Do they think people have literally no memory of past events at all?
07.03.2026 21:59 β π 1405 π 375 π¬ 63 π 40@zaminiqbal.bsky.social
Professor of Algorithmic and Microbial Genomics at the University of Bath (UK). Pangenomes, drug resistance (esp TB), data structures for DNA search, plasmid evolution, global microbial surveillance. Open Data, reproducibility
Do they think people have literally no memory of past events at all?
07.03.2026 21:59 β π 1405 π 375 π¬ 63 π 40Caitlin KALINOWSKI over X I resigned from OpenAl. I care deeply about the Robotics team and the work we built together. This wasn't an easy call. Al has an important role in national security. But surveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got. This was about principle, not people. I have deep respect for Sam and the team, and I'm proud of what we built together.
OpenAI head of robotics just resigned over company deal with the Pentagon sayingβ¦
βSurveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they gotβ
"novel transformer-based model, Popformer, for learning encodings of general patterns of genetic variation...includes site-wise and haplotype-wise attention.... The model is pre-trained with an analog of the masked language modeling objective across a range of real human genomic data"
07.03.2026 07:47 β π 13 π 3 π¬ 0 π 0
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
The good kind of infighting. Creasy explaining very clearly why the policy sucks. More power to the backbenchers
05.03.2026 22:53 β π 251 π 35 π¬ 9 π 3
New post just out:
"Sovereignty for sale"
The UK is uniquely exposed to the power of US tech monopolies - from Palantir to AI and cloud services.
Why is it so dangerous? What are other countries doing? And what should we do?
(Β£/free trial)
open.substack.com/pub/samf/p/s...
FT comments section this morning - saying what everyone else is thinking, right?
05.03.2026 07:15 β π 17133 π 5975 π¬ 514 π 493
Postdoc on AI/biology funded by asanger and DeepMind!
www.sanger.ac.uk/news_item/ne...
Spanish PM SΓ‘nchez responds to Trump:
"Spain is against this disaster... Govts are here to improve people's lives... It is absolutely unacceptable that those leaders who are unable to fulfill that mission use the smoke of war to hide their failure, and in the process, fill the pockets of a few."
right, deleted, sorry both!
04.03.2026 14:20 β π 0 π 0 π¬ 1 π 0
er...yes, sorry :-(
i'm not really adding to the discourse am I, maybe i should delete
"According to the New York Times, the United States asked PAHO for the deadline to be pushed back, with some experts again noting the administration does not want to lose elimination status before the midterm elections."
www.cidrap.umn.edu/measles/bhat...
"..used the tech to identify peptide inhibitors of transcription factor CREB1, which is overactive in a broad range of cancers ..
They have shown these peptides can enter human cancer cells grown in the lab, shut down CREB1-controlled pathways, and selectively kill cancer cells."
New paper showing that bacteria with more genes for cooperation can live in a broader range of habitats and that genes for cooperation are more more likely to be in the accessory genome www.pnas.org/doi/10.1073/... @lauriebelch.bsky.social
04.03.2026 10:23 β π 55 π 24 π¬ 2 π 0I felt the need to write to my MP yet again
03.03.2026 11:01 β π 76 π 23 π¬ 6 π 2
Coming to Europe...
#Medsky #IDsky #influenza #COVID #vaccines #Moderna #MRNA
Starmer's ability to find the exact sweetspot which pisses off everyone is extraordinary. He's a savant.
01.03.2026 21:25 β π 2878 π 534 π¬ 84 π 24Help me out #MicroSky - what are your favourite microbial/viral genetics practical classes for undergraduates that can be run across just two sessions?
23.02.2026 12:43 β π 2 π 3 π¬ 0 π 0If Reform had won Starmer would be on his knees to those voters begging for forgiveness. Greens win and it's like "Fuck off cranks"
28.02.2026 11:04 β π 422 π 63 π¬ 13 π 3Not even hiding the belief that there are Real Voters (white, male, perhaps working class when politically convenient) and Fake Voters (graduates, Muslims, people in cities, anyone on the left regardless of class or ethnicity). A belief shared by many on the Labour right
28.02.2026 11:20 β π 128 π 27 π¬ 12 π 3This is terrible from Starmer. Just absolute trash. And any normal voter looking at the new MP and then reading this description will think him mad.
27.02.2026 21:15 β π 1507 π 274 π¬ 122 π 14The man who brought you βIsland of Strangersβ has views on divisive politics
27.02.2026 14:13 β π 113 π 28 π¬ 1 π 0Nobody here needs to be told that Reform's claims of cheating are utterly absurd but it's worth noting that the anti-Reform vote was so large that even if the Greens and Labour had split it 50/50 Reform would still have lost β in fact they would have come third. They only value democracy if they win
27.02.2026 10:54 β π 426 π 98 π¬ 10 π 2There are millions of people who have that stomach. We just need to get them into public office.
27.02.2026 13:48 β π 772 π 109 π¬ 14 π 3@martamiori.bsky.social and I have been writing, since 2024, about why Labour's 'Reform' challenge and emphasis was based on a misunderstanding of Labour's vote. Here for anyone interested: politicscentre.nuffield.ox.ac.uk/news-and-eve...
27.02.2026 08:02 β π 129 π 70 π¬ 3 π 18Unrooted phylogeny indicating the major clades of Mycobacterium bovis found in Ireland.
The genomic diversity and spatial patterns of Mycobacterium bovis in Ireland revealed by whole genome sequencing rdcu.be/e50zT
27.02.2026 10:42 β π 8 π 3 π¬ 0 π 0This could be true, I don't know how we'd know really. What do you mean by load/requirements for ENA - they are under load, and submission has tricky requirements?
27.02.2026 09:16 β π 0 π 0 π¬ 1 π 0This spoof is fantastic, not least for the aging choices theyβve made for the tech broligarchs. youtu.be/zdkBUuWCZg4?...
27.02.2026 08:26 β π 30 π 11 π¬ 3 π 0One interesting thing we did is look at rate of submission bacterial sequence data to INSDC. Logistic fit is surprisingly good - inflexion during the pandemic. So, the data no longer looks like exponential growth. Projected to 2035 just for the hell of it, let's see.
27.02.2026 09:02 β π 16 π 1 π¬ 2 π 0Cheers Ewan! Was very glad to see our protein structure predictions get into EBI AFDB! Open databases FTW!
26.02.2026 16:18 β π 1 π 1 π¬ 0 π 0