Prediction of designer-recombinases for DNA editing with generative deep learning

Abstract

Site-specific tyrosine-type recombinases are effective tools for genome engineering, with the first engineered variants having demonstrated therapeutic potential. So far, adaptation to new DNA target site selectivity of designer-recombinases has been achieved mostly through iterative cycles of directed molecular evolution. While effective, directed molecular evolution methods are laborious and time consuming. Here we present RecGen (Recombinase Generator), an algorithm for the intelligent generation of designer-recombinases. We gathered the sequence information of over two million Cre-like recombinase sequences evolved for 89 different target sites with which we trained Conditional Variational Autoencoders for recombinase generation. Experimental validation demonstrated that the algorithm can predict recombinase sequences with activity on novel target-sites, indicating that RecGen is useful to accelerate the development of future designer-recombinases.

Recent Advances in Double-Strand Break-Free Kilobase-Scale Genome Editing Technologies

“…Despite these potential limitations, the discovery and characterization of additional recombinases has begun to yield new enzymes with unique and useful properties. Leveraging these data sets could lead to a machine-learning framework that can ab initio predict high probability recombinase enzyme variants with user-defined sequence specificities …”