Chai Discovery is an exciting new startup developing AI models for drug discovery and design. The company raised $70 million in a series A and will use the funding to build on the functionality of its Chai platform, specifically concerning targets that have been difficult to drug.
The round was led by Menlo Ventures and Chai Discovery, attracting investment from groups such as Yuri Milner’s DST Global Partners, SV Angel, and Yosemite. A few existing backers also chipped in, including Thrive Capital, OpenAI, and Dimension.
This financing comes just a month after Chai Discovery debuted its latest model, Chai-2, which can design antibodies from scratch, with a near 20% hit rate. This greatly surpasses the 0.1% observed in other computational models, which typically screen millions to billions of antibodies to find hits.
Matthew McParlton, Co-founder of Chai Discovery, said: “Before Chai-2, the process was not unlike searching a giant bunch of keys for the right fit for a lock — but there are millions of keys. Now, it’s like having a master locksmith design exactly the right shape key, based only on your description of the lock.”
Even when just the target antigen and epitope are input to Chai-2, the model can use this information to generate accurate binder hits across a range of targets. The company envisions that Chai-2 will be able to design de novo antibodies that can precisely target disease-causing proteins.
Chai is aiming to take on more established rivals such as Isomorphic Labs, the drug discovery start-up spun out of Google’s DeepMind AI unit. Overall, Chai could flip the script on antibody design and ultimately play a key role in creating better therapeutics faster.
