Making Metabolomics Data More Useful in the Study of Health Risks

Image

AMHERST, Mass. – The National Institutes of Health (NIH) has awarded University of Massachusetts Amherst biostatistician Raji Balasubramanian a four-year, $1.37 million grant to advance the analysis of data from metabolomics research. These studies examine at the molecular level the links between metabolic action and the risk of chronic health conditions, from heart disease and stroke to breast cancer and other complex disorders.

Balasubramanian, associate professor of biostatistics and epidemiology in the School of Public Health and Health Sciences, will co-lead an interdisciplinary team in the development and application of new statistical models to more effectively mine the expanding base of metabolomics data sources. The other lead investigator is Denise Scholtens, chief of biostatistics and director of the Northwestern University Data Analysis and Coordinating Center.

Metabolomics focuses on the study of the universe of small molecule metabolites involved in the body’s chemical reactions, as measured in blood, urine, saliva and other fluids.

“It provides a view of our biochemical state, aggregating the influences of genetic and environmental factors,” Balasubramanian explains. “Our small molecule metabolites are influenced not just by our genetic profile but also by what we eat, how much we exercise, the air we breathe and so on.”

In recent years, research into the metabolome has grown exponentially in population-based studies, and processing the voluminous data has created methodological challenges for researchers. Balasubramanian’s NIH grant will fund work to address those challenges.

The first goal, she says, is to come up with a statistical approach for variable selection specially suited to the characteristics of data generated in metabolomics studies. “We want to do this in a way that incorporates the fact that metabolites have dependence on each other and co-act in various known biological pathways,” Balasubramanian says.

Secondly, the researchers also will develop methods to identify differential networks of metabolites, where differing patterns of connectivity between exposure or phenotype groups can shed light on the underlying biology.

“In many studies, we view metabolites as mediating the relationship between an exposure and the outcome,” Balasubramanian says. “We want to develop models that can identify not just individual metabolites that act as mediators but networks of metabolites that can act as mediators.”

The network-based approaches for data analysis that Balasubramanian is working to develop will provide new tools that can be broadly applied in metabolomics studies of chronic disorders in human populations.

In her lab, Balasubramanian is currently conducting metabolomics research in collaboration with medical investigators from Harvard Medical School in other NIH-funded studies to understand the molecular underpinnings of common conditions, including stress and post-traumatic stress disorder, as well as cardiovascular disease, stroke, breast cancer and kidney disease.