Tingting Zhao, Ph.D., has dedicated her research to developing new tools to process the world around us. By applying that skillset to the health sciences, she’s finding ways to use cutting edge data science techniques to aid others. “This is a topic that I am really passionate about, the idea that something I was able to work on might make life easier for patients someday,” says the assistant professor of Information Systems and Analytics and a faculty fellow with the School of Health and Behavioral Sciences at Bryant University.
Her recent publication, “Identification of significant gene expression changes in multiple perturbation experiments using knockoffs,” published in the scientific journal Briefings in Bioinformatics, offers insight into how studies in biology and data science can augment one another to great effect. Using machine learning — a branch of artificial intelligence and computer science that focuses on the use of data and algorithms to imitate the way that humans learn — Zhao and her collaborators developed and honed algorithms that can help identify how genes respond to specific stressor stimuli. They then compared those reactions to the effects of various “small molecule” drugs.
Zhao was the lead author for the paper. Her co-authors include Harsh Vardhan Dubey, a Ph.D. student in Statistics at the University of Massachusetts Amherst; Guangyu Zhu, an assistant professor in the Department of Computer Science and Statistics at the University of Rhode Island; and Patrick Flaherty, associate professor of Mathematics and Statistics at the University of Massachusetts Amherst.
“This is a very fast evolving field, and there is a lot of competition because people all over the world want to make the next breakthrough.”
The information they discovered through their work can help us develop a more detailed understanding of the molecular pathways that respond to genetic and environmental changes as well as the underlying mechanisms of disease. Zhao is particularly excited for the potential the algorithms have regarding drug repurposing. “Creating and approving a new drug takes a great deal of work and costs a great deal of money,” Zhao explains. “We want to explore if we can use cheap, previously existing, drugs for new purposes, including being used in place of more expensive drugs.”
The study is part of the growing field of bioinformatics, an interdisciplinary space that involves the development of methods and tools for understanding biological data, with applications to the understanding of health, disease, and medical care. That health-based focus, she says, adds a greater purpose to her data science work. “We are not just developing algorithms in a vacuum,” says Zhao. “We develop these techniques to solve a real problem and that problem, which should always motivate people doing this work, is ‘How can we help people?’”
The bioinformatic field’s rise, she says, speaks to a larger embrace of computational analysis within the health sciences. “People’s opinions have changed over time. At the start, there was some skepticism of this sort of research because it was not just biology, but very computational as well,” says Zhao. “But over time people became more comfortable with it because they came to realize how helpful these tools could be.”
Today, bioinformatics is a hotbed for research. “This is a very fast evolving field, and there is a lot of competition because people all over the world want to make the next breakthrough,” says Zhao, who notes that this competition is part of what draws her to the field. “There is a thrill in knowing that I could create an algorithm that could beat all of the others.”
But while healthy competition can be a spur for innovation, Zhao states, in the end, collaboration is the real goal. “You always hope that others can use your research and make their own improvements and help more people,” she reflects.