Did you miss a session from the Future of Work Summit? Head over to our Future of Work Summit on-demand library to stream.
In a new study published in the journal Nature Biomedical Engineering, researchers at IBM say they’ve developed an AI model that can assist in the rapid design of antimicrobial peptides — the building blocks of proteins. The researchers say that the model outperforms other AI methods at designing such peptides and increases the success rate of identifying a viable candidate by 10%.
Antibiotics have transformed the world of medicine over the past century or so, but they’ve also been overused, leading to the emergence of bacteria with powerful resistance. According to the Centers for Disease Control and Prevention (CDC), antibiotic resistance is one of the biggest public health challenges of our time. In fact, in the U.S. alone, nearly 3 million people die annually as a result of antibiotic-resistant infections.
Unfortunately, few new antibiotics are being developed to replace those that no longer work, in part because drug design is an extremely difficult, lengthy, and capital-intensive process. IBM’s proposed solution is generative modeling, a subfield of AI that allows researchers to decide upfront what characteristics they want peptides to have versus guessing combinations.
Historically, material design of molecules, proteins, and altogether new peptides has been a complex simulation problem. Even small molecules made of only a few atoms have hundreds of possible combinations. To combat this, IBM’s AI model pulls from a large dataset to reverse-engineer a peptide’s design and produce the desired peptide framework. Effectively, it shortens the time needed to create high-quality peptide candidates from years to potentially days while increasing the likelihood of identifying successful candidates to fight antibiotic drug resistance.
Within 48 days, IBM says its AI-boosted molecular design approach enabled it to identify, synthesize, and experimentally test 20 AI-generated novel candidate antimicrobial peptides. Two of them turned out to be potent against pathogens, very unlikely to trigger drug resistance in E. coli, and had low toxicity when tested both in vitro and in mice.
Beyond antibiotics, IBM says the generative AI system could potentially accelerate the design process of molecules for new drugs and materials. “Our proposed approach could potentially lead to faster and more efficient discovery of potent and selective broad-spectrum antimicrobials to keep antibiotic-resistant bacteria at bay — for good,” IBM’s Saska Mojsilovic and Payel Das wrote in a blog post. “And we hope that our AI could also be used to help address the world’s other most difficult discovery challenges, such as designing new therapeutics, environmentally friendly and sustainable photoresists, new catalysts for more efficient carbon capture, and so much more.”
IBM’s latest work builds on an earlier study published in the journal Advanced Science by the company’s researchers. It demonstrated a technique that enabled the coauthors to create up to 100 bacteria-fighting polymers in nine minutes, using AI and machine learning.
VentureBeatVentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact. Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:
- up-to-date information on the subjects of interest to you
- our newsletters
- gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
- networking features, and more