This was done by VSCode Agent + Opus 4.6 (provided by Copilot)
10.03.2026 14:06
π 0
π 0
π¬ 0
π 0
This was done by VSCode Agent + Opus 4.6 (provided by Copilot)
I used to think the same but a week ago I tasked it with speeding up a C++/CUDA MolDynamics framework (hoomd), based on ideas from another framework (openmm). It identified the differences and transferred them to hoomd. The sloptimized hoomd passes all tests, is computationally stable and 3x faster
Flyer for symposium: https://ctrlepiedit.sciencesconf.org/
Very excited to announce the FIRST symposium on epigenome editing! These tools are becoming widely used in mol bio, ag & therapy. It's time to bring leaders together to discuss this rapidly growing and exciting field. And why not in Paris! Please register & share! (1/2) ctrlepiedit.sciencesconf.org
somehow stonks don't seem to mind him Β―\_(γ)_/Β―
trump said words
A consortium of group leaders from IMP, IMBA (Vienna BioCenter) and ISTA has been awarded an FWF Emerging Fields grant to study how germ cells safeguard and pass on genetic information across generations.
Two weeks left to get in your applications! βοΈ
www.sussex.ac.uk/study/fees-f...
...but what should theorists-in-biology do? I typically aim for general biologists as my target audience, which means that most of them care about the finding, rather than the technique. Yet, very often, new theoretical findings require new techniques - but then the target audience isn't interested
Postdoc positions available in my lab in Aarhus, Denmark on 'Mammalian Nuclear RNA Production and Turnover Systems'. Please get in touch for further information or simply apply here:
mbg.au.dk/en/news-and-...
Delighted to see this published: H3K115 acetylation is associated with fragile nucleosomes at CpG island promoters and active regulatory sites. Great work from all contributors; Yatendra Kumar, @illingworth-lab.bsky.social @eliasfriman.bsky.social @helink.bsky.social
doi.org/10.7554/eLif...
Excited to share this preprint that describes my latest work on using GPUs to accelerate processing of RNA-seq data.
The title says it all: "RNA-seq analysis in seconds using GPUs" now on biorxiv www.biorxiv.org/content/10.6... and github github.com/pachterlab/k...
Figure 1 shows they key result
π¨ Job Alert - Please share! π
Interested in 3D gene regulation in development & evolution? π€π§¬
π₯ Our lab at @cabd-upo-csic.bsky.social is expanding!
Weβre recruiting:
β
PhD students
β
Postdocs
π»π§ͺ Experimental or computational backgrounds welcome
π Details below
Ubiquitin & Friends Fiesta in Vienna, 29-30 April 2026.
Great speakers, small size, a friendly community, perfect for early-career researchers to connect with fellow ubiquitin enthusiasts. Sign up at www.protein-degradation.org/symposium/ and submit abstracts for talks and awards.
#ubfriends26
Pls re-post: My department @oxfordbiochemistry.bsky.social are recruiting for several new faculty positions (links below). Broad search in molecular biology/biochemistry, across prokaryotes and eukaryotes. Interested in understanding life at the molecular level, this job might be for you!
1/n
i agree with the message - but has this Claude-using-old-data hypothesis been supported by anything? All I saw so far is another output by Claude, even if the hypothesis is entirely plausible
I have so many examples of students questioning whether I know enough about a topic to teach it. I asked in faculty meeting "who has been questioned whether they are qualified to teach a class?" The women raised their hands, the men didn't. It's definitely rooted in sexism and racism.
Exciting news:
Our RNA community in @uniregensburg.bsky.social is set to grow!
We are opening a Junior Group Leader position in RNA biochemistry / ribonucleases / RNA stability. A great opportunity to start your own team within our collaborative RNA network.
Details & application π
Our Lab is searching a postdoc for an exciting ERC-funded project that will start this summer. Apply if youβre passionate about multi-omics, cutting-edge computational approaches, and exploring biological questions across the animal πͺΌ πͺ± π π¦ π¦ π·οΈ tree of life www.biozentrum.unibas.ch/open-positio...
If you want to explore ncRNA dynamics during mammalian development, love microscopy, and are not afaraid of transposons biology, applyβΌοΈ
We are looking for a postdoc to join our team at @imbavienna.bsky.social
More details ππ»
Black and white illustration featuring a forest setting with central hippo like creature, hedgehog in the foreground and birds flying overhead
Illustration from Tove Jansson's book 'The Moomins and the Great Flood'Β (1945) #Womensart
But the usage limits are tough. When I use opus 4.6, i run out of tokens every day. Sonnet 4.6 is cheaper, but still quite good
I now use opus 4.6 for similar questions in an easier project and, for the first time, it feels actually useful. Eg I asked it to generate a checklist to evaluate a narrative and the logical structure of a paper and it gave me something very reasonable (i spent a decade formulating this for myself)
A bit hard to compare, although, I did have access to gpt 5.2 and opus 4.6 temporarily. I used both to solve difficult writing questions (e.g. exploring narratives and framing) and both were more or less a waste of time. Opus had a better style, but gpt 5.2 seemed to think harder.
I switched two or three years ago and never looked back.
It's the markets
Incredibly proud to share our new preprint, lead by the Incomparable Rithika Sankar.
Here we temporally dissect the role of FACT in mES cells, finding that FACT loss drives progressive deterioration of chromatin architecture, leading to transcriptional collapse.
www.biorxiv.org/content/10.6...
1/New preprint just dropped! π₯π₯ We investigate how the genome is destroyed in apoptotic cells in a way that prevents DNA fragments from spreading beyond the dying cell π§¬β°
Done here at @imbavienna.bsky.social & in the super @rcollepardo.bsky.social & Rosen labs:
www.biorxiv.org/content/10.6...
𧬠@danielibrahim.bsky.social, moderated by @stefanschoenfelder.bsky.social, on enhancer-promoter specificity & how promoter-proximal regions act as molecular filters.
Register: us02web.zoom.us/webinar/regi...
March 5, 4PM UK.
#Genomics #GeneRegulation
I've been trying to make LLMs help me analyse my drafts and explore alternative narratives for the past 1-2 years. Opus 4.6 is the first model that doesn't feel like a waste of time. Previous LLMs were best at rubber-ducking, with neutral-to-negative time gains; Opus 4.6 feels insightful
MDAnalysis is participating in #GSoC 2026!
π Read our blog post for details on how you can apply to work with MDAnalysis through GSoC: www.mdanalysis.org/2026/02/19/g...
ποΈ Pre-proposals must be submitted by March 9.
βFor transparency and fairness, use public forums for all #GSoC communications.