PhD student Sophia Hu and Professor Jianhua Xing’s lab have developed a computational method to demystify what happens when cells go through epithelial-to-mesenchymal transition (EMT).
EMT is a biological process where epithelial cells change from being rectangular in shape and tightly packed together to a mesenchymal cell. Mesenchymal cells are detached from other cells, elongated in shape, and can move around. This process occurs during wound healing and cancer spread (metastasis).
Hu and researchers discovered that the cell cycle is coupled with EMT. Additionally, cells don’t all take the same path during EMT. Instead, there are two main routes cells can take, and which path a cell takes depends on where it is in its cell cycle. Cells can start transitioning to EMT from either the G1 phase, which is early in the cell cycle, or the G2/M phase, which occurs later in the cycle.
Previous studies have had conflicting results, with some seeing one EMT path and others seeing multiple paths. The way you look at the data matters, and this study figured out why.
Hu explained it’s like trying to understand a 3D object by only looking at its 2D shadow: You might miss important details. Most previous methods “flattened” complex data too much, hiding the true picture.
The Xing lab developed and applied computational methods that analyze cell information in 2D and in high dimensions.
“We wanted to explore ways where we could perform computational analyses in higher dimensions to show that the cell cycle is coupled with EMT,” Hu said. “Sometimes, people want to stop at 2D analysis. However, higher dimensions are much more informative.”
A better understanding of EMT could lead to improved cancer treatments. “The idea is that if you know the paths or trajectories that cells can take, you can more easily figure out how to block them, and then that would help treat people,” Hu said. “And hopefully, for instance in cancer treatment, you can block metastasis.”
“This work is not just about EMT,” Hu said. “In a much general context, we asked ourselves how one could extract mechanistic information about cellular processes from high throughput single cell data. We challenged several routine practices in the field and emphasized the importance of incorporating biology and biological physics into the analyses.”
Read the full research article in Communications Biology.