
In a breakthrough that marries the complexity of biology with the precision of machine learning, Xinyi Zhang, a fourth-year PhD student at MIT, is revolutionizing the way scientists analyze cellular data. As MIT News reports, Zhang has spearheaded efforts to build computational frameworks capable of tackling multimodal biological data. By integrating various measurements from cells, including type, gene expression, and location within tissue, Zhang's methods are poised to enhance our understanding of diseases like Alzheimer's.
While working under the guidance of Professor Caroline Uhler in the Laboratory for Information and Decision Systems, Zhang has led the development of these new tools for data analysis. Working with modal data that are multimodal saw them facing the challenge of how to interpret simultaneously different kinds of measurements, which scientists previously could only tackle one at a time. Zhang's process begins with an autoencoder designed in reverse, expanding rather than shrinking data dimensions, allowing for an integration of various data forms, and teasing out meaningful biological differences from technical variations.
Discussing her motivations, Zhang stated in a statement obtained by MIT News, "There are still many unanswered questions," and "I want to pick questions that are biologically meaningful, that help us understand things we didn’t know before." One of her key projects demonstrated this technique, abbreviated as STACI, in identifying tissue and cellular changes indicative of Alzheimer's progression using a fusion of spatial and imaging technologies.
Zhang's pioneering work extends beyond the lab as well. Boasting an array of hobbies that include everything from rock climbing to earning her pilot's license in November 2022, her advisor Caroline Uhler points out Zhang's humility and passion for learning, "Every time, you learn something like, ‘Okay, so now she’s learning to fly,’" Uhler told MIT News. "It’s just amazing. Anything she does, she does for the right reasons."
With a background in bioengineering and a keen eye for solving intense biological puzzles, Zhang's research is, without a doubt, carving new paths in the scientific community. Whether she's analyzing the frontal cortex to understand neurodegeneration or predicting protein images from sequences, Zhang's work has already contributed to the Nature Communications paper, unraveling some of the intricacies involved in our biological makeup—and there's a promise of more to come.