On the hunt for a better antibody, BigHat Biosciences debuts with $19M Series A to scale its 'wet lab' AI platform
Next-gen antibodies have become somewhat of a Holy Grail in the drug development world, but the process of actually creating those antibodies has proven to be difficult and time consuming. A brand-new California outfit is looking to solve that time crunch by bringing AI and machine learning into the lab.
BigHat Biosciences debuted Wednesday with a $19 million Series A round to help scale up its “wet lab” AI and machine learning platform to speed development of next-gen antibodies.
The San Carlos, CA team led by Google.ai and Broad institute veteran Mark DePristo and chief scientist Peyton Greenside, a Stanford grad with Broad bloodlines, are aiming to bring AI and machine learning into the experimental process in an attempt to dramatically increase the speed of antibody discovery.
“The real goal of our experimental platform is to enable us to go from a prediction to an in silico design through to a synthesized, purified and characterized antibody in a matter of days,” Greenside told Endpoints News. “What that means is there’s this core work cell that allows us to go from DNA to protein very quickly and have a very rapid characterization of that protein.”
What that looks like in practice is turning out hundreds of antibodies in very short order — an output that could play well in terms of future outlicensing deals with biopharma players. But in the short term, BigHat, which takes its rather nerdy moniker from the “hat” function in statistics denoting an estimated value, will work on developing and scaling its platform to give it even more predictive and computational power to produce more antibodies, DePristo said.
The biotech will also look to turn drugmaker, shepherding therapeutic programs through the clinic in the far future, DePristo said. In terms of what that could look like, DePristo said BigHat was looking more at design challenges for antibodies rather than specific therapeutic areas, but he did specifically call out autoimmune diseases, oncology and anti-infectives.
“Because our platform can optimize for pretty much any measurable property of a molecule — so not just affinity — we can optimize for biological function,” DePristo said. “We’re looking for programs where the biological effect is not just maximized by high affinity. The goal is not to permanently stick the cell to a molecule but to really activate the cell and get its biological function moving.”
With $19 million in hand from the Series A round led by VC firm Andreessen Horowitz and joined by 8VC, AME Cloud Ventures and Innovation Endeavors, BigHat will look to grow its team of nine employees, adding team members in the biology and computational spaces. The expanded team will add power to BigHat’s platform “10 to 100x,” DePristo said.
The company comes with a star-studded scientific advisory team, including Nobel laureate Brian Kobilka, AI and deep learning experts in Jake Gardner and Andrew Gordon Wilson, and a slew of experts in life sciences, entrepreneurship and drug discovery. That advisory team was specifically constructed to help BigHat not only scale its platform but prepare for the tough translational work that would be required to take its antibodies into the clinic, DePristo said.