New Tools Trace How Single Cells Build an Entire Mouse

Every cell in the body descends from another cell, tracing back to a single fertilized egg, much like a family tree tracks generations of ancestors. Scientists have long been able to build these cellular family trees for simple, transparent animals like roundworms, but mapping how a mammal develops from one cell into hundreds of millions in just weeks, hidden inside the mother, has remained out of reach. HHMI Investigators Jonathan Weissman and Jay Shendure, working separately with their own teams, have now each found a way to do it, reconstructing cellular family trees spanning millions of cells in developing mice and tracing how those cells commit to their eventual fates. “It’s really proof of principle that we can do what was done with the roundworm in 1983, but for mammals like you and me,” Weissman says. 

Both teams engineered cells to record their own history directly into their DNA, adding a small, permanent, inheritable mark each time a cell divides. In 2025, Weissman’s team unveiled PEtracer, a technique using prime editing to install these marks at more than a hundred genomic sites. “The cell divides and each of the sisters gets a mark, and those are inherited by their daughters, and they get additional marks, and so on and so forth,” Weissman says. “And so, by looking at the end at the marks in these DNA, we’re able to reconstruct what this relationship is.” Weissman’s team at the Whitehead Institute engineered stem cells with these marks, injected them into a mouse embryo, and sequenced individual cells as it developed to build detailed cellular family trees.

Shendure’s team at the University of Washington took a related approach with DNA Typewriter, a technology they developed in 2022 that logs each cell division onto a string of DNA, which can later be read like a ticker tape. In the new work published in Science, they injected the technology’s components into a fertilized mouse egg and tracked the marks as cells divided. “Most biological measurements are based on either live imaging, which is limited by the fact that most animal tissues are not transparent, or genomics, which are destructive and only measure a single timepoint,” Shendure says. “Recording techniques like the ones in these studies enable measurements over time including in settings that we can’t directly visualize.” 

The resulting maps could help researchers understand mammalian development and disease, including when embryos are vulnerable to genetic or environmental stressors, and how tumors form, spread, and develop treatment resistance. “There are so many key things that happen during development or during the evolution of a tumor that are happening at a time when we can’t observe them directly, so if we can record that information in the DNA, then we can infer and reconstruct exactly how it’s happening,” Weissman says. The data could also help train AI models of embryogenesis, with researchers aiming to eventually build a “virtual embryo” capable of predicting developmental outcomes. Weissman notes that, much like the roundworm map led to unanticipated discoveries decades ago, these tools may also lead to findings no one can yet predict. “That was just by understanding the process and getting at the underlying molecular mechanisms that led to those discoveries,” he says.