Scientists are exploring the possibility of using AI to develop antibiotics with fewer side effects
In the new study, the scientists used explainable AI (XAI), a branch of AI that provides reasoning for model decisions, which is increasingly being used by scholars to investigate predictive AI models. The study was conducted by researchers at the University of Manitoba, Canada. Read on to learn more about the study.
Scientists are using AI to develop antibiotics with fewer side effects
A new study explores its possibilities artificial intelligence ,Hey) in better development AntibioticsWith accurate and efficient predictive models. The use of AI has grown tremendously, enabling applications such as content development, email proofreading, and driverless cars. However, in the new study, the scientists used explainable AI (XAI), a branch of AI that provides reasoning for model decisions, which is increasingly being used by scholars to investigate predictive AI models.
The findings will be presented at the American Chemical Society (ACS) Fall 2024 meeting to be held August 18-22. The study was conducted by researchers from the University of Manitoba, Canada.
Although XAI can be applied in a variety of contexts, the team used it to develop antibiotics. Despite the nearly ubiquitous use of AI, many of its models act as “black boxes,” obscuring the decision-making process. This can lead to mistrust, especially in critical areas such as drug discovery.
To overcome this, the team used XAI to train AI drug discovery models, particularly those that identify potential new antibiotic candidates. Predictive models are essential given the urgent need for efficient antibiotics in the wake of rising resistance.
“AI in chemistry and drug discovery is the way of the future. Someone has to lay the groundwork for it, and I feel like I’m doing that,” said Hunter Sturm, a graduate student at the university. To predict biological effects, scientists fed drug chemical databases into an AI model. An XAI model was then used to examine the exact molecular properties behind these predictions.
Interestingly, XAI has discovered elements that human chemists would not have discovered, such as the fact that non-core structures are more important than the core in penicillin compounds. To test predicted antibiotic compounds, researchers collaborate with microbiology laboratories to improve AI models through the application of XAI insights. “AI creates a lot of mistrust,” said Rebecca Davis, a professor of chemistry at the University of Manitoba in Canada.
“Still, there is a greater chance that this technology will be accepted if we can seek clarification from the AI itself,” he added.
(With inputs from IANS)
Get the latest news on Times Now as well as breaking news and top headlines from across health and the world.