Companies are increasingly exploring alternative protein sources in response to the growing consumer demand for sustainable options. One key trend for 2024, according to dairy cooperative FrieslandCampina, is "Alternative proteins go global". The shift toward alternative protein is driven by various factors, including environmental concerns associated with animal-based protein production, such as greenhouse gas emissions, deforestation, and biodiversity loss.
In 2024 the alternative protein innovation space is diverse, ranging from microalgae to beer brewing by-products. Yet formulating these proteins into appealing finished products remains a challenge.
Enter ProteinDiscovery.ai, Shriu’s artificial intelligence (AI) and machine learning platform launched earlier this month. The platform streamlines the discovery of optimal protein formulations, which according to Shiru can save companies millions of dollars and years of R&D efforts.
The California-based protein discovery company, founded in 2019, launched Oleopro in March of 2023, a plant protein-based fat ingredient designed to be used in a range of alternative protein food products. Just over a year later, in May 2024, Shiru launched ProteinDiscovery.ai.
“We’ve been using AI to identify high-value, novel, scalable proteins for years, fueling our own product development. With significant recent interest from CPGs [consumer packaged goods] and ingredients companies alike, we decided to open our toolbox to everyone,” said Jasmin Hume, Shiru founder and CEO.
Dual-approach platform
The platform has a dual approach: It is a marketplace for protein-based products and a discovery platform. According to Shiru, users can search, discover, pilot, and buy molecules for various applications, including food, agriculture, and personal care.
In a press release, ProteinDiscovery.ai describes itself as the “world's first protein marketplace and discovery platform”. Fi Global Insights spoke with Hume to understand how the platform differs from other platforms, such as Nuritas’ AI platform launched in 2023, which identifies and delivers plant-based peptides for food and beverage companies.
Hume said: “Shiru’s platform is the only one with a user interface and access to software capabilities, bringing radical clarity on IP [intellectual property] ownership and the BD [business development] process for novel ingredients.”
According to Hume, ProteinDiscovery.ai is the world’s largest natural protein archive – a searchable database of more than 33 million proteins, by sequence, metadata, expression, and functional use.
“Users can leverage our marketplace features and our discovery tools to search our catalogue for proteins with specific functionalities, predict protein expression, and search for proteins with similar functionalities that come from unexpected organisms,” she added.
Ordering a protein sample is 'as easy as finding a product of interest on Amazon'
Users can approach Shiru with a problem statement, such as the need for a clean-label and high-stability sweet protein. Next, Shiru uses proprietary machine-learning methods to predict proteins that meet the desired performance parameters.
The platform can then facilitate access to these proteins. “The site is extremely easy to use and is designed for simplicity. Ordering a protein sample is as easy as finding a product of interest on Amazon,” Hume said. “Once the user locates a protein of interest, they can purchase a sample through a familiar e-commerce flow and can license proteins on our site,” she added. Users can secure intellectual property rights, either exclusively or non-exclusively.
According to Hume, the protein samples available on ProteinDiscovery.ai are research-grade and are for the sole use of R&D, non-commercial use. “The samples are not intended for consumption. These are novel proteins offering major potential solutions and functionality; however, it is up to each individual company to own the product development and resulting regulatory approval process,” she said.
The platform currently offers two discovery tools: the protein search engine and the expressor. According to the ProteinDiscovery.ai website, the protein search engine enables users to search for proteins with similar functions and structures, providing them with the most relevant candidates. Meanwhile, the expressor tool predicts and enhances protein expression across various production hosts by utilising general protein language and structural representations.
Additionally, the platform will soon introduce two more features: the interactor and the secretor tools. The interactor tool will predict the likelihood of interaction between any pair of sequences, while the secretor tool will forecast the likelihood of protein secretion.
