A DNA typewriter recorded a mouse embryo’s family tree from its first two cells

A mouse embryo has now been made to write its own family tree. A recording device called the DNA Typewriter, built in Jay Shendure’s laboratory at the University of Washington, kept a running log of cell divisions inside a developing mouse from the moment the fertilized egg split into two cells. The log ran for the first 13.5 days of development, and it let researchers sort nearly every cell they profiled back to one of those two founders.

Help yourself and the planet: The Cool Down makes it easy to help yourself while helping the planet, with smart, simple changes to save more and waste less in your everyday life.

Get The Cool Down →

We may earn a commission if you sign up.

The DNA Typewriter in Science

The work was published in Science on October 8, 2026, as “A DNA Typewriter records the cell lineage history of a mouse, from zygote to late organogenesis.” Shendure, a professor of genome sciences at the University of Washington School of Medicine and a Howard Hughes Medical Institute investigator, and Chengxiang Qiu of Dartmouth College are the co-senior authors. Haedong Kim, Qi Yu and Sophie Seidel share first authorship, and Junhong Choi of Sloan Kettering Cancer Center, who co-invented the technique with Shendure, is credited in the release.

The full paper is on the Science website, and the University of Washington’s own news release summarizes it. The experiments sit inside the Seattle Hub for Synthetic Biology, a partnership of the Allen Institute, UW Medicine and Biohub.

A tape that writes into a living genome

Kim, a postdoctoral scientist in genome sciences at UW Medicine and co-first author, offered the simplest description: “Think of it like an actual typewriter, except it types onto a cell’s own DNA instead of paper.” The team places a redesigned recording sequence, called a tape, into the genome of a fertilized mouse egg. Each cell division then adds one new character to the tape, in strict order, and every daughter cell inherits the record written so far.

According to a detailed account in The Brighter Side of News, the characters are short DNA symbols inserted by prime editing, and each insertion closes one recording site while opening the next, which is how the order of events is preserved. The embryo carried eleven independently recording tapes, which the team redesigned so the records could be read out together with gene-activity measurements. The University of Washington release does not name the editing chemistry, so that detail rests on the outlet’s account.

The contrast with older lineage tracers is the cutting. Earlier tools, as the News-Medical summary puts it, scar the genome, can erase earlier marks and place marks in no particular order. Kim said the new tape writes without fully severing the DNA, records relatively steadily and writes everything in strict order.

What lineage resolution was achieved

The researchers attempted recording in about 100 fertilized eggs, and 10 embryos were available for analysis. They concentrated on Embryo No. 3, whose tape preserved the most extensive history. In that embryo a permanent mark made at the first division separated the two founding cells, and nearly every profiled cell traced back to one or the other.

The Brighter Side account gives the scale: the lineage tree connects 1,281,141 sampled cells to the fertilized egg, roughly 10 percent of the embryo’s cells. About 57.1 percent of cells with recovered recordings descended from one founder and 41.9 percent from the other, with about 1 percent unassigned. At day seven the team reconstructed 3,849 ancestral lineages, and just 170 of them produced half the cells in the final tree.

The founders turned out to be close to equals in what they built. One produced more descendants overall, but both gave rise to similar mixes of cell types at almost equal ratios, according to the ScienceDaily reprint of the release. Cell types also kept different schedules: blood cells and the retina committed to their fates relatively early, while the outer layer of the skin committed later.

Limits, and how the method compares

The tree is a partial sample, not a complete record, and branch dates are approximate because the editing rate varied over development. The same account notes that the result comes from a single richly recorded embryo, that cells which died are not captured, and that genes near some insertion sites changed their activity, so the recorder’s influence on development is not yet known. The work was done in mice and is not a medical treatment. Even so, a record that survives from the first division through late organ formation, and can be read cell by cell, gives developmental biologists something older methods could not: a way to ask whether a cell’s ancestry, and not only its neighborhood, helps decide what it becomes.

It also arrives alongside a rival approach. A separate system called PEtracer, published in Cell with Jonathan Weissman named for that paper, mapped more than 1.4 million cells from 16 mouse embryos, as GEN Engineering and Biotechnology News covered. Shendure said his group wants to track how cells’ early ancestry in an embryo shapes their later fates, and the authors suggest the approach could eventually help explain birth defects, neurodevelopmental conditions and cancer.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

Help yourself and the planet: Get The Cool Down →

We may earn a commission if you sign up.