GoKawiilA new AI tool has generated an atlas of more than one billion predicted protein structures. Plus, a biology lab run by ten robots and whether we can truly trust eyewitness accounts during criminal trials.
Hello Nature readers, would you like to get this Briefing in your inbox free every day? Sign up here.
Laboratory robots with built-in artificial-intelligence software can look after stem-cell cultures independently.Credit: Masatoshi Okauchi/Shutterstock
In a biology laboratory in Tokyo, ten two-armed robots run the show. The robots can handle liquids, grow cells on plates and operate scientific instruments, among other basic tasks. The lab does still require some human input to function, but largely frees researchers up to focus on designing experiments and interpreting results. The ultimate goal is to create a “factory-scale” facility with thousands of robots that could be used by local and international scientists by 2040 or 2050, says automation researcher Genki Kanda.
Nature | 5 min read
Dwindling daylight hours in autumn usually prompt mosquitoes (Culex pipiens) to head to dark spaces such as caves, where they’ll hide out over the winter. But outdoor lights in urban areas can disrupt that timeline, which means that the insects stay active for longer. A delayed dormancy keeps the window open for longer for the mosquitoes to bite and transmit West Nile virus. It also leaves more time for the insects to reproduce, which could lead to a bigger population in the following spring.
Science | 4 min read
Reference: Journal of Applied Ecology paper
A newly released artificial-intelligence tool called ESMFold2 has generated an atlas of 1.1 billion predicted protein structures and billions more protein sequences. The database, known as the ESM Atlas, vastly increases the known protein universe, eclipsing the AlphaFold Database of predicted protein structures by more than 800 million entries. The freely accessible atlas should be “an extraordinary resource for biology,” says computational biologist Gemma Atkinson. And the open-source nature of ESMFold2 means that it could find wide-ranging uses, says computational biologist Sergey Ovchinnikov.
Nature | 5 min read
... continue reading