Sign in

iiinova is part of the Atmosphere — an open, public network where you own your identity.

Cas Blaauw

@casblaauw.eurosky.social
329 followers 531 following 147 posts

(machine) learning to turn one genome into 959 different cells 🪱 | bioinfo phd 🧬🖥️ in Leuven (Aerts lab @ vib.ai 🇧🇪) & Utrecht (AvO lab @ hubrecht.eu 🇳🇱) also urbanism & (cross-border) transport, occasionally | prev 🇳🇱/🇭🇰/🇩🇰 | he/him | ook in het nederlands

PostsRepliesMedia
Reposted by Cas Blaauw
Evgeny Kvon @evgenykvon.bsky.social · 14/10/2026
Are weak TF motifs important? Weilert et al use deep learning to systematically map low-affinity pioneer binding sites in mESCs and show that they cooperate with strong pioneer motifs to drive chromatin opening. Paper: www.cell.com/cell-genomic... Preview w/ Grace Bower: www.cell.com/cell-genomic...
cell.com
Deep learning uncovers hidden power of low-affinity motifs
Low-affinity transcription factor binding sites are functionally important but difficult to map and mechanistically poorly understood. In this issue of Cell Genomics, Weilert et al. leverage sequence-...
05322 53
Reposted by Cas Blaauw
Brandon Bishop @brandontbishop.bsky.social · 18/10/2026
Approximately 1/3 of our students pursue some form of graduate education, 1/3 pursue other opportunities outside of the field, 2/9 go directly into K12 education, energy, water or consulting sectors, and 1/9 transform into sand tiger sharks swimming off beyond the horizon never to be seen again.
19113 91
Cas Blaauw @casblaauw.eurosky.social · 18/10/2026
Wait, all of this just to spend like an hour in the sky? That must hurt even more...
000 0
Reposted by Cas Blaauw
#SingleCell preprints @prepub-singlecell.bsky.social · 18/10/2026
Order Matters: Ambient RNA correction confounds doublet detection in single-cell RNA sequencing #SingleCell 🧪🧬🖥️ www.biorxiv.org/content/10.64898/20…
053 5
Cas Blaauw @casblaauw.eurosky.social · 16/10/2026
After going through a bunch of supplementary tables recently: we desperately need to stop providing these as excel files. So many genes turned into dates, so many typos, so many weirdly formatted sheets...
020 2
Reposted by Cas Blaauw
Shannon Mattern @shannonmattern.bsky.social · 15/10/2026
"Men can generally be bad dept citizens as long as their research is valued, but women can’t." "Meritocracy is really just a way [to support] the status quo of a dominant white, male envrnmnt…I thought that if I worked hard enough, my work wld be recognized. But [that's] not how the system works."
academic.oup.com
Devaluation underlies gendered US faculty departures across disciplines and institutions
Abstract. Women faculty experience academia differently from men in many ways, which can lead them to consider leaving their positions. Using a large-scale
3537233 537
Reposted by Cas Blaauw
Gabriel Flynn @gabrielflynn.bsky.social · 14/10/2026
AI researchers warn technology 'months away' from being able to connect to wifi using eduroam
241244331 1.2K
Cas Blaauw @casblaauw.eurosky.social · 15/10/2026
Congrats!
110 1
Reposted by Cas Blaauw
Vicki Metzis @vmetzis.bsky.social · 07/10/2026
How do DNA sequence features tune the timing, level or duration of gene expression? Join us for a 4 year PhD @mrc-lms.bsky.social with @perez-carrasco.bsky.social @jamesware.bsky.social to build models and help us better predict the impact of non-coding sequence variation #devbio #quantitativebio👇
01614 16
Cas Blaauw @casblaauw.eurosky.social · 13/10/2026
This was a very interesting read, and it's led to some good discussion with colleagues here so far. I'm gonna have to let it simmer a bit longer to make up my mind about this, but thanks for sharing!
110 1
Reposted by Cas Blaauw
Evgenii Protasov @evgenii-protasov.bsky.social · 10/10/2026
The archaeon Sukunaarchaeum mirabile, associated with dinoflagellates, has a genome of only 238 kbp🧬🤯 #archaea #symbiosis #eukaryotes #protists #MicroSky @currentbiology.bsky.social www.cell.com/current-biol...
cell.com
A hidden archaeal lineage with an ultra-reduced genome retaining only its replicative core
Harada et al. describe an archaeon, recovered from a single marine plankton cell, whose 238-kb genome is the smallest known for the domain. It has lost nearly all metabolism but keeps the machinery to...
29036 90
Reposted by Cas Blaauw
Jesse Engreitz @jengreitz.bsky.social · 09/10/2026
Systematic mapping of enhancers, silencers, and topological elements using prime editing deletion screens www.biorxiv.org/content/10.6... Check out our preprint on next-generation tech for mapping distal regulatory effects! We’re excited to help set this up - please get in touch Thread 👇
biorxiv.org
Systematic mapping of enhancers, silencers, and topological elements with prime editing deletion screens
Noncoding regulatory elements in the human genome can control gene expression through distinct mechanisms, including acting as enhancers, silencers, or topological elements. However, we lack tools to ...
06022 60
Reposted by Cas Blaauw
Judith Kribelbauer @jk-swietek.bsky.social · 09/10/2026
ui, a new regulatory sequence models from Kelley/Linder to play with - Cerberus! a new architecture with Hydra blocks replacing the transformer blocks. www.biorxiv.org/content/10.6...
biorxiv.org
011 1
Reposted by Cas Blaauw
James Davies @jojdavies.bsky.social · 07/10/2026
Delighted to see our latest paper out in @natgenet.nature.com ! The brilliant @jhamley98.bsky.social developed MCCv, allowing chromatin architecture to be visualised at base-pair resolution for each allele—a powerful way to identify functional genetic variants. www.nature.com/articles/s41... 🧵
12818 28
Reposted by Cas Blaauw
Alejandro Fábregas-Tejeda @alejandrofabregastejeda.com · 06/10/2026
Please do get in touch if you are a philosopher or scientist (from any discipline) interested in these topics—I would love to collaborate and learn from you! If you would like to learn more about my research or find my contact info, please go to www.alejandrofabregastejeda.com
2151 15
Reposted by Cas Blaauw
Alejandro Fábregas-Tejeda @alejandrofabregastejeda.com · 06/10/2026
I am really happy to share that I have started as PI of an @fwovlaanderen.bsky.social-funded 3-year project @clpskuleuven.bsky.social. I'll investigate how scientists conceptualize, model & measure environments across the life, cognitive, social & Earth sciences! #philsci #philsky #evosky #envhum 🧪
1215215 152
Reposted by Cas Blaauw
Kaili Fan @kailifan.bsky.social · 05/10/2026
1/13 The human body runs on one genome—but every cell reads it differently. How? To find out, we need to measure both gene expression and chromatin accessibility in the same nucleus. So we did, across 21 adult tissues and nearly half a million nuclei. 🧵 🔗 www.biorxiv.org/content/10.6...
biorxiv.org
Single-Nucleus Multi-Omic Atlas Maps Regulatory Architecture and Non-Coding Variant Effects across Adult Human Tissues
Diverse human cell types establish specialized functions through lineage- and context-specific regulatory programs. Interpreting non-coding genetic risk requires integrated multi-omic reference maps t...
1179 17
Cas Blaauw @casblaauw.eurosky.social · 05/10/2026
Has everyone just forgotten to write all of a sudden? I understand writing is not as easy when English is not your native language, but it's not improved by adding more "not just this — it's also that". Nobody's forcing you to throw your writing through the slop machine.
100 1
Reposted by Cas Blaauw
Julia M. Rohrer @dingdingpeng.the100.ci · 04/10/2026
This is very good. “Over 2.2 million cancer deaths can be attributed to Costco; that’s more than 20% of all cancer deaths between 2000 and 2018.”
Ridiculous things you can “prove” caused cancer mortality

Everything causes cancer. Sounds cliché; maybe those California warning tags were right all along. But thanks to the methodology created by Alwadi et al., we can now prove that anything and everything causes cancer. Oh, sorry, that is too strong of an assertion. To use the authors’ words, since they have claimed their papers don’t prove causality, we can create an association between any physical landmark and cancer and, from that association, figure out how much cancer is attributable to that thing. Which is totally not the same as saying “that thing causes cancer.”

For instance, living near a private four-year university is associated with a 15-fold increase in cancer mortality when compared
1111523 115
Reposted by Cas Blaauw
Matt Henderson @matthen.com · 03/10/2026
Visualising the distribution of substrings in the genomes of different organisms. Going up to depth eight by recursively splitting into quadrants, we get these fractal-like pictures.
817135 171
Reposted by Cas Blaauw
samuel mehr @mehr.nz · 02/10/2026
early-career academics, if you don't have a website, make yourself one. yes it's annoying but just do it it can be plain text on a white background, or google sites, or even a free Wix one with ads. nobody cares we just need to be able to look you up, quickly, bc there's like 100 other apps here
1119058 190
Cas Blaauw @casblaauw.eurosky.social · 02/10/2026
There's www.micropublication.org, specifically focused on one-figure papers. I've seen some nice stuff in there from the worm community of small work you just wanna get out there. Not sure whether they're open for hypothesis papers, tho - it could fit in their categories?
micropublication.org
microPublication - Get Your Data Out, Be Cited
110 1
Reposted by Cas Blaauw
Dr Hannah M. Rowland @hannahmrowland.bsky.social · 01/10/2026
Hope your day was better than Darwin’s, on this day 165 years ago
I am very poorly today & very stupid & hate everybody &everything.

Charles Darwin
Letter to Charles Lyell
1st October 1861
201100382459 10K
Reposted by Cas Blaauw
VIB.AI @vibai.bsky.social · 30/09/2026
Want to present your work in this room? The VIB Conference on AI & Computational Biology is coming to Leuven, with several talk slots selected from abstracts. Submit yours by 7 Oct: www.vibconferences.be/events/ai-and…
031 3
Reposted by Cas Blaauw
Hubrecht Institute @hubrechtinstitute.bsky.social · 30/09/2026
Group leaders Danny Sahtoe and Alexander van Oudenaarden will supervise 2 of the 15 PhD positions that open in the ORFeus network, a Horizon Europe program studying the #DarkProteome, a group of recently discovered proteins linked to disease and new therapies. Apply 1–22 Oct via orfeus-dn.eu/apply
orfeus-dn.eu
Apply to ORFeus — 15 fully funded PhD positions
15 fully funded PhD positions across Europe, starting 2027. Applications open 1 October and close 22 October 2026. Rank up to three projects.
073 7
Reposted by Cas Blaauw
Uwe Ohler @uweohler.bsky.social · 30/09/2026
Our #genome still holds many mysteries… Want to help figuring out what is actually translated into proteins and how it is regulated? Our research network #ORFeus is looking for motivated #PhD students to join projects across Europe. Deadline is soon (Oct 22), have a look! www.orfeus-dn.eu/apply/
orfeus-dn.eu
Apply to ORFeus — 15 fully funded PhD positions
15 fully funded PhD positions across Europe, starting 2027. Applications open 1 October and close 22 October 2026. Rank up to three projects.
088 8
Reposted by Cas Blaauw
Brendan O’Kane @bokane.org · 15/07/2025
in the process of having an idea
An orange cat with white socks and a sweet heart, emerging slowly from croissant mode on the sofa. He appears to be thinking about something, which he is - sorry, little homie - not really equipped to do.
27146580 1.5K
Reposted by Cas Blaauw
Maitreya Dunham @maitreya.bsky.social · 24/09/2026
Exciting news: @uwgenome.bsky.social will be hiring for two tenure-track faculty positions this year, one open rank and one assistant professor. Both are “Genome Sciences” interpreted broadly so we welcome applicants working in a variety of fields and methods. Ad will be out shortly.
1108105 108
Reposted by Cas Blaauw
Michalis Averof @michalis-averof.bsky.social · 23/09/2026
Nice example of how AI predictions are limited by the training set. AlphaFold has not learned the rules of protein folding, it's just great at predicting structures based on what it has already seen. Added constraints (e.g. alignment revealing conservation patterns) can help make a good prediction.
1106 10
Cas Blaauw @casblaauw.eurosky.social · 23/09/2026
Yes! Corgi is probably the closest comparable model currently out there (and I'm pretty sure we do mention it as such!), but ofc with differences: we explicitly encode the full GRN, making tf2r and r2g links easier to interpret but less capable of modelling non-linear effects.
110 1
Cas Blaauw @casblaauw.eurosky.social · 22/09/2026
This has been a long time in the making (and my part was very small) - cheers to (blueskyless?) Gabriele and the rest of the lab for finally getting this out there.
010 1
Cas Blaauw @casblaauw.eurosky.social · 22/09/2026
Excited to show deepSCENIC to the world! GRNs meet sequence-to-function models: explicit TF-region and region-gene links, with the power of enhancer DL models, predicting RNA & ATAC at single-cell resolution. Check out the examples in the paper or try it on your own data: github.com/aertslab/Dee...
github.com
GitHub - aertslab/deepSCENIC: Deep learning for single-cell Gene Regulatory Networks
Deep learning for single-cell Gene Regulatory Networks - aertslab/deepSCENIC
2145 14
Reposted by Cas Blaauw
bioRxiv Bioinfo @biorxiv-bioinfo.bsky.social · 22/09/2026
DeepSCENIC: transfer learning from sequence-to-function models enables causal gene regulatory network inference www.biorxiv.org/content/10.64898/20…
0116 11
Reposted by Cas Blaauw
John Timmer @jtimmer.bsky.social · 22/09/2026
Ok, this "there are two separate brains" thing appears to be making the rounds, and I want to be absolutely clear: this is a press office overhyping some perfectly reasonable science that won't come as a surprise to most developmental biologists.
32756238 756
Cas Blaauw @casblaauw.eurosky.social · 22/09/2026
Oh don't you worry, it's live and running: askjev.net
askjev.net
AskJev — Have a Question?
Ask Jev to make a choice, give a score, or judge a statement. A loving tribute to the early web.
110 1
Reposted by Cas Blaauw
Matt Hodges @matthodges.bsky.social · 19/09/2026
Very quick and dirty Jev test for hiring bias. One resume for a Wall Street job; asked whether the applicant should get a first-round interview. 76 evaluations, changing only the first name. 19 of each: White-associated men/women, Black-associated men/women. console.typesafe.ai/playground?s...
19474116 474
Cas Blaauw @casblaauw.eurosky.social · 21/09/2026
Finally - so many PIs and CEOs are trying to automate their workers away, now we can turn the tables and automate the boss instead 😉
000 0
Reposted by Cas Blaauw
Stefan Schoenfelder @stefanschoenfelder.bsky.social · 20/09/2026
The number of MSCA Postdoctoral Fellowship applications in 2026 (21,627) has more than doubled since 2024 (10,360 proposals) and more than tripled since 2022 (7,044). Stating the obvious here but this is not sustainable.
01212 12
Reposted by Cas Blaauw
Arnaud Krebs @arnaudkr.bsky.social · 21/09/2026
We are recruiting! Wet&Dry either combined or separate. If you are excited about gene regulation, new methods in genomics and AI-based interpretation! Talk to us!
0126 12
Cas Blaauw @casblaauw.eurosky.social · 21/09/2026
In the end, is it better to have a more reproducable version of papers, where you can rerun code directly? I also think so, and LLMs can probably help. Is there potential for ML-based tools in science? Yes, my entire career focuses on this. None of that is 'The future of reading papers is agents'.
030 3
Cas Blaauw @casblaauw.eurosky.social · 21/09/2026
- Finally, I'm very tired of the 'you have to be positive about AI or the conversation will move elsewhere'. Critical about the effects of AI on science? Sorry, suck it up because we don't listen to critics, only to boosters. It's just confirming that this field refuses to engage with criticism.
120 2
Cas Blaauw @casblaauw.eurosky.social · 21/09/2026
- The response was clearly about the effects of LLMs on science rather than AlphaFold and the like. That's what this paper is about, that's what people have the most problems with, and conflating the two misrepresents the argument of the opponents as much as you feel you've been misrepresented.
120 2
Cas Blaauw @casblaauw.eurosky.social · 21/09/2026
- Posting that talking to chatbots is the future of papers is not just 'a mildly optimistic post', it's a fundamental reshaping of the concept of reading and understanding science. If you feel this is where we should go, own up to that opinion, but don't be surprised if others disagree.
130 3
Cas Blaauw @casblaauw.eurosky.social · 21/09/2026
I understand that being dogpiled sucks, and I'm sure you've gotten all kinds of stupid responses. That said, I do think these arguments also misrepresent the arguments that your fellow scientists are making. Three thoughts:
111 1
Reposted by Cas Blaauw
Christophe 🔬 Leterrier @christlet.bsky.social · 20/09/2026
Acronyms in biology tools these days are either already used by another method or "Here we present sMLFNJCOIMZNJQM"
4456 45
Reposted by Cas Blaauw
Joe Bak-Coleman @jbakcoleman.bsky.social · 20/09/2026
For whatever reason the fact that LLMs manifest their task completion with language has had folks wanting to jump the gun on a big thorny question of whether/when they “think” in the sense that we all have come to understand it. They may simply not need “think” to to complete complex tasks.
723231 232
Reposted by Cas Blaauw
Richard Sever @richardsever.bsky.social · 20/09/2026
"All corners of academia will soon be flooded with AI-generated research....Maths and CS go first because their entire output is symbol manipulation...Biology comes later not because biologists are cleverer but because the rate-limiting step is physical" giorgio.gilest.ro/the-weimar-o...
giorgio.gilest.ro
The Weimar of knowledge – la bottega
56641 66
Reposted by Cas Blaauw
Christophe 🔬 Leterrier @christlet.bsky.social · 19/09/2026
I asked chatGPT and - Well I asked Toko the starling and he said he loves me so much and that's everything I need
071 7
Reposted by Cas Blaauw
Eris ⌬ @eriskii.net · 18/09/2026
"Pope gifted green orb" "Trump threatens EU over Canada" "Chipotle partners with Palantir" What the fuck is going on anymore they're just making this shit up now
12529101 529
Reposted by Cas Blaauw
Nicola Rennie @nrennie.bsky.social · 17/09/2026
I love overly honest READMEs
Screenshot of GitHub README saying known issues: coding style only a mother could love
19921 99