Unlocking the Protein Production Potential of Non-Conventional Yeasts with Artificial Intelligence

Non-Conventional Yeasts (NCYs) offer unique benefits for protein production over traditional model organisms due to their ability to metabolise complex carbon sources, exhibit favorable posttranslational modifications, and maintain higher protein stability.

Despite these benefits, the full potential of NCYs remains untapped due to a limited understanding of their genomic grammar and gene regulation. Experimentally mapping every regulatory element in non-model species is infeasible given the vast size of DNA space.

To address this challenge, we propose a project that employs artificial intelligence (AI) approaches to map information encoded in DNA to molecular data on a genome scale, to enable efficient development of protein-producing hosts.

The project aims to characterize and experimentally verify gene expression regulatory elements in 96 NCYs to adopt non-model species for protein production and deepen our understanding of gene expression regulation arising from genetic diversity spanning over 500 million years of evolution.

We will combine omics assays, data-driven machine learning, and high-throughput microfluidics to develop a panel of hosts favorable for heterologous expression of dairy proteins.

This project will provide fundamental knowledge on expression regulation, gene design, and genome-editing technologies, significantly impacting biotechnology. The ultimate goal is to uncover the hidden potential of NCYs for sustainable and cost-effective protein production.