In the rapidly evolving field of life sciences, researchers are inundated with data from advanced imaging and genomics technologies. Xinyi Zhang, a fourth-year PhD student at the Massachusetts Institute of Technology, is at the forefront of this challenge. Working under the guidance of Professor Caroline Uhler in the Department of Electrical Engineering and Computer Science, Zhang is developing innovative computational tools to analyze multimodal data, which combines various types of measurements from biological samples. This endeavor is crucial for understanding the complexities of cellular behavior, particularly in the context of diseases like Alzheimer’s.
Zhang’s journey into the world of biology began in her high school years in Hangzhou, China. She found the subject intriguing, especially when her teachers struggled to answer her questions. This curiosity led her to pursue bioengineering at the University of California, Berkeley, where she also studied electrical engineering and computer science. After graduating in 2020, she joined MIT’s PhD program, ready to tackle the pressing questions in biology that traditional methods could not address. Despite the delays caused by the Covid-19 pandemic, Zhang quickly made her mark in the scientific community.
One of her significant contributions is a paper published in Nature Communications, which arose from collaborative efforts involving researchers at the Broad Institute. This project aimed to develop a method for spatial cell analysis that integrates multiple forms of imaging and gene expression data. The groundbreaking aspect of this work was its ability to map the spatial context of cells within tissue samples, an achievement that had never been accomplished before. However, the challenge remained to analyze the vast amounts of multimodal data generated by this method, which is where Zhang’s expertise in machine learning came into play.
Zhang focused on chromatin staining as a cost-effective imaging technique that provides rich information about cellular structures. To tackle the data integration challenge, she designed an autoencoder, a type of neural network that transforms high-dimensional data into a lower-dimensional representation and vice versa. In this case, her autoencoder increased the dimensionality of the input data, allowing for the combination of datasets from different organisms while eliminating technical variations that did not reflect meaningful biological differences. This innovative approach, dubbed STACI, enabled the team to investigate the progression of Alzheimer’s disease more effectively.
The implications of Zhang’s work extend beyond Alzheimer’s research. The computational frameworks she develops have the potential to analyze various diseases, making her contributions highly versatile. Despite her ambitious goals, Zhang remains grounded, expressing a desire to tackle biologically meaningful questions that can lead to new understandings in the field. Currently, she is working on projects related to neurodegeneration and protein imaging, aiming to push the boundaries of what is known in cellular biology.
Outside the lab, Zhang leads an equally adventurous life. Her interests range from sailing and skiing to rock climbing and performing with MIT’s Concert Choir. She even earned her pilot’s license in November 2022, showcasing her determination to explore new horizons. Professor Uhler praises Zhang’s humility and dedication, noting that her diverse interests often lead to surprising conversations and insights.
As Zhang continues her research, she embodies the spirit of interdisciplinary collaboration that characterizes modern scientific inquiry. By merging machine learning with biological analysis, she is paving the way for new discoveries that could transform our understanding of cellular mechanisms and diseases. Her work not only highlights the potential of computational tools in life sciences but also serves as an inspiration for future researchers aiming to bridge the gap between technology and biology.
