In biopharma, protein discovery has been completely disrupted by the use of artificial intelligence in recent years, whereas the food industry has been slower to adopt such tools, says Shiru founder and CEO Dr. Jasmin Hume. “But I think that disruption is certainly going to happen.”
California-based startup Shiru has developed a first-of-its-kind ingredient discovery platform and marketplace for proteins, enabling companies in food and cosmetics to harness its AI and machine learning capabilities to identify and test proteins from high-intensity sweeteners to binders and emulsifiers.
Via a simple web interface, ProteinDiscovery.ai lets users search a database of millions of molecules by protein sequence, functional use, and successful expression (how efficiently the protein might be expressed in microbes via precision fermentation), Hume told AgFunderNews at the Future Food-Tech summit in San Francisco.
Discovering the ‘most functional proteins from nature’
Unlike firms making ‘designer proteins,’ Shiru is exploring proteins that already exist in nature, said Hume. “Shiru has created a novel way to discover the most functional proteins from nature for applications in flavor and texture as well as bioactivity for food and cosmetics.
“Shiru has built the most extensive library of natural proteins from plants, algal sources, and microbes. It’s about 33 million sequences. So we leverage that, as well as certain bioinformatics and machine learning capabilities, to effectively cluster that data based on a protein’s functional properties.”
In a nutshell, Shiru looks at the job a protein needs to perform in a food product such as emulsifying or enhancing sweetness, and translates that into prompts for its search algorithms so it can identify proteins that can deliver.
Shiru can then produce samples of those proteins for partners that don’t have the capabilities to produce them in-house, she said.
Given that a protein isn’t much use if it can’t be manufactured cost-efficiently at scale, Shiru can also predict how well, or indeed if, a given protein can be expressed in a microbial host, she added.
It has also forged a partnership with molecular farming startup GreenLab, which has developed tools to express certain proteins in corn kernels, she said. “We want to create the best and most efficient paths to market, whether that be through precision fermentation in microbial hosts, or expressing those proteins in crops such as corn, which is another really scalable way to bring them to market.”
She added: “We’re applying new tools that, frankly, are not the most common tools from a food industry standpoint, such as large databases of proteomic information, biotechnology, AI, machine learning, although they are becoming more familiar. And what we aim to do is make this really plug and play for companies… to be able to say you don’t need to hire a team of molecular biologists, you just need to know what your ingredient needs to do. Let us all the rest for you.”
