Using AI and 'Extinct' Proteins to Battle Antimicrobial Resistance

Thursday - August 3 2023, 14:07 UTC - 1 year ago

The Machine Biology Group (MBG) at the University of Pennsylvania has used AI to explore the molecular diversity of extinct species in order to find novel drug therapies. MBG has found promising antimicrobial peptides from modern and ancient proteins. This breakthrough has led to the discovery of antibiotics from extinct species of humans, such as Neanderthals and Denisovans. While this field has exciting potential, it also brings up ethical considerations.

Antimicrobial resistance is one of the most pressing medical challenges of the modern era, and scientists are turning to the potential of long-extinct organisms for help. The Machine Biology Group (MBG) at the University of Pennsylvania has leveraged new AI tools to explore the molecular diversity of ancient organisms to find novel drug therapies we can use today, even turning to the "de-extinction" of a molecule found in extinct species of humans.

Antimicrobial resistance (AMR) is a pressing global issue that arises when microbes become resistant to antibiotics, thus increasing a host’s susceptibility to infectious diseases. Multiple factors contribute to AMR, such as antibiotic misuse and a lack of access to clean water, hygiene, and sanitation facilities. And, at present, AMR results in 700,000 fatalities annually and is predicted to claim ten million lives by 2050, according to MBG.

In looking for new therapies, MBG has turned to AI and a new machine-learning model called panCleave to search for antimicrobial peptides within both extinct and existing human proteins. The results of in vitro experiments have been promising, revealing the antimicrobial potential of both modern and archaic protein fragments. Further comprehensive assessment of these peptides, their action mechanism, resistance to breakdown, and efficacy as anti-infective agents in two preclinical mouse models—published last week in the journal Cell Host & Microbe—revealed certain peptides demonstrating stability, non-toxicity, and potent antimicrobial properties. These results indicate a potential pathway to the development of novel anti-infective agents that may not have been considered up until now.

This breakthrough has resulted in the successful discovery of antibiotics from extinct organisms, notably Neanderthals and Denisovans, which are two species of humans that have gone extinct but whose DNA is incorporated into our own to varying degrees. This new area of research, termed "molecular de-extinction" by the paper’s authors, focuses on reviving specific extinct molecules rather than resurrecting entire organisms, which comes with several practical and ethical concerns.

The approach utilizes advancements in machine learning, synthetic biology, and chemistry to discover, synthesize, and test extinct molecules in the lab, the researchers say. The convergence of these fields promises to open up uncharted territory in terms of molecular sequences, allowing researchers to delve into evolutionary history and unlock a wealth of potential therapeutic agents. Furthermore, this transformative approach helps us understand the significance of peptide sequences and their roles in immunity throughout evolutionary history.

"Our findings suggest that molecular de-extinction holds tremendous potential as a framework for antibacterial drug discovery," lead author César de la Fuente said.

Molecular de-extinction is not without problem, however. Even as the potential for these new therapeutics is certainly exciting, the authors acknowledge the ethical considerations of ("ab)using ancient genomes and molecular forms of life to serve our own invention and interests.