Abstract
The CRISPR/Cas genome editing approach in non-model organisms poses challenges that remain to be resolved. Here, we demonstrated a generalized roadmap for a de novo genome annotation approach applied to the non-model organism Macrobrachium rosenbergii. We also addressed the typical genome editing challenges arising from genetic variations, such as a high frequency of single nucleotide polymorphisms, differences in sex chromosomes, and repetitive sequences that can lead to off-target events. For the genome editing of M. rosenbergii, our laboratory recently adapted the CRISPR/Cas genome editing approach to embryos and the embryonic primary cell culture. In this continuation study, an annotation pipeline was trained to predict the gene models by leveraging the available genomic, transcriptomic, and proteomic data, and enabling accurate gene prediction and guide design for knock-outs. A next-generation sequencing analysis demonstrated a high frequency of genetic variations in genes on both autosomal and sex chromosomes, which have been shown to affect the accuracy of editing analyses. To enable future applications based on the CRISPR/Cas tool in non-model organisms, we also verified the reliability of editing efficiency and tracked off-target frequencies. Despite the lack of comprehensive information on non-model organisms, this study provides an example of the feasibility of selecting and editing specific genes with a high degree of certainty.