M5 ~ M9, M = A/C). The editing efficiency of C‐A‐D‐SpGn was as high as 95.2% at certain target sites, with frequencies of simultaneous C‐to‐T and A‐to‐G conversions as high as 81.0%. In summary, PhieDBEs (especially C‐A‐D‐SpGn) can produce diverse mutants and may prove useful in a wide variety of applications, including plant functional genomics, precise mutagenesis, directed evolution and crop genetic improvement, among others.

BEdeepon: an in silico tool for prediction of base editor efficiencies and outcomes

BEdeepoff: an in Silico Tool for Off-Target Prediction of ABE and CBE Base Editors

Prediction of Base Editor Off-Targets by Deep Learning

Deep learning models to predict the editing efficiencies and outcomes of diverse base editors

Comprehensive evaluation and prediction of editing outcomes for near-PAMless adenine and cytosine base editors

Annotation and Evaluation of Base Editing Outcomes in Multiple Cell Types Using CRISPRbase.

Optimization of C-to-G base editors with sequence context preference predictable by machine learning methods

Web-based design and analysis tools for CRISPR base editing

Improving adenine and dual base editors through introduction of TadA-8e and Rad51DBD

Comparison of Cytosine Base Editors and Development of the BEable-GPS Database for Targeting Pathogenic SNVs

OPED: an Optimal Prime Editing Guide RNA Designer Based on Deep Learning and Transfer Learning

ACBE, a new base editor for simultaneous C-to-T and A-to-G substitutions in mammalian systems

BES-Designer: A Web Tool to Design Guide RNAs for Base Editing to Simplify Library

PhieDBEs: a DBD‐containing, PAM‐flexible, high‐efficiency dual base editor toolbox with wide targeting scope for use in plants

EditR: A novel base editing quantification software using Sanger sequencing

Human Cell Based Directed Evolution of Adenine Base Editors with Improved Efficiency

Joint genotypic and phenotypic outcome modeling improves base editing variant effect quantification

Design of prime-editing guide RNAs with deep transfer learning

Eliminating Base-Editor-induced Genome-Wide and Transcriptome-Wide Off-Target Mutations

Expanding C–T base editing toolkit with diversified cytidine deaminases

CHANGE-seq-BE enables simultaneously sensitive and unbiased profiling of base editor genome-wide activity