MMEJ-based Precision Gene Editing for applications in Gene Therapy and Functional Genomics

Abstract

Experiments in gene editing commonly elicit error-prone non-homologous end joining for DNA double-strand break (DSB) repair. Microhomology-mediated end joining (MMEJ) can generate more predictable outcomes for functional genomic and somatic therapeutic applications. MENTHU is a computational tool that predicts nuclease-targetable sites likely to result in MMEJ-repaired, homogeneous genotypes (PreMAs) in zebrafish. We deployed MENTHU on 5,885 distinct Cas9-mediated DSBs in mouse embryonic stem cells, and compared the predictions to those by inDelphi, another DSB repair predictive algorithm. MENTHU correctly identified 46% of all PreMAs available, doubling the sensitivity of inDelphi. We also introduce MENTHU@4, an MENTHU update trained on this large dataset. We trained two MENTHU-based algorithms on this larger dataset and validated them against each other, MENTHU, and inDelphi. Finally, we estimated the frequency and distribution of SpCas9-targetable PreMAs in vertebrate coding regions to evaluate MMEJ-based targeting for gene discovery. 44 out of 54 genes (81%) contained at least one early out-of-frame PreMA and 48 out of 54 (89%) did so when also considering Cas12a. We suggest that MMEJ can be deployed at scale for reverse genetics screenings and with sufficient intra-gene density rates to be viable for nearly all loss-of-function based gene editing therapeutic applications.