AI for Antibiotic Discovery

Would you like to regularly receive event updates from the Max Planck Institute for Infection Biology? Subscribe to our Newsletter!

Computers excel at superhuman pattern recognition in images and text; however, their application in biology and medicine is still in its infancy. In this talk, I will discuss our advances over the past decade, which are accelerating discoveries in the crucial and underinvested area of antibiotic discovery. We have pioneered the design of antibiotics using artificial intelligence (AI), achieving proven efficacy in preclinical animal models and demonstrating that machines can effectively create therapeutic molecules. For the first time, we successfully mined the human proteome to identify antibiotic candidates. Building on this success, we hypothesized that similar compounds could be found throughout evolution. We expanded our efforts to extinct species, where our AI-driven approach led to the discovery of the first therapeutic molecules from organisms such as Neanderthals and the woolly mammoth. This work launched the field of molecular de-extinction and yielded preclinical candidates such as neanderthalin, mammuthusin, and elephasin. Furthermore, my lab has broadened our antibiotic discovery initiatives to explore other branches of the tree of life beyond eukaryotes. By computationally analyzing microbial dark matter, we identified nearly one million new antibiotic molecules. These molecules have been made freely available and open access to the scientific community to encourage researchers worldwide to synthesize, characterize, and further develop them. This collaborative effort leveraged machine learning to explore the vast diversity of the microbial world by analyzing 63,410 metagenomes and 87,920 microbial genomes. Additionally, through the computational exploration of thousands of human microbiomes, we and our collaborators discovered a myriad of new antimicrobial agents, including prevotellin-2 produced by the gut microbe Prevotella copri. Collectively, our efforts have dramatically accelerated antibiotic discovery, reducing the time required to identify preclinical candidates from years to just a few hours. I believe we are on the cusp of a new era in science where advances enabled by AI will help control antibiotic resistance, infectious disease outbreaks, and pandemics.

Presenting author details
Full name: Prof. Cesar de la Fuente

Title: Presidential Associate Professor

Affiliation: University of Pennsylvania

Contact number: +1-215-746-6083

Website: https://delafuentelab.seas.upenn.edu/

Twitter account: @delafuenteupenn

LinkedIn account: https://www.linkedin.com/in/cesardelafuentenunez/