UC Riverside

Microbial biochemical production as a renewable alternative to traditional methods is a rapidly growing sector of industrial biotechnology. Non-conventional microbes are attractive targets for metabolic engineering to produce biochemicals as they can present a range of desirable traits that may help avoid complex and intensive engineering of less suitable model hosts. Yarrowia lipolytica is one such non-conventional yeast with an abundant acetyl-CoA pool and native capacity to produce and accumulate lipids to high levels. While there have been significant advances in the metabolic engineering of this yeast for the biosynthesis of oleochemicals and other value-added products, there is also a dearth of synthetic biology tools for genome engineering, functional genomic screening and rapid strain development. We have sought to overcome these limitations by developing CRISPR-Cas9 and Cas12a systems for multiplexed gene knockout, integration, regulation, and genome-wide screening. However, prediction of highly active guide RNA (gRNA) which are crucial in effective genome editing and improving confidence in hit calling, remains a challenge. To address this, we constructed two genome-wide libraries, one using SpCas9 and the other using LbCas12a, to target all protein coding sequences. A negative selection screen in the absence of DNA repair, was used to generate gRNA activity scores for both endonucleases. This genome-wide data served as input to a deep learning algorithm, DeepGuide, that could accurately predict high activity gRNA for both Cas9 and Cas12a. Another critical challenge in accurately assessing screening outcomes is accounting for the variability in gRNA activity. Poorly active guides targeting genes essential to screening conditions obscure the growth defects that are expected from disrupting them. Thus, we also developed acCRISPR, an end-to-end pipeline that used gRNA activity scores to provide an activity correction to the screening outcomes, thus accurately determining the fitness effect of disrupted genes. acCRISPR analysis of the Cas9 and Cas12a screens in Yarrowia enabled the determination of a high-confidence set of essential genes for growth under glucose, a common carbon source used for the industrial production of oleochemicals. acCRISPR was also used in high salt and low pH tolerance screens, to identify known and novel genes related to stress tolerance. Collectively, this thesis presents an experimental-computational framework for CRISPR-based functional genomics studies that may be expanded to other non-conventional organisms of interest.