Leveraging AI in cultivated seafood research
Two students from the faculties of Science and Mathematics are exploring new ways to bring cultivated seafood to the table. Kevin Shen (BSc '24) and Rikard Saqe's idea to apply computational modeling to understand how fish cells grow and transform, has earned them more than $700,000 in grants to scale their research efforts.
Shen is a Master of Science (Biology) student and a recent graduate in the Computer Science program, while Saqe is completing his undergraduate studies in the Faculty of Mathematics with hopes of pursuing a Master of Science afterward.
Throughout their undergraduate degrees at Waterloo, the two students completed co-op placements where they used AI and machine learning techniques to analyze large data sets. Their shared skill sets and passions led them to pursue bioinformatics, a field where they could tackle the complex problems of animal welfare and food security. Shen and Saqe believe that applications of AI are just starting to be explored and they are eager to be involved in these efforts.
"Climate change and food security are two of the biggest challenges we face as a society," Shen says. "Alternative proteins and cellular agriculture use far less land and water, providing a sustainable alternative to traditional meat products."
Their research is supported by grants and the guidance of Dr. Brian Dixon, a professor in the Department of Biology at Waterloo and a Canada Research Chair in Fish and Environmental Immunology, Dr. Brian Ingalls, a professor in the Department of Applied Mathematics, and Dr. Nguyen Vo (Ph.D. '15), a professor at Wilfrid Laurier University.
The students aim to establish a scientific basis that could contribute to the large-scale production of cultivated foods, specifically fish. These research goals were developed within the Waterloo Alt Protein Project design team, part of a global network of more than 60 student-led organizations.
It all starts with leveraging AI to analyze the data sets they generate. By using AI, they hope to facilitate an improved understanding of predicting and enhancing cell behavior, create detailed genomic and proteomic profiles of several fish species, and leverage multi-omics technologies including single-cell RNA sequencing.
These techniques will offer insights into the cellular makeup of tissues, revealing cells and showing how these differences impact biological functions and development.
The collected data will be used to develop computational models that enhance our understanding of gene regulation, cell growth and development. These models are intended to guide numerous problems within cultivated seafood, such as formulating optimized animal-free serum media, reliably inducing cell immortalization, and improving cell proliferation and differentiation at large scales.
"Cultivated seafood has the potential to reduce commercial fishing's environmental toll while lessening reliance on live animals for food," Saqe says. "It could also provide safer food options by eradicating harmful elements like mercury and microplastics."
All data and models will be publicly shared on a "cultivated meat atlas" platform to support other researchers in the field. This collective research aims to develop lab-grown fish that closely mimic an animal-based filet for commercial distribution, without the negative repercussions of fishing and aquaculture.
Their work so far, a comprehensive review of machine learning in cultivated meat created in collaboration with New Harvest and the Alberta Machine Intelligence Institute, is now available as a preprint.
More information:
Artificial Intelligence and Machine Learning in Cellular Agriculture. zenodo.org/records/11193570
Provided by University of Waterloo