Colossal unveils GenoMorph AI to find dodo gene-editing targets
Representative image: fernandozhiminaicela / Pixabay
Colossal Biosciences on 14 October 2026 unveiled GenoMorph, an AI platform that has flagged about 2,300 candidate DNA edits for dodo skull and beak traits.
What Happened
Colossal Biosciences, the US company working to bring back extinct animals, on Wednesday, 14 October 2026, unveiled GenoMorph, an artificial intelligence platform built to find gene-editing targets when there is very little genetic data to work with. The company said the system has already been used on its dodo project, where it identified about 2,300 candidate stretches of DNA that may shape the bird's distinctive skull and beak. More than 650 of the candidates relate to beak size and close to 1,700 to beak thickness and breadth. Colossal said these candidates are now going into laboratory validation, meaning they are predictions, not confirmed edits.
Key Facts
- Company: Colossal Biosciences.
- Announcement: 14 October 2026.
- Platform: GenoMorph, an in-house AI system for choosing gene-editing targets when genomic data is scarce.
- Method: it combines comparative genomics, developmental biology and high-resolution 3D scans of anatomy to predict which regulatory regions of DNA are active as an animal develops.
- Training data: developmental data from chickens, quail, ducks and pigeons, plus genetic and skull data from more than 70 bird species.
- First output: about 2,300 candidate DNA targets for dodo head and beak traits, including more than 650 for beak size and nearly 1,700 for beak thickness and breadth.
- Data gap: Colossal has a single dodo genome to work with, compared with about 100 genomes for its woolly mammoth project.
- Colossal also reported a lipid-nanoparticle method to deliver gene-editing tools into developing bird embryos, which it said worked in three bird species.
Why It Matters
Most of the public talk about artificial intelligence is about chatbots and coding assistants. GenoMorph is an example of a different use: AI as a research tool in biology, where it helps scientists narrow down a huge number of possibilities before they spend time and money in the laboratory.
The problem it tries to solve is a real one. To recreate a trait such as the dodo's heavy, hooked beak, scientists need to know which parts of the genome control it. For the mammoth, Colossal could compare many ancient genomes. For the dodo, it has only one. GenoMorph attempts to fill that gap by learning how genes and DNA switches shape the skull in living birds, then using 3D measurements of skulls from dozens of species to predict which regions matter for the features the company wants to reproduce.
The numbers need careful reading. Some 2,300 candidates is a list to test, not a recipe. Each candidate must now be checked in the laboratory, and many may turn out to have little or no effect. Independent scientists will also want to see the method published and reviewed before judging how well it works. De-extinction itself remains controversial, with critics arguing that resources would be better spent protecting species that still exist. But tools like this one, which predict what a stretch of DNA does from limited data, could prove useful well beyond the dodo, for example in conservation genetics for endangered birds.
| Item | Detail |
|---|---|
| Platform | GenoMorph |
| Announced | 14 October 2026 |
| Species studied for training | Chickens, quail, ducks, pigeons, plus skull data from 70+ species |
| Dodo candidates found | about 2,300 |
| Next step | Laboratory validation |
Impact
Short-term: Colossal will test the candidate DNA regions in the laboratory, using its new method for delivering gene-editing tools into bird embryos. Results from those tests will show whether the AI's predictions hold up.
Long-term: If the approach works, it could shorten the search for genetic targets in species with little genetic data, both extinct and endangered. It would also add to the growing list of AI tools that help biologists decide which experiments to run first.
Who is affected: Colossal and its investors, researchers in genomics and developmental biology, conservation groups working with endangered birds, and the wider debate about whether de-extinction is a good use of science.
Key Takeaway
Colossal's GenoMorph AI has produced a shortlist of about 2,300 candidate DNA targets for the dodo's beak and skull, but each one still has to be proven in the laboratory.
Questions and Answers
What is Colossal's GenoMorph?
GenoMorph is an AI platform unveiled by Colossal Biosciences on 14 October 2026. It combines genomics, developmental biology and 3D anatomy data to predict which DNA regions shape physical traits, especially when only limited genetic data exists.
How is GenoMorph being used for the dodo?
It has identified about 2,300 candidate DNA targets linked to the dodo's skull and beak, including more than 650 for beak size and nearly 1,700 for beak thickness and breadth. These are now being tested in the laboratory.
Why does the dodo project need AI?
Colossal has only one dodo genome to study, compared with about 100 genomes for its woolly mammoth work, so it cannot rely on comparing many genomes. GenoMorph learns from living birds to fill that gap.
Has Colossal brought back the dodo?
No. The candidates found by GenoMorph are predictions that still need laboratory validation, and the work has not yet been independently peer reviewed.
Disclaimer: Prepared by the Peepals newsroom from publicly available sources with AI assistance. Information is accurate to the best of our knowledge at the time of publication and may change. Images may be representative. Not professional advice. Report an error via our contact page.
Sourced and fact-checked by the Peepals Global Editorial Team
Apple TV 4K gets Siri AI and A19 chip, but price more than doubles to $299
Freshworks to buy Torii to track company spending on AI and software
Reported, fact-checked and published by the Peepals Global Editorial Team.
