Last month, US biotech startup Climax Foods announced it had discovered the “first ever” plant-based casein, which can replicate the functional performance of conventional casein to make realistic animal-free cheese. Today, the company reveals that these innovations and developments have been made possible through a partnership with biotech research and data cloud platform Benchling.
“Sustainable foods are the future and data science is how we achieve this”
The collaboration, which uses Benchling’s platform for R&D data and insights, is helping Climax Foods manage high volumes of data to improve its formulation process and digital recipes. Through an innovative data-driven approach to R&D, the collaboration will accelerate the launch of plant-based cheese innovations onto the market.
“There’s a lot of excitement for how machine learning and artificial intelligence can be used in the food industry. At Climax, we’re delivering on that promise today,” says Karthik Sekar, Head of Data Science at Climax Foods.
Millions of combinations
Climax Foods uses a “precision process,” where AI and machine learning (ML) discover ingredients that can replicate dairy proteins among thousands of edible plants at an incredible speed. The ML process also includes the development of prototypes using data from plant-based cheeses, taking into account factors such as flavor, texture, nutritional value, and sustainability. Climax is initially focusing on dairy products, but says this technology could also be applied to other animal-based foods.
In January, the company launched its first product range, “zero compromise,” which includes Blue, Brie, Feta, and Chèvre cheeses — all developed using AI and ML to make them realistic and low-cost. A few months later, the revolutionary biotech announced a partnership with global cheese leader The Bel Group to introduce a new generation of plant-based cheeses.
“Our building blocks are the three hundred thousand varieties of plants, which can be combined in millions of different ways to reach specific textures, flavors, smells, and environmental impact. This is a huge combinatorial screening problem that even the largest labs can’t crack when done manually. Climax Foods is uniquely positioned to meet this challenge with our focus on data and machine learning,” argues Sekar.
“Ideal setup”
After undergoing significant expansion, Climax Foods needed the capability to handle more complex data science, optimizations, and machine learning. Specifically, its scientists required models to track “meltability” and cheese flavor changes while keeping cost increases below 10%.
The company wanted to replace its data systems with a consistent ontology across all teams to meet a standardized classification of concepts. Benchling’s centralized platform has proven to be a valuable tool for organizing concepts such as ingredients, formulations, and assays; it also came with a pre-established ontology that the Climax team could further build upon to support their research, saving time for scientists and resources for the company.
“Sustainable foods are the future and data science is how we achieve this. We continuously train our portfolio of machine intelligence tools to enable the plant-based recreation of any taste and texture, while optimizing for nutrition and lowering costs. Benchling specifically gives us a solid foundation where data is findable, accessible, interoperable, and reusable with FAIR data principles. This quality data and Benchling’s flexible system provide the ideal setup for developing and training our machine learning models,” says Sekar.
