Alphafold 3.0: the AI protein predictor gets an upgrade

Nature Podcast

In this episode:

00:45 A nuclear timekeeper that could transform fundamental-physics research.

Nuclear clocks — based on tiny shifts in energy in an atomic nucleus — could be even more accurate and stable than other advanced timekeeping systems, but have been difficult to make. Now, a team of researchers have made a breakthrough in the development of these clocks, identifying the correct frequency of laser light required to make this energy transition happen. Ultimately it’s hoped that physicists could use nuclear clocks to probe the fundamental forces that hold atoms together.

News: Laser breakthrough paves the way for ultra precise ‘nuclear clock’

10:34 Research Highlights

Why life on other planets may come in purple, brown or orange, and a magnetic fluid that could change shape inside the body.

Research Highlight: Never mind little green men: life on other planets might be purple

Research Highlight: A magnetic liquid makes for an injectable sensor in living tissue

13:48 AlphaFold gets an upgrade

Deepmind’s AlphaFold has revolutionised research by making it simple to predict the 3D structures of proteins, but it has lacked the ability to predict situations where a protein is bound to another molecule. Now, the AI has been upgraded to AlphaFold 3 and can accurately predict protein-molecule complexes containing DNA, RNA and more. Whilst the new version is restricted to non-commercial use, researchers are excited by its greater range of predictive abilities and the prospect of speedier drug discovery.

News: Major AlphaFold upgrade offers huge boost for drug discovery

Research Article: Abramson et al.

Subscribe to Nature Briefing, an unmissable daily round-up of science news, opinion and analysis free in your inbox every weekday.

---

Hosted on Acast. See acast.com/privacy for more information.