For centuries, people chewed willow tree bark to relieve pain, but scientists at chemical firm Bayer didn’t isolate its active ingredient until the 1800s and eventually patented its modified version as Aspirin.
Aspirin is just one example of a medicine derived from natural sources. In fact, the World Health Organization estimates that around 40% of modern pharmaceutical products have roots in remedies used by our ancestors.
Even with this impressive success of harnessing nature’s bounty, scientists estimate that they have discovered only a tiny fraction of natural chemical compounds that could be developed into powerful medicines.
In part that’s because identifying, isolating and testing molecules from nature is complex and more time-consuming than synthesizing new compounds in a lab.
Viswa Colluru, an early employee of Recursion Pharmaceuticals, which went public in 2021, decided that AI and other techniques can expedite the process of discovering new medicines from nature.
In 2019, Colluru left Recursion to start Enveda Biosciences, a Boulder, Colorado-based biotech that analyzes plant chemistry to unearth potential medicines.
Colluru told TechCrunch that Enveda tapped all of the world’s digital information about how humans across cultures have used plants to cure pain and disease.
“We discovered that geographically separated cultures from across the world were much more likely to use similar plants for similar diseases and symptoms, even though they never talked to each other,” he said. “They discovered that a certain plant helps stomach ache, or a certain plant helps like a fever or a headache, and that is literally thousands of years of experiential human wisdom.”
Today, the company’s database has 38,000 medicinal plants linked to about 12,000 diseases and symptoms.
Once Enveda’s AI identifies plants with the highest likelihood of providing cures, team members gather the materials and test them using the company’s laboratory and AI model. Unlike traditional methods for studying individual molecules, Enveda’s transformer model can decipher the “chemical language” of the entire sample.
“Once we know their shape, we can prioritize the right sets of molecules and say, this will one day be a medicine,” Colluru said.
Enveda’s approach is starting to bear fruit. Two of the company’s drugs — one for treating skin conditions, including eczema and the other for inflammatory bowel diseases — are expected to begin clinical trials later this year, according to Colluru.
The company’s scientific progress has attracted the attention of investors. On Thursday, Enveda announced that it has raised a $55 million Series B2 from new investors, including Microsoft, The Nature Conservancy, Premji Invest and Lingotto Investment Fund, and existing backers Kinnevik, True Ventures, FPV, Level Ventures and Jazz Venture Partners. The fresh funding brings the company’s total capital to $230 million.
The new round allows Enveda to add long-term strategic partners to its cap table, and the company plans to raise a Series C later this year after the start of clinical trials, Colluru said.
Microsoft is also providing some cloud credits as part of the deal, but this is separate from its cash investment, according to Colluru.
While sampling plants to find medicines is an age-old approach, Enveda is one of the few companies doing this with AI’s help. U.K.-based Pangea Bio is also studying plants to discover drugs for treating neurological conditions.
Of course, much of the attention in this field has gone to marijuana and the natural sources best known for having produced psilocybin in so-called “magic mushrooms” or other psychedelics that have the potential to cure mental health disorders, but Enveda is not interested in studying their compounds.
“Everybody is focused on cannabis and psychedelics, which are just a tiny fraction of the natural world,” Colluru said. “The natural world is so rich in its chemical diversity and biological effects that studying just 100 plants is enough to give so many potential drugs that we don’t know what to do with them.”
Comment