Oral biology postdoc to examine periodontal disease through AI lens

BUFFALO, N.Y. — A University at Buffalo researcher will employ artificial intelligence (AI) to better understand periodontal disease development and progression.

Lu Li, a postdoctoral associate in the Department of Oral Biology in the School of Dental Medicine, was recently awarded a $993,098 Pathway to Independence Award (K99/R00) from the National Institutes of Health. It funds a five-year study that will utilize an AI subset known as machine learning to study bacterial communities in the mouth collected from more than 2,600 individuals, some of whom were followed over a five-year period.

“This work will uncover novel patterns in oral microbial ecology, improve our ability to predict periodontitis progression, and lay the groundwork for personalized prevention and treatment strategies,” explains Li, who works in the laboratory of his mentor, Patricia Diaz, Sunstar Robert J. Genco Endowed Chair, Empire Innovation Professor of Oral Biology and director of the UB Microbiome Center.

K99/R00 awards are designed for promising postdoctoral scientists seeking to complete mentored research that will facilitate their transition to a tenure-track or equivalent faculty position.

“It’s a prestigious grant,” Diaz says. “Lu is the first post-doctoral associate in the dental school to receive it.”

Li is using manifold learning, a type of advanced machine learning that turns complex, high-dimensional data into a simpler form while keeping important patterns and features to help construct microbiome landscapes. They will serve as a type of map showing different types of bacteria linked to periodontitis in various individuals and help predict the ones who will get worse over time.

“The goal is to identify distinct microbial states linked to periodontitis phenotypes,” he says. “I will then examine how these states change over time and predict disease progression using five-year longitudinal data.”

The other members of Li’s mentorship committee are Jean Wactawski-Wende, PhD, SUNY Distinguished Professor in the Department of Epidemiology and Environmental Health and dean of the School of Public Health, and Michael Buck, PhD, professor and director of Genetics, Genomics and Bioinformatics in the Department of Biochemistry in the Jacobs School of Medicine and Biomedical Sciences.

Work under mentors, then independently

The grant is divided into two phases.

In the first phase, Li will work with his mentors to study dental plaque samples of participants in two major projects: the Buffalo Myocardial Infarction (MI) Perio Study and the Buffalo OsteoPerio Study.

The former, led by the late UB dental researcher Robert Genco, SUNY Distinguished Professor of Oral Biology, Periodontics and Microbiology, studied the connection between periodontal disease and recurring cardiovascular events. The latter, led by Wactawski-Wende and Genco, studied the association between osteoporosis and periodontal disease in older women, following  individuals over five years.

The second phase of the grant is focused on independent research, and Li says he plans to examine oral bacteria at the strain level.

“This will allow me to build a strain-level landscape, revealing strain-specific genetic and functional features associated with disease severity,” he says. “Ultimately, this could lead to more precise ways of predicting which patients are at higher risk and open the door to personalized strategies for preventing and treating periodontitis.”

Drawn to study of microbiome

Li joined Diaz’s lab shortly after earning his PhD in computer science from UB in 2021. He says he was drawn to study of the microbiome.

“This opportunity opened a new door for me to apply my computer science techniques into such an amazing field to help solve real-world problems and improve human health,” he says.

Although Li’s doctoral training wasn’t in biology, Diaz says he expresses interest in everything he does and works diligently.

“Over a short time, Lu has acquired a very broad knowledge of microbiology and oral diseases, especially periodontitis,” she says. “He understands the minutia of how we define it and how we measure disease progression.”

Digging deeper into data

Li says that many of the advances made in machine learning have not yet been applied to biomedical fields, and the tools that do exist are often not tailored to the unique challenges inherent in microbiome research.

“I see my role as bridging this gap by developing advanced, customized algorithms that can help us understand these complex data in ways that were not possible before,” he explains. “With this grant, we now have the opportunity to dig deeper into the data and create new tools that can take our understanding to the next level.”

Diaz says she and the other mentors will be available to advise Li throughout the five years of the grant, including the independent study. They want to help him reach a key goal of the grant: to develop and hone all the skills he needs to secure an independent research position.

She says even now, he’s well on his way.

“Lu is an amazing trainee and a really outstanding person,” Diaz says. “To see him develop like this is really a satisfying moment in my career.”