The expanse of disease-causing proteins is fertile searching floor for drug analysis. Many biotech startups are looking for small molecules able to binding to those targets. That’s nice when it really works, however many protein targets stay elusive, mentioned Raphael Townsend, founder and CEO Atomic AI. Somewhat than pursue disease-causing proteins, Townsend’s startup goals for a special goal: the RNA that carry directions for making these proteins. Atomic AI makes use of synthetic intelligence to seek out methods to drug RNA and it’s now out of stealth backed by $35 million.
“By focusing on the RNA as a substitute, you’re giving your self new methods of focusing on these untreatable illnesses,” Townsend mentioned.
The Collection A spherical of financing introduced Wednesday was led by Playground World.
With a purpose to drug RNA, scientists must first get a greater understanding of it. Proteins are comparatively effectively understood, with a whole lot of hundreds of recognized protein buildings, Townsend mentioned. By comparability, the human transcriptome, the entire set of all RNA, is poorly understood. The a whole lot of recognized RNA buildings are much less effectively mapped out in comparison with proteins, Townsend mentioned. That’s key as a result of there’s rising recognition that RNA performs a serious position in illness by itself, he added.
Proteins fold and alter form, which might make them tough to hit with a small molecule. However RNA is considerably extra versatile making it extra of a transferring goal, Townsend mentioned. The know-how of San Francisco-based Atomic AI maps out the transcriptome with an strategy that mixes moist lab experiments with computational evaluation. Information generated by the moist lab are used to coach the AI to find novel targets on the three-dimensional construction of RNA, Townsend mentioned. The AI makes predictions that inform further moist lab experiments. These outcomes feed further AI evaluation, persevering with a virtuous cycle.
Atomic AI’s know-how relies on analysis stemming from Townsend’s PhD work at Stanford College. That analysis was printed within the journal Science in 2021, the identical 12 months Atomic AI fashioned. Since then, the corporate has made advances with its algorithms and its moist lab, Townsend mentioned. The know-how, now referred to as Platform for AI-driven RNA Construction Exploration (PARSE) has additionally improved in velocity and accuracy.
The brand new capital allows Atomic AI to scale the platform, enabling the startup to turn out to be a drug discovery group, Townsend mentioned. The corporate will start to slim down the targets it is going to pursue. Townsend declined to establish particular illnesses Atomic AI might pursue, however he mentioned the know-how might be used to find small molecules to be used in oncology, neurodegenerative problems, cardiology, uncommon illness, and infectious illness. The startup’s preliminary analysis will deal with figuring out the components of the transcriptome which are even targetable, Townsend mentioned.
Atomic AI isn’t the primary biotech aiming to drug RNA, and along with having an earlier begin a few of these startups have already got partnerships with massive pharma corporations. Essentially the most superior program of Arrakis Therapeutics is an oncology compound in lead optimization. The Waltham, Massachusetts-based firm has a drug discovery alliance with Roche. Skyhawk Therapeutics is one other Waltham-based firm growing RNA-targeting small molecules. That firm has alliances with Bristol Myers Squibb, Merck, and Takeda Pharmaceutical. Somewhat than straight goal RNA, Remix Therapeutics is growing medicine that focus on components of the cell that course of it. Practically a 12 months in the past, Cambridge, Massachusetts-based Remix inked a analysis alliance with a Johnson & Johnson subsidiary. Extra lately, Boulder, Colorado-based Arpeggio Biosciences unveiled a $17 million Collection A spherical of funding.
Townsend acknowledges the opposite corporations pursuing RNA-targeting small molecules, however he says what units Atomic AI aside is the moist lab part of its platform. Firms that take a purely AI strategy to RNA could have problem as a result of there’s simply not a lot RNA knowledge on the market for these applied sciences to investigate, he defined.
Now that Atomic AI is out of stealth, Townsend mentioned he’s in search of potential partnerships. Whereas the startup’s inner analysis will deal with growing small molecule medicine, Townsend mentioned partnerships will deal with utilizing PARSE to develop new RNA-based medicines. The platform’s means to foretell how RNA folds and types new buildings can be utilized to design new RNA medicines, he defined. The know-how additionally holds potential for bettering sure elements of RNA-based medicines, equivalent to stability. For instance, a extra secure RNA molecule might keep away from the ultra-cold storage required of the messenger RNA-based Covid-19 vaccines.
Atomic AI initially raised $7 million in a 2021 seed financing led by 8VC. That agency additionally invested within the Collection A spherical, which included the participation of Manufacturing unit HQ; Greylock; NotBoring; AME Cloud Ventures; and angel buyers together with former GitHub CEO Nat Friedman; Doug Mohr; Curai CEO Neal Khosla; and Patrick Hsu, a College of California, Berkeley professor, and Arc Institute co-founder.
Picture by libre de droit, Getty Photos
