Toggle light / dark theme

In the brain, timekeeping is done with neurons that relax at different rates after receiving a signal; now memristors—hardware analogs of neurons—can do that too.

Artificial neural networks may soon be able to process time-dependent information, such as audio and video data, more efficiently. The first memristor with a ‘relaxation time’ that can be tuned is reported today in Nature Electronics, in a study led by the University of Michigan.

Energy Efficiency and AI.

Joint research led by Yutaro Shuto, Ryoya Nakagawa, and Osamu Nureki of the University of Tokyo determined the spatial structure of various processes of a novel gene-editing tool called “prime editor.” Functional analysis based on these structures also revealed how a “prime editor” could achieve reverse transcription, synthesizing DNA from RNA, without “cutting” both strands of the double helix. Clarifying these molecular mechanisms contributes greatly to designing gene-editing tools accurate enough for gene therapy treatments. The findings were published in the journal Nature.

The 2020 Nobel Prize in Chemistry was awarded to Jennifer Doudna and Emmanuelle Charpentier for developing a groundbreaking yet simple way to edit DNA, the “blueprint” of living organisms. While their discovery opened new avenues for research, the accuracy of the method and safety concerns about “cutting” both strands of DNA limited its use for gene therapy treatments. As such, research has been underway to develop tools that do not have these drawbacks.

The prime editing system is one such tool, a molecule complex consisting of two components. One component is the prime editor, which combines a SpCas9 protein, used in the first CRISPR-Cas gene editing technology, and a reverse transcriptase, an enzyme that transcribes RNA into DNA. The second component is the prime editing guide RNA (pegRNA), a modified guide RNA that identifies the target sequence within the DNA and encodes the desired edit. In this complex, the prime editor works like a “word processor,” accurately replacing genomic information. The tool has already been successfully implemented in living cells of organisms such as plants, zebrafish, and mice. However, precisely how this molecule complex executes each step of the editing process has not been clear, mostly due to a lack of information on its spatial structure.

Scientists used machine learning to discover what they say could be a new way to speed up the process of breaking down plastic significantly, Vice reports.

As detailed in a new paper published in the journal Nature, a research team from the University of Texas at Austin modified an enzyme to break down the individual components of polyethylene terephthalate (PET), a commonly used plastic that makes up a staggering 12 percent of global waste.

Impressively, the modified enzyme also reduced the amount of time it takes for the plastic to degrade from months to a just single week.

AMD on Monday announced its new artificial intelligence chips for everything from cutting-edge data centers to advanced laptops, ramping up its challenge to the runaway market leader Nvidia.

Demand has exploded in the past two years for the specialized processors that help develop, train and run AI applications such as ChatGPT.

AMD has emerged as one of Nvidia’s most serious contenders and CEO Lisa Su said the firm’s next-generation processors will rival the top offerings from competitors.

A small, seemingly unremarkable fern that only grows on a remote Pacific island was on Friday crowned the Guinness World Record holder for having the largest genome of any organism on Earth.

The New Caledonian fern, Tmesipteris oblanceolata, has more than 50 times more DNA packed into the nucleus of its cells than humans do.

If the DNA from one of the fern’s cells – which are just a fraction of a millimetre wide – were unravelled, it would stretch out to 106 metres (350 feet), scientists said in a new study.