A generative AI model that designs new antibiotics
A research team at Penn Engineering has developed and validated a new ‘diffusion model’ that can generate antibiotic candidates the same way AI creates images.
2 min. read
What if generative AI could design lifesaving antibiotics, not just art and text? In a new Cell Biomaterials paper, Penn researchers introduce AMP-Diffusion, a generative AI tool used to create tens of thousands of new antimicrobial peptides (AMPs)—short strings of amino acids, the building blocks of proteins—with bacteria-killing potential. In animal models, the most potent AMPs performed as well as FDA-approved drugs, without detectable adverse effects. “We’re leveraging the same AI algorithms that generate images, but augmenting them to design potent new molecules,” says Pranam Chatterjee, assistant professor in bioengineering and in computer and information science at the School of Engineering and Applied Science, and the paper’s senior co-author.
The lab of Pranam Chatterjee (pictured), in collaboration with the lab of César de la Fuente, developed and validated a new “diffusion model” that can generate antibiotic candidates the same way AI creates images.
(Image: Sylvia Zhang)
While some generative AI models like ChatGPT work by predicting the next word or element in a sequence, “diffusion” models start from random “noise” and iteratively refine it into a coherent output—the principle behind tools like DALL·E and Stable Diffusion.
AMP-Diffusion works the same way, only instead of “denoising” pixels, it refines sequences of amino acids. “It’s almost like adjusting the radio,” says César de la Fuente, Presidential Associate Professor in bioengineering and in chemical and biomolecular engineering at Penn Engineering, in psychiatry and microbiology in the Perelman School of Medicine, and in chemistry in the School of Arts & Sciences, and the paper’s other senior co-author. “You start with static, and then eventually the melody emerges.”
Because ESM-2, a widely-used protein language model, already has a rich “mental map” of how real proteins fit together, AMP-Diffusion doesn’t need to relearn basic biology. That means it can generate candidate AMPs faster, and its outputs are more likely to follow the intricate patterns that make peptides effective.
Chatterjee’s team also designed AMP-Diffusion to consult ESM-2’s built-in rules while “denoising,” essentially giving the new tool a coach that keeps it grounded in biological reality.
“Instead of teaching the model the ABCs of biology, we started with a fluent speaker,” says Chatterjee. “That shortcut lets us focus on designing peptides with a real shot at becoming drugs.”
Applying design to agriculture for a new approach to agroforestry
Penn researchers are leading a multidisciplinary land-use project that blends trees, livestock, and streamside buffers to turn New Bolton Center into a living lab for stewardship agriculture.
Future mRNA vaccines may prevent food and seasonal allergies
Early research from Penn Medicine finds a new mRNA vaccine stops allergens from causing immune reactions and life-threatening inflammation, with promise for future treatment for a variety of seasonal and food allergies.
Helping robots work together to explore the Moon and Mars
Penn Engineers, NASA, and five other universities tested robotic systems designed to help unmanned explorers cooperate in the dunes of White Sands, New Mexico, paving the way for Moon and Mars exploration.
Deepfakes, digital doubles, and the law: Jennifer Rothman on protecting identity in the AI era
The evolution of AI technology may require more robust copyright laws and guardrails that protect people’s control of their own identities, says the Penn Carey Law professor.
