What happened
Biotech startup Enveda has reached a $2 billion valuation following $461 million in recent funding, driven by its use of AI to accelerate the development of natural product-based drugs. Simultaneously, researchers at the Korea Institute of Science and Technology (KIST) are utilizing proprietary AI models to map the therapeutic potential of natural compounds, such as ginsenosides, to identify new drug candidates.
Enveda, a U.S.-based biotech firm, has raised $461 million in combined Series D and E funding, bringing its total funding to over $845 million since 2019. The company’s valuation has reached $2 billion, doubling in one year. Enveda attributes this growth to its AI-driven R&D platform, which has enabled the development of 12 drug candidates at four times the speed and one-tenth the cost of industry averages.
At the Korea Institute of Science and Technology (KIST), researchers are employing an in-house AI model called ECBS (evolutionary compound binding similarity). This model pre-trains on chemical structures to predict how natural compounds fit into human protein targets. KIST researchers have successfully used this approach to identify how ginseng-derived ginsenosides suppress breast cancer cell growth via calcium pump signaling.
The broader scientific community is also seeing results from AI-assisted discovery. A 2024 study from the University of Pennsylvania, published in Cell, demonstrated that AI could identify hundreds of thousands of antibiotic-potential peptides from microbial genomes, with a high success rate in experimental validation.
Source details: biz.chosun.com
Why it matters
The integration of AI into natural product research addresses a historical bottleneck: the difficulty of identifying how specific natural compounds interact with human biological targets. By using AI to predict these mechanisms, researchers can bypass years of trial-and-error, potentially reviving the use of natural sources—which have historically provided foundational medicines like aspirin and penicillin—for modern drug development. This shift is economically significant, as evidenced by Enveda’s rapid valuation growth and the ability to develop candidates at a fraction of the industry-standard cost and time.
Historically, natural product research stalled because identifying the active ingredient was only the first step; determining the mechanism of action was often prohibitively difficult and time-consuming. AI changes this by predicting efficacy and binding affinity before physical synthesis begins.
The economic impact is substantial. By reducing the time and expense required to move from a natural extract to a clinical candidate, AI makes natural product research more viable for pharmaceutical companies that had previously moved away from the field.
The potential for 'global blockbusters' is high, particularly in high-demand areas like dermatology and obesity management, where Enveda’s candidates are currently being tested.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
crm_get_transaction(id='4092').Which component of an AI application is the machine-learning model itself?
What to watch next
The industry is currently monitoring the progress of AI-discovered candidates in human clinical trials, such as Enveda’s ENV-294 and ENV-308. Additionally, the field faces a critical need to digitize traditional medicinal knowledge and share 'failed' experimental data, which are currently underutilized resources that could further improve the predictive accuracy of AI models in drug discovery.
The primary milestone to watch is the outcome of human clinical trials for AI-discovered natural product drugs. While AI has proven its ability to identify candidates, no such drug has yet received full marketing approval.
Data availability remains a hurdle. Researchers emphasize the need for a culture of sharing 'failed' experimental results, which are currently rarely published but are essential for training more robust AI models.
There is an urgent push to digitize traditional medicinal knowledge from Indigenous languages and cultures before this information is lost to language extinction and environmental degradation, as noted by researchers at the University of Zurich.