Juan LiAI4Science
Reading updates
EN中

AI-generated weekly synthesis · selected by a reproducible rule, not curated by a human · not individually human-reviewed · verify original sources before citing

Google DeepMind publishes AI-powered predictions for effect of all 9B mutations to human DNA

What changed

Google DeepMind said Tuesday that it has used artificial intelligence to predict the biological consequences of all 9 billion possible single-letter changes to human DNA, and is making the resulting database available free to academic researchers worldwide. AlphaGenome Atlas, as DeepMind calls the database, is a precomputed catalogue of what each substitution of a single DNA base is likely to do to the machinery that switches genes on and off. Until now researchers had to run such a model one variant at a time or had to test variants in the laboratory, a process that was painstakingly slow. It would have taken many human lifetimes to discover the consequences of all 9 billion possible single-letter mutations. The Atlas promises to make the job of biologists and medical researchers considerably easier, potentially speeding up the understanding of genetic diseases and the hunt for possible cures. Pushmeet Kohli, DeepMind’s vice president for research and head of its AI for science team, told reporters on a briefing call that this was the first time any researcher in the world could reach a comprehensive map of human genetic variation “by simply opening a browser.” Kohli also framed the release as helping to complete the unfinished business of the Human Genome Project, which in 2003 succeeded in mapping the entire human DNA sequence. “As the saying goes, we bought the book,” he said, “but we did not understand how to read it.” Atlas is available for non-commercial use from today through a website Google DeepMind has set up for it. The company said it would be available for commercial use through a licensing arrangement through Google Cloud “soon.” Kohli said that Google DeepMind’s sister company, Isomorphic Labs, which is using AI for drug discovery, would have access to Atlas but that it would also require a commercial license for access. He did not specify exactly what the terms would be for commercial licensing. A paper describing the Atlas and how it was created is being released on bioRxiv, a repository for biomedical preprint academic papers. Helping understand mutations in DNA’s vast ‘non-coding’ segments DNA provides the recipe for the proteins a living cell can make. It consists of two chains of nucleotides, or molecules containing nitrogen, that coil around one another to produce a double helix. The nucleotides in DNA are formed of one of four differ…

Sources and evidence

外部发现 · Google DeepMind · Original source ↗ · Site entry ·

Self-similar study of stable singularity in unforced Euler equations on R^3

What changed

[This is a guest post by Anima Anandkumar . This blog post was initially written in a different file format and converted using AI. — T.] The last few days have been a flurry of activity. After months of sleepless nights and diving deep into different mathematical and computational tools to tackle the open problem of singularity in fluid dynamics, we were finally ready to send our work on Euler equations (without forcing) to Tom Hou and Terry Tao for feedback. Almost immediately I heard back from Terry encouraging us to release our work publicly, given that the NYU team had just released their results on forced Euler and OpenAI was rumored to release the next day. We scrambled and put together a blog post and attached our paper with code, which went live about 30 minutes after, in the evening of September 7, the day before the OpenAI’s announcement. A lot of mainstream media religiously followed their press release that fails to acknowledge our work even after we informed them. On the bright side, many researchers have reached out to me. I want to thank Terry for highlighting our work, as well as for giving me the opportunity to write this blog post. The starting point for us is different from the route that the NYU team and OpenAI pursued. Their initial solution has a specific construction with a multi-scale structure with discrete jumps in scales. On the other hand, we consider a self-similar ansatz, and so far, no one has been able to construct a self-similar singular profile for the Euler equations in without forcing. Modified cases such as the Euler equation with a boundary have been successfully solved before ( Chen and Hou ), where the boundary helps confine the singularity, but such techniques fail in free space. We wanted to see if AI could come up with a self-similar profile for Euler. LLMs are not suited for generating such singularity candidates without analytical closed forms. Physics informed neural networks (PINN), on the other hand, can be optimized to find such solutions. Although there have been prior attempts to use PINNs for this case, they were not successful. There are optimization and precision challenges: with standard ansatz and optimizers, a common failure mode is getting the trivial solution. The other challenge is getting precision to be high enough, so we can then attempt to convert the numerical solution into an analytical proo…

Sources and evidence

陶哲轩 · What's new 博客 · Original source ↗ · Site entry ·

Google DeepMind maps 9 billion DNA mutations: how AI could help solve rare diseases

What changed

Google DeepMind has created an AI-powered map of about 9 billion possible single-letter changes in human DNA, giving scientists a new way to search for genetic mutations that could matter for disease and biology. Called AlphaGenome Atlas, the platform contains precomputed predictions for every possible single-nucleotide variant in the human genome. DeepMind says testing billions of possible changes individually in laboratories would be practically impossible. The company has made the resource available for academic research through a free website, alongside its AlphaGenome API. What does AlphaGenome Atlas actually do? Think of it as a giant search map for DNA. The Atlas uses predictions from AlphaGenome, DeepMind’s AI model designed to predict how genetic variants affect biological processes. It covers both the part of DNA that produces proteins and the much larger non-coding portion that helps control gene activity. DeepMind says the Atlas is a 1-petabyte dataset, more than 30 times larger than the AlphaFold Database. It also introduces an AlphaGenome Variant Impact (AVI) score. The score gives researchers a single way to rank potentially important genetic changes and investigate what biological processes they may disrupt. Why could 9 billion DNA changes matter? The human genome contains an enormous number of possible variations. Finding the few changes that actually matter for a disease can be like searching for a needle in a haystack. DeepMind says researchers have already used AlphaGenome Atlas in rare-disease research and studies of common traits. Trending Stories In one example, researchers working with the Broad Institute used the AVI score to prioritise a variant linked to DNM1, a gene associated with epileptic encephalopathy. Experimental work then validated the predicted effect. Another study using data from more than 54,000 UK Biobank participants identified 22% more non-coding genetic associations by grouping variants according to their predicted molecular effects. What happens next? DeepMind says AlphaGenome Atlas is a starting point rather than an endpoint. As its AI models improve, the company expects the predictions to become more comprehensive and precise. The goal is not to replace laboratory research. Instead, the Atlas could help scientists decide which genetic changes deserve closer investigation first. That could speed up research into…

Sources and evidence

外部发现 · Google DeepMind · Original source ↗ · Site entry ·

Controversy over OpenAI's claimed Navier-Stokes breakthrough

What changed

The controversy over OpenAI's maths miracle ‘Historic’ mathematical breakthrough by the tech company has been overshadowed by data privacy concerns of human researchers This year has been one of “unthinkable” progress for maths, said The Wall Street Journal. But OpenAI’s proof of the Navier-Stokes problem – one of the seven esteemed Millennium Prize Problems accepted as maths’ “deepest and most difficult unresolved questions” – “represents the most significant advance yet”. The tech company took around three days to find a solution with an internal model that is “significantly more capable” than its most advanced commercial model, said New Scientist. It first used 1,000 AI agents to crack a similar problem, then around 10,000 to complete the task. If a customer wanted to run the same problem, it would cost them “around $15 million” (£11 million). However, there has been a “dispute” about how the solution was reached. Two human researchers had announced they had “cracked three significant problems” linked to the Navier-Stokes problem, also using OpenAI and Anthropic large language models, which has raised questions over the privacy and integrity of the data integral to the breakthrough. The Week Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives. Sign up for The Week's Free Newsletters Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles. ‘Potentially historic breakthrough’ The “Navier-Stokes existence and smoothness problem” essentially looks at equations which describe the movement of liquids and gases, said The New York Times. These equations are used to predict the weather and “model the behaviour of ocean currents or the climate as a whole”. The problem challenges whether, theoretically, the “laws of physics would break down under certain extreme conditions”, such that water, for instance, could “spontaneously explode”. This “potentially historic breakthrough” has been “shadowed by questions over unpublished research by outside mathematicians”, said Axios. New York University mathematician Tristan Buckmaster and Anthropic researcher Levent Alpöge had been using large language models, including OpenAI’s Codex, where information was stored privately. Just hours before OpenAI’s announcement, the…

Sources and evidence

外部发现 · OpenAI · Original source ↗ · Site entry ·

A collaborative agent with two lightweight synergistic models for autonomous crystal materials research

What changed

Nature Machine Intelligence, Published online: 10 September 2026; doi:10.1038/s42256-026-01298-6 Shi et al. demonstrate a dual architecture approach for materials research, integrating two lightweight large language models for collaborative reasoning and scientific tool execution. The method achieves competitive performance while remaining affordable and locally deployable.

Sources and evidence

Nature Portfolio · Nature Machine Intelligence · Original source ↗ · Site entry ·