Genome editing holds great potential for significantly improving the way we treat and understand diseases. Modern genome editing tools like CRISPR-based systems require the design of guide RNA sequence (gRNA) that binds to an area of interest within the DNA. Guide sequences vary considerably in efficacy. In this project, we will develop computational tools that predict the efficacy of a given guide sequence.
Computational tools for predicting guide efficacy are used when designing genome editing experiments. The accuracy of available tools for predicting the activity of guide sequences is low and this delays the progress of genome editing experiments.
In this proposal, our goal is to shorten the time needed to carry successful genome editing experiments by developing better tools for predicting guide-sequence activity. Results obtained using the new prediction algorithms will also be used to gain insights into the factors influencing the efficiency of different guide sequences. Modern tools based on resent innovations in artificial intelligence and computer vision research will be utilized to develop the proposed tools.