The past is the key: tackling antibiotic resistance with molecular de-extinction

In our latest Lexicon episode , we sit down with Professor César de la Fuente , a pioneer in molecular de-extinction, and we explore how ancient biology and modern technology can be united to tackle one of the most pressing challenges of our time: antibiotic resistance.

As de la Fuente explained, the rise of antimicrobial resistance (AMR) is a major global health threat. “Antimicrobial resistance is… one of the greatest and existential threats facing humanity,” he said, noting that it’s currently responsible for 5 million deaths each year.

If left unchecked, this figure could rise to 10 million annually by 2050. Let’s find out how. Also, check out our latest merchandise at Interesting Engineering Shop .

Molecular de-extinction is the key to the future

Molecular de-extinction , as de la Fuente explained, is the process of resurrecting ancient molecules that have not existed in nature for thousands or even millions of years. These molecules, buried in the genetic code of extinct organisms, could offer novel insights into biological functions and potential medical applications.

As de la Fuente explained to IE, his team uses deep learning algorithms to mine proteomes for these ancient molecules. Proteomes are the complete sets of proteins produced by an organism.

The theory goes that we could reintroduce tried-and-tested means of fighting pathogens in the present day by finding older pathogen-fighting sections of DNA (peptides). They worked in the past, and existing pathogens are likely not exposed to them, so they prove potent again today.

“We believe that by comparing molecules throughout evolutionary history, you can first learn more about our past, our present, and perhaps help predict the future,” de la Fuente explained. “Bringing back molecules from the past may give us a better chance at addressing present-day problems,” he added.

As de la Fuente explained, his team has successfully tested several ancient molecules in preclinical mouse subjects so far. This has showed their potential to combat modern pathogens resistant to existing antibiotics.

Back to the future

When asked how they resurrect these molecules, de la Fuente emphasized that the work builds on decades of foundational science. “We stand on the shoulders of giants,” he said, referencing previous researchers who developed methods for sequencing ancient DNA.

One of the most notable advancements in this area came from Svante Pääbo, who won a Nobel Prize for sequencing archaic DNA samples . With access to this wealth of ancient genetic data, de la Fuente’s lab developed algorithms to mine these sequences for antibiotic peptides.

“We developed AI methods… to find molecules with antibiotic properties hidden within that information,” he said. These models can then predict which molecules are likely to have antimicrobial properties.

Once identified, his team synthesizes those molecules in the lab using semi-automated robots. According to de la Fuente, what’s remarkable is the speed at which these discoveries can now be made.

“I didn’t even have time to go for a coffee, and the algorithm had completed the exploration within about one hour,” he said, recalling a breakthrough moment.

As you can imagine, this is a significant leap from traditional antibiotic discovery methods, which can take years to yield a promising candidate. With AI, the process has been accelerated dramatically.

De-extinction: opening the genetic gifts from our ancestors

One of the most exciting discoveries from de la Fuente’s lab was a molecule called neanderthalin-1 . As the name suggests, this is a peptide found in the DNA of Neanderthals, whose DNA is found in most of European or Asian descent.

When tested in preclinical mouse test subjects, this peptide showed impressive anti-infective properties, potentially opening the door to a new class of antibiotics. Another, called mammuthosin (derived from the woolly mammoth), has also shown promise in combating infections.

As de la Fuente explained, both compounds were found using AI to sift through ancient genetic data. He also explained the significance of these discoveries: “The modern compounds target the outer membrane of bacteria, whereas the ancient ones tend to go after the cytoplasmic or inner membrane. ”

This distinction is important because it offers new strategies for fighting bacteria that have developed resistance to traditional antibiotics. “By comparing molecules over time, we can unlock new biological insights,” de la Fuente noted.

This approach generates new questions for further research, such as why ancient molecules target different bacterial structures than modern ones.

A long road ahead

Despite these breakthroughs, de la Fuente cautioned that the road to real-world application is a long one. While AI has vastly accelerated the discovery process, moving a drug from the lab to clinical trials and eventually to patients remains a significant challenge.

“It takes around ten years from the moment you discover something to the time that it can impact patients,” he explained. However, de la Fuente is optimistic about the future of AI-driven antibiotic discovery.

In the past, traditional methods took years to identify potential drug candidates. Now, thanks to AI, the process can be completed in hours. “We can discover hundreds of thousands of candidates,” de la Fuente said.

Looking forward, de la Fuente believes that AI and machine learning will continue to play a major role in molecular biology and medicine. He envisions a future where AI can predict how pathogens will evolve and design antibiotics to combat them in real-time.

“Imagine if you could program a molecule on a computer that targets a specific infection, and within hours, you have a personalized treatment,” he said. While this remains in science fiction, for now, it represents the ultimate goal of personalized medicine.

Man and machine working for the better good

De la Fuente also emphasized the importance of collaboration between humans and machines. “I believe it’s going to be a combination between human ingenuity and machine intelligence,” he said.

This blend of skills will be essential for navigating the complex and evolving landscape of antimicrobial resistance. Collaboration extends beyond the lab, too.

De la Fuente’s discoveries have sparked discussions in bioethics and patent law, raising questions about the ethical implications of resurrecting molecules that have been extinct for millennia. “Is it okay for us to synthesize these molecules in the lab?” he asked.

These are questions that scientists, ethicists, and legal experts must tackle as molecular de-extinction continues to evolve.

As antibiotic resistance continues to rise, these discoveries offer hope for the future. “We tend to think we know everything about ourselves, but here we are with a simple algorithm, finding thousands of new things,” de la Fuente said. And in those thousands of discoveries may lie the key to defeating the superbugs of tomorrow.