Connect with us

Healthcare

Modulus Therapeutics Secures Funding, Looks to Expand Cell Therapy Design Platform

Published

 on

Seattle-based AI-powered cell therapy design company Modulus Therapeutics has announced it completed an oversubscribed $3.5 million Seed round. Led by Madrona Venture Group, there was also participation from KdT Ventures and the Allen Institute for AI (AI2). 

The new funding will go toward expanding the company’s Modulus’ Convergent Design platform. Along with the funding announcement, the company has formed their Scientific Advisory Board, which was founded by Alana Welm and Raphael Gottardo.

Developing the System

Max Darnell, Ph.D, founded Modulus back in 2020 along with bioinformatician Bryce Daines, Ph.D. The two were part of the Allen Institute of Artificial Intelligence (AI2), which is a research institute established by Paul Allen, co-founder of Microsoft. According to the founders, they wanted to invent a platform for systematic and repeatable discovery of next-generation cell therapies, which could be used to improve the immune system’s fight against solid tumors.

Bryan Hale is Managing Director at the AI2 Incubator. 

“The AI2 Incubator’s mission is to help founders launch AI companies that have the potential to change the world,” said Hale. “By harnessing cutting edge AI to seek out life-saving treatments for solid tumors, Modulus has the potential for a truly big impact.”

Chris Picardo is Investor at Madrona Venture Group. 

“Modulus has an ambitious goal and an innovative platform combining modern machine learning, synthetic biology, and cutting-edge lab automation to treat tumors and eradicate cancer,” said Picardoas. “Their method unlocks true high throughput screening for cancer fighting NK cells and creates a massive new dataset that is perfectly suited for AI.  We are excited to work with them as they build out their team and capabilities.”

Developing Therapies

The company’s top priority is to develop therapies based on natural killer cells for the treatment of metastatic breast cancer, and it is building on recent progress made in deploying them against solid tumors.

Alana Welm is a scientific advisor. 

“High throughput discovery approaches in cell types such as natural killer cells have lagged behind those in other cell types,” said Welm. “We’re really excited about the therapeutic prospects of exploring the different ways that these cells can be engineered.” 

The design process for these cells is still a complex process that is limited by human understanding of biology, so Modulus is developing its Convergent Design platform to combine key technology components, such as gene editing, machine learning, multi-omics, and high-throughput in vivo screening. This will help bring an intelligent and unbiased approach to cell therapy design. 

Max Darnell, Ph.D. is co-founder of Modulus. 

“When combined, these technologies provide a path for discovering the enhancements that cells need to accomplish a complex therapeutic task, like fighting a solid tumor,” said Darnell. “Current cell therapy design is like trying to find your way out of the woods without a GPS. This platform gives us the tools to navigate the cell’s design space systematically and effectively.”

Instead of focusing on a single feature like targeting, Convergent Design works to “enable simultaneous improvement along multiple axes of therapeutic importance converging on optimized cell designs,” according to the release. 

“Modulus is the only company we’ve seen which is focused on engineering the chassis of these cells,” said Rima Chakrabarti, MD, Principal at KdT Ventures. “By modulating the innate cellular machinery, Modulus can optimize not only for therapeutic efficacy but for improved manufacturability and storage as well, bringing us closer to curative, off-the-shelf immune cell therapies.”

Instead of learning the impact of one gene at a time, one of the keys of Modulus is that it can learn how entire networks of genes impact a cell’s function.

Bryce Daines, Ph.D., is co-founder of Modulus. 

“Combining high-throughput screening with machine learning to interpret and predict genetic interactions we get a multiplier on our experimental throughput that dwarfs previous approaches, repeated over cell types and diseases; the result is a flywheel of discovery,” Daines says.