Skip to main content
Photo of Melissa Kinney

Reading the roadmap of stem cell differentiation

Written By: Tom Ziemer


The path a stem cell takes toward its identity as a fully defined cell, the analogy goes, is like a ball rolling down a hill. Along the way, it encounters bumps, obstacles and varied terrain that influence its final destination.

Unfortunately for stem-cell researchers, seeing each transition in a cell’s state—and understanding what drives that change—isn’t yet possible in real time.

Stem cell and systems biologist Melissa Kinney, an assistant professor of biomedical engineering at the University of Wisconsin-Madison, is working to uncover new details of dynamic biological processes, such as stem cell differentiation, with the hopes of better guiding cells down that path toward a desired identity.

In a new paper in Nature Biotechnology, Kinney and collaborators outline a method they developed to glimpse transitional phases in the differentiation of hematopoietic stem cells toward red blood cells and describe how it pinpointed a key regulator of blood cell development.

She sees it as a proof of concept for a versatile computational approach that researchers could use with different types of cells and processes beyond stem cell differentiation, including CAR-T cell activation.

“The data that we collect tend to give us snapshots. What we’re trying to answer is: Can we use those static snapshots to figure out the ‘why’—why cells transition from one state to the other?” she says. “We can’t actually see everything that is happening in the cell in real time, so we use computational methods to infer what’s happening from one stage to the next during these dynamic processes.”

The project, which Kinney completed during her time as a postdoctoral researcher at Boston Children’s Hospital and Massachusetts Institute of Technology, builds off a previously created algorithm from Harvard Medical School that allows researchers to judge the veracity of their stem cell-derived, differentiated cells. But Kinney says this new computational approach delves deeper into the nuances of differentiation.

“We are calling it a pipeline or a roadmap. Our roadmap is intended to help investigators connect the types of questions they want to ask with relevant computational tools, because everyone’s questions are going to be slightly different,” she says.

The roadmap led Kinney and her collaborators to a protein, ErbB4, that was not previously known to influence red blood cell development. The finding moved the researchers incrementally closer in their quest to produce fully developed red blood cells in a dish.

Kinney, who joined the Department of Biomedical Engineering in January 2019, says she plans to expand upon the model as part of her lab’s ongoing efforts to shape stem cell behaviors.

“I think there’s so much power in being able to use computational methods to understand really complex systems,” she says. “We’re studying blood in general, but there are a lot of different questions that we can ask with the same computational methods. I’m excited to see how this applies across a whole spectrum of different applications.”