A novel AI-based approach has been developed to discover novel deaminase proteins used for base editing through structure prediction and classification. This study focuses on an approach that develops a series of new technologies and uses only the cytidine deaminase superfamily to uncover new protein functions.
Researchers led by Dr. Caixia Gao at the Institute of Gene and Developmental Biology, Chinese Academy of Sciences, who are focused on plant genome editing, are eyeing the development of a base editor suite with unique AI-based capabilities. – Structural Prediction Assistance – greatly expands the utility of base editors in therapeutic and agricultural applications, opening up a wide range of applications for the discovery and creation of desirable plant genetic traits.
Result is, cell Paper “Discovery of deaminase function by structure-based protein clustering”.
Efforts to unearth novel proteins currently generally rely on amino acid sequences, which fail to provide robust links between protein structural information and function.
Base editing has the potential to revolutionize molecular crop breeding by introducing desirable traits into elite germplasm. The discovery of several deaminases has expanded the capacity for cytosine base editing. Although traditional sequence-based efforts have identified many proteins that can be used as base editors, there are still limitations to editing specific DNA sequences or species.
Standard approaches based solely on protein engineering and directed evolution have helped diversify base-editing properties, but challenges remain. Using AlphaFold2 to predict the structure of proteins within the deaminase protein family, researchers clustered and analyzed deaminases based on structural similarity. They identified five new deaminase clusters with cytidine deamination activity from the perspective of DNA base editors.
Moreover, this approach further reclassified SCP1.201, a group of cytidine deaminases previously thought to act on dsDNA, primarily to deaminate ssDNA. Through subsequent protein profiling and engineering efforts, we have developed a series of novel DNA-based editors with remarkable features. These deaminases exhibit properties such as higher efficiency, fewer occurrences of off-target editing events, editing at different preferred sequence motifs, and much smaller size.
The researchers emphasized that the development of a set of basic editors will enable customized applications for different therapeutics and agro-breeding efforts in the future. They developed a minimal single-strand-specific cytidine deaminase, allowing the first efficient cytosine base editor to be packaged into a single adeno-associated virus.
They also found a highly effective deaminase from this clade, focusing on soybean plants, a globally important crop that had previously exhibited poor editing by cytosine base editors.
In general, the recent advent of protein structure prediction using growing genomic databases will greatly accelerate the development of new bioengineering tools. Moreover, this work will be of broad interest to the larger research community in phylogeny, metagenomics, protein engineering and evolution, genome editing, and plant breeding.
