Allozymes puts its accelerated enzymatics to work on a data and AI play, raising $15M

Allozymes’ innovative approach to rapidly testing millions of bio-based chemical reactions has transcended being merely a valuable service to becoming the foundation of a unique and invaluable dataset. This dataset has attracted significant interest from investors, leading to the company securing a $15 million Series A funding round, propelling it from a fledgling startup to a globally recognized resource.


In our initial coverage of the biotech startup back in 2021, Allozymes was in its nascent stages, with CEO and founder Peyman Salehian reminiscing about the humble beginnings: “Back then we were less than five people, and at our first lab — a thousand square feet.”

Since then, the company has experienced exponential growth, expanding to a team of 32 individuals spread across the United States, Europe, and Singapore. Furthermore, Allozymes now boasts 15 times the lab space it had initially, which has facilitated the acceleration of its already lightning-fast enzyme-screening technique.


While the core technology of the company remains unchanged since 2021, with detailed descriptions available in our original article, the essence lies in its revolutionary method of exploring enzymes. Enzymes, comprised of chains of amino acids that carry out specific tasks in biological systems, have traditionally been challenging to discover or engineer due to the vast number of variations. Each molecule can consist of hundreds of acids, with 20 options available for each position, leading to billions of potential permutations.


Traditionally, testing these variations could only be done at a rate of a few hundred per day in a standard laboratory setting. However, Allozymes has revolutionized this process by employing a method wherein millions of enzymes can be tested daily.

This is achieved by encapsulating them in tiny droplets and passing them through a specialized microfluidics system. A fitting analogy would be envisioning a conveyor belt with a camera positioned above it, swiftly scanning each passing item and automatically sorting them into distinct categories.


Enzymes play a crucial role across various industries within biotech and chemistry, facilitating processes such as converting raw materials into specific molecules or vice versa. However, finding cost-effective and efficient enzymes has historically been challenging, with the industry collectively testing approximately a million possibilities per year until recently.

Allozymes aims to revolutionize this landscape by targeting a staggering 7 billion enzyme variants in 2024, a thousandfold increase from previous efforts.


Peyman Salehian, CEO and founder of Allozymes, highlighted the company’s progress since 2021, emphasizing the successful operation of their enzyme-screening machines. Currently, Allozymes can screen up to 20 million enzyme variants per day, a substantial leap in throughput capacity. This increased efficiency has attracted customers from various industries, although some partnerships remain undisclosed due to non-disclosure agreements.


Among the documented case studies is Allozymes’ ability to produce phytoene, an enzyme naturally found in tomatoes, in bioreactors with a 99% reduction in water usage. Similarly, the company developed a bio-identical bisabolol, derived from the endangered candeia tree, through enzymatic pathways, ensuring a sustainable source of the chemical.

Additionally, Allozymes received funding to accelerate the production of “soluble sweet fiber” from plant and fruit fibers, offering an alternative to conventional sugars.


While microplastics-degrading enzymes are on Allozymes’ radar, their current business model primarily focuses on customer-driven projects. However, future initiatives may include ventures into plastics recycling and handling. Moving forward, Allozymes plans to expand beyond enzyme optimization services into molecule discovery to meet specific industry needs.


Allozymes’ enzyme-tailoring service, now dubbed SingZyme, will continue to provide entry-level options for clients seeking faster or more cost-effective solutions. Meanwhile, the company’s MultiZyme service will offer a broader range of enzyme discovery and refinement options.

Salehian envisions the vast amount of data collected through these services forming the foundation of “the biggest enzyme data library in the world,” safeguarding Allozymes’ intellectual property.


Allozymes puts its accelerated enzymatics to work on a data and AI play, raising $15M

CEO Peyman Salehian and CTO Akbar Vahidi, co-founders of Allozymes. Image Credits: Allozymes

Salehian highlighted the limitations of existing machine learning models in predicting chemical reactions, noting that while tools like AlphaFold can determine molecular structure, they fall short in forecasting interactions with other chemicals, a critical concern for the industry.


He emphasized the scarcity and fragmented nature of available data, citing industry-wide reliance on a mere 300 samples per day over two decades, a volume easily surpassed by Allozymes’ machines in a single day.


However, Allozymes is actively working on developing a machine learning model utilizing its extensive dataset. Salehian disclosed that initial testing of the model produced promising results, including suggesting a new molecule currently undergoing testing. While not unprecedented, this integration of machine learning into chemical research offers added confidence, albeit without replacing traditional processes.


The recent $15 million Series A funding round, inclusive of investments from Seventure Partners, NUS Technology Holdings, Thia Ventures, and ID Capital, along with continued support from existing investors like Xora Innovation, SOSV, Entrepreneur First, and Transpose Platform, underscores Allozymes’ promising trajectory.


Salehian expressed confidence in the company’s financial stability and outlined plans for potential future fundraising to facilitate expansion into the pharmaceutical sector and the establishment of a U.S. office.

Leave a Reply

Your email address will not be published. Required fields are marked *

error: Content is protected !!