Explainable AI could help scientists design better antibiotics

By Amit Malewar 19 Aug, 2024

Collected at: https://www.techexplorist.com/explainable-ai-could-help-scientists-design-better-antibiotics/87366/

Artificial intelligence (AI) has gained tremendous popularity, revolutionizing various aspects of our lives, from driving vehicles to designing new medications. Despite its incredible capabilities, understanding AI’s decisions can be challenging.

Explainable AI (XAI) offers a solution by providing justifications for AI’s decisions, allowing for greater transparency. Researchers are now leveraging XAI to scrutinize predictive AI models more closely and gain deeper insights into the field of chemistry.

These groundbreaking findings by researchers at the University of Manitoba will be presented at the upcoming fall meeting of the American Chemical Society (ACS).

AI has become incredibly prevalent in today’s technology, but many AI models operate as black boxes, making it difficult to understand how they arrive at their results. This lack of transparency can lead to skepticism, especially when it comes to important tasks such as identifying potential drug molecules.

“As scientists, we like justification,” explains Rebecca Davis, a chemistry professor at the University of Manitoba. “If we can come up with models that help provide some insight into how AI makes its decisions, it could potentially make scientists more comfortable with these methodologies.”

Explainable AI (XAI) can address this issue by providing visibility into the decision-making process of AI models. Researchers like Davis are particularly interested in applying XAI to AI models for drug discovery, especially for predicting new antibiotic candidates. With thousands of molecules needing to be screened to find just one viable drug and the ongoing threat of antibiotic resistance, accurate and efficient prediction models are essential.

“I want to use XAI to better understand what information we need to teach computer chemistry,” says Hunter Sturm, a graduate student in chemistry in Davis’ lab who’s presenting the work at the meeting.

The researchers began by utilizing AI to analyze databases of known drug molecules and predict their biological effects. They then employed an XAI model to identify the specific components of the drug molecules that influenced the predictions. This allowed for a better understanding of the AI model’s criteria for determining molecule activity. The researchers discovered that XAI can identify nuances that may be overlooked by humans and process a greater amount of data simultaneously. For instance, when examining penicillin molecules, the XAI uncovered intriguing information.

“Many chemists think of penicillin’s core as the critical site for antibiotic activity,” says Davis. “But that’s not what the XAI saw.” Instead, it identified structures attached to that core as the critical factor in its classification, not the core itself. “This might be why some penicillin derivatives with that core show poor biological activity,” explains Davis.

The researchers aim to leverage eXplainable AI (XAI) to enhance predictive AI models and guide the creation of new antibiotic compounds. By understanding how computer algorithms prioritize molecular structures for antibiotic activity, they can train AI models to better identify effective compounds.

Partnering with a microbiology lab, the team will synthesize and test the compounds identified by the improved AI models. Ultimately, they hope that XAI will enable the development of more effective antibiotic compounds to tackle antibiotic-resistant pathogens.

By using AI to explain its decision-making processes, the researchers believe that it can help build trust and acceptance for this technology. They view AI applications in chemistry and drug discovery as the future of the field and aim to pave the way for its widespread use.