Snakebite Crisis: Can AI Transform The Century-Old Approach To Antivenoms?

Researchers have used AI to design proteins that neutralize key toxins in snake venom, offering a potential pathway to more effective and accessible snakebite treatments.

The study, published in Nature, highlights how AI can pave the way for new, more accessible snakebite treatments, potentially saving thousands of lives each year.

Nature reports that snakebites kill an estimated 100,000 people annually and cause permanent disability for many more, particularly in low-resource areas. The World Health Organisation defines snakesbites as a “neglected public health issue in many tropical and subtropical countries.’’

Current treatments, primarily derived from antibodies in the blood serum of immunized animals like horses and sheep, have remained largely unchanged for over a century. These antivenoms often require refrigeration and trained medical professionals to administer, limiting their utility in rural or underserved regions, as noted by Nature.

The new approach utilizes an AI tool called RFdiffusion, developed by David Baker’s team at the University of Washington. Inspired by image-generating AI programs like DALL-E, RFdiffusion can design proteins capable of binding tightly to specific toxins.

In collaboration with biochemist Susana Vázquez Torres, the team targeted three types of toxins in elapid snake venom, known to cause paralysis, tissue damage, and death.

Using RFdiffusion, the researchers created “mini-binders” that strongly attached to these venom toxins. Experiments showed that these mini-binders could neutralize toxins’ effects in lab-grown cells.

In live animal tests, the team demonstrated their efficacy: mice injected with otherwise lethal doses of venom survived when treated with mini-binders either beforehand or shortly after exposure. “This is probably the coolest experimental result I’ve had in my career so far,” said Vázquez Torres.

The AI-designed proteins boast advantages over traditional antivenoms. They are highly stable and could eliminate the need for refrigeration, making them ideal for remote settings. Additionally, they can be mass-produced at low cost using industrial bacteria.

However, these proteins address only a fraction of snake venom’s complex composition, meaning future antivenoms may need to combine multiple mini-binders tailored to regional snake species.

Despite promising results, challenges remain in funding and clinical development. While Baker’s protein design ventures for diseases like cancer have secured substantial investments, resources for neglected diseases like snakebites are scarce.

“The path forward for anything to do with infectious disease or developing-world diseases like snakebites, it’s just harder,” Nature reports Baker saying. Still, this groundbreaking research signals a transformative shift in tackling global health challenges with AI.