Scientists have traced the family tree of about 1.3 million cells as a mouse embryo grew from a single fertilized egg into an animal with its major organs in place. The study, published Oct. 8 in Science, used a genetic recording tool called DNA Typewriter that writes a running record of cell divisions into the cells' own DNA.

A second team, publishing the same day in Cell, used a related approach to map lineages in more than 1.4 million cells from 16 mouse embryos. Together, the papers show that mammalian development can now be traced cell by cell at a scale that was not possible before.

The work matters because nearly every organ, and many diseases, trace back to decisions cells make in the embryo. A detailed map of when cells commit to becoming blood, skin or eye tissue could help researchers understand birth defects, developmental disorders and cancer.

Writing a Record Into Each Cell's DNA

The Science study, led by genome scientist Jay Shendure of the University of Washington and Chengxiang Qiu of Dartmouth College, built on DNA Typewriter, a tool Shendure developed with Junhong Choi. As cells divide, the system adds marks one after another to a stretch of DNA, like letters typed onto a tape. Cells that share early marks share ancestors, so reading the marks reveals the family tree.

The researchers used prime editing, a gentler form of gene editing that writes changes without fully cutting both strands of DNA. Earlier CRISPR-based barcodes could scar or harm cells when too many edits piled up, according to Nature. "It writes without fully severing the DNA, keeps recording relatively steadily," said Haedong Kim, a postdoctoral scientist at UW Medicine and co-first author.

The team followed development for about 13.5 days, close to two weeks into a mouse pregnancy that typically lasts 19 to 21 days, by which point the major organ systems have formed, according to News-Medical. The reconstructed lineage covered about 1.3 million cells, roughly 10 percent of the embryo's total.

Two Founding Cells, Unequal Families

The results revealed a permanent mark separating the first two cells formed after the egg divided, and nearly every profiled cell traced back to one or the other. One founding cell produced more descendants, but both gave rise to a wide range of cell types in roughly equal proportions.

The timing of cell decisions varied by tissue — and that variation is the core of the finding. Blood cells and the retina committed to their fates relatively early, while the outer layer of the skin committed later. That suggests different lineages lock in their identities on separate schedules rather than all at once.

"Twins look the same, humans kind of look the same, yet even twins develop through very different sets of cell divisions," Shendure told Nature, pointing to how much variation hides beneath similar-looking bodies.

A Second Map From the Whitehead Institute

The Cell study, led by molecular biologist Jonathan Weissman of the Whitehead Institute in Cambridge, Massachusetts, used a prime-editing system called PEtracer. It captured most of the cell divisions in embryos as organs formed, analyzing more than 1.4 million cells across 16 embryos, according to GEN.

"It's really proof of principle that we can do what was done with the roundworm in 1983," Weissman said. He was referring to John Sulston's landmark work tracing every one of the 959 body cells of the tiny worm C. elegans using only a microscope, which remains the only complete map of an animal's development.

Limits Before a Complete Mammal Map

These studies are a large step — but not a complete map. The Science team attempted recordings in 100 fertilized eggs and obtained 10 usable embryos, and its detailed analysis centered on a single embryo with the richest record. Its 1.3 million cells represent about a tenth of that embryo.

That means some findings, such as the timing of lineage decisions, come from one mouse and need to be confirmed in more animals. Both teams say the methods must be scaled across more embryos and conditions before they can support predictive models of development.

The work also involved genetically engineered laboratory mice, engineered as fertilized eggs to carry the recording system, and it does not directly study human embryos or offer any immediate medical application. Its near-term value lies in basic research, giving scientists a tool to ask how organs form and where development goes wrong in models of disease.

Shendure framed the goal in terms of time. "Recording techniques like the ones in these studies enable measurements over time," he said, according to GEN. Most methods for studying embryos capture only snapshots, showing which cells exist at a given moment but not how they got there.

Lineage recorders fill that gap by storing history inside the cells themselves, so scientists can read it out later with DNA sequencing. That makes it possible to ask whether two similar cells share a recent ancestor or arrived at the same identity by different routes.

Researchers expect to apply the recorders to questions about congenital malformations, neurodevelopmental and genetic disorders, cancer and stem cell engineering.

The bottom line is that scientists can now record, inside living cells, much of the family tree that builds a mammal. The next challenge is extending these maps to whole embryos and many animals, to see which developmental paths are universal and which vary from one individual to the next.

What Readers Want to Know

What did the scientists do?

They used genetic recorders written into cells' DNA to reconstruct the family tree of about 1.3 million cells in a developing mouse embryo, published in Science on Oct. 8, 2026.

How does DNA Typewriter work?

It adds sequential marks to a stretch of DNA as cells divide. Cells sharing early marks share ancestors, revealing the lineage tree.

What did the second study find?

A Whitehead Institute team used a prime-editing tool called PEtracer to map lineages in more than 1.4 million cells from 16 mouse embryos, published in Cell.

Is this a complete map of mouse development?

No. The 1.3 million cells were about 10 percent of one embryo, and most detailed findings came from a single animal.

Why does this matter for human health?

It could help researchers understand birth defects, developmental disorders and cancer, though the work was done in mice.

What remains unknown?

Researchers need to scale the methods across more embryos to confirm which developmental patterns are universal.