Ekin Deniz Aksu

I'm excited to share that Corgi is out in @natcomms.nature.com! We refine context-aware sequence-to-function models by using FiLM to condition the model on trans-regulator expression levels. This way Corgi can generalize to unseen cell types. www.nature.com/articles/s41... 🧵Thread 1/n

The achievement of Serbian students overthrowing Aleksandar Vučić can't be overstated. In 2024, they began protesting. In 2025, they transformed a protest movement into civil dissidence. In 2026, they became a political party — and two months later, crushed a corrupt state-entrenched authoritarian.

Learning how single cells respond to perturbations In this issue of Molecular Cell, we wrote a Spotlight with @bergerlab.bsky.social on the recently published State model, where we briefly discuss open problems in AI-based perturbation response prediction. authors.elsevier.com/c/1nx~I3vVUP...

We really need to make Meng's Big Data Paradox central to teaching, don't we? Researchers don't know this stuff bc they have never been exposed to it. But it is fundamental. (Link to paper PDF statistics.fas.harvard.edu/sites/g/file...)

Cerberus: bidirectional state space blocks improve accuracy and efficiency of regulatory sequence models https://www.biorxiv.org/content/10.64898/2026.10.02.756377v1

I recently posted the first output of our partnership with Zack Smith's lab, here I post it again this time with the other 2 manuscripts that were generated from this partnership thanks to the fantastic postdoc Yusuke Kijima: www.biorxiv.org/content/10.6...

In our Nature paper, we introduce the first superhuman Stratego AI, which we built using general techniques that we developed for RL & test-time compute under imperfect information. www.nature.com/articles/s41...

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/

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...

Want an efficient way to delete cis-regulatory elements in stem cells? 🧬 Our lab just published a methods paper in JoVE's collection on 2D/3D Stem Cell-Based Models of Early Mammalian Development. shorturl.at/QB1xo Thank you @ancheldjs.bsky.social for the invitation to contribute!

Academia has seen inflation in the past 30 years but it is now entering Republic of Weimar territory, with value of knowledge moving from "paper to land". In all this, Open hardware is an act of armed resistance to Technofeudalism. giorgio.gilest.ro/the-weimar-o...

I resigned from ExxonMobil today. I spent the last three years doing fossil fuel extraction research at both BP and ExxonMobil. Neither company is acting responsibly. They are racing to maximize extraction while the window to act closes, and gambling with our lives. More below.

In 2023, @nature.com published 'Papers and patents are becoming less disruptive over time', receiving world-wide media attention. Our Matters Arising, published after a 32 month delay (more on that soon), shows that the reported decline can largely be attributed to dataset artefacts. 🧵

"Here we describe a systematic effort (‘Codebook’) to determine the sequence specificity of 332 putative and poorly characterized human TFs." www.nature.com/articles/s41...

In a new study in Nature Communications, Martin Vingron and @ekindea.bsky.social present "Corgi", an AI-based tool that allows researchers to predict how the genome behaves across different cell types. Read more: www.molgen.mpg.de/news-2026-le... Original publication: doi.org/10.1038/s414...

1/ I'm excited to share that we're launching NECB 2026, the inaugural New England Computational Biology Symposium. Oct 1-2 at Microsoft Research New England, Cambridge. Two days of keynotes,talks, and posters to bring our community together across institutions.Space is limited. newenglandcompbio.org

My first PhD paper is out in Nucleic Acids Research! We describe "Ab-trapping" - an antibody artifact that distorts assays relying on antibody diffusion (microscopy, CUT&Tag, CUT&RUN). The revisions made the story much stronger. Check it out! doi.org/10.1093/nar/...

I have been thinking since yesterday about why I find the 1% thing so, so disturbing. It's a visceral reaction of 'just no', which goes beyond the general yuck towards the enshittification of/via AI. I have finally put my finger on it: it is the feeling you get when

(1/n) Very excited to share tri-lab collab (Mirny & Zechner) led by Harvey, Henrik & Jack: Q: How do enhancers & promoters interact in space (contact vs. action-at-a-distance) and time (stable vs. transient)? A: Transient E-P contact (~25-42 nm lasting ~10-20 sec): www.biorxiv.org/content/10.6...

A new Science study shows that bumble bees can position a ball underneath a fake “flower” to reach a reward, suggesting they can exhibit spontaneous problem-solving and challenging the notion that such advanced cognitive abilities are exclusive to large-brained vertebrates. https://scim.ag/4vs08dC

Exciting breakthrough technology from the lab, now live in @cellcellpress.bsky.social ! Instead of cutting the genome where proteins bind (e.g., Cut&Tag), D&D-seq scars the DNA with a deaminase, allowing single cell genome mapping of TFs and chromatin remodellers!

This is insane.

TL;DR: We've identified more than 100 cases of apparent manipulation in Thermo Fisher Scientific's antibody verification data. @sholtodavid.bsky.social @johanduchene.bsky.social reeserichardson.blog/2026/05/28/h...

🔬 Latest research in Bioinformatics Advances: "Modeling strategies for in vivo transcription factor binding predictions"  Read it here: https://doi.org/10.1093/bioadv/vbag123 Authors include: @ekindea.bsky.social

Very excited to share our newest findings! A very long team effort pursued by multiple team members -to frantically reproduce everything over and over-, we show mTOR-Polycomb synergy at promoters of developmental master regulator TFs in ESCs prior to their expression. www.biorxiv.org/content/10.6...

Does this mean the end of the central dogma?

I am so excited to share our new findings with you! We provide the structural evidence for a direct protein-to-DNA information pathway, showing how a bacterial enzyme 'reads' its own structure to 'write' DNA. www.science.org/doi/10.1126/...

Spectral multiplexing is typically limited to 4-5 channels. Our new preprint introduces a framework that utilizes DNA barcoding and signal tuning to enable robust spectral unmixing and ground-truth benchmarking to achieve 15-plex subcellular profiling without cycling www.biorxiv.org/content/10.6...

🚨 Why can’t mammals regenerate limbs like frog tadpoles or salamanders? In our new paper in @science.org , we show that species-specific oxygen sensing acts as a gatekeeper for initiating limb regeneration 🐭🐸 🔗 www.science.org/doi/10.1126/... #EvoDevo

MPRAs are the gold-standard tool for measuring how DNA sequences drive gene expression and prioritizing variant effects. In this preprint we asked: does it matter WHERE you place a variant in an MPRA? Spoiler: yes, and it might lead you to miss disease-causing variants. 1/6 doi.org/10.64898/202...

Ever wondered how a eukaryotic transcription factor finds its specific DNA motif in the vast genome? In this preprint, we directly measured the dynamics of this search process in living cells, revealing a cooperative mechanism mediated by disordered regions. 1/10 doi.org/10.64898/202...

New OpenFold3 preview out! (OF3p2) It closes the gap to AlphaFold3 for most modalities. Most critically, we're releasing everything, including training sets & configs, making OF3p2 the only current AF3-based model that is functionally trainable & reproducible from scratch🧵1/9

🧬 How do immune disease-relevant variants affect gene regulatory networks in CD4+ T cells? 🧪🖥️ We coupled two large-scale CRISPRi screens (>4M cells) to map the downstream cascades of thousands of SNPs More details in the thread below 👇 or in the article 📖 on biorxiv tinyurl.com/CD4screens

Some molecular machines, like ribosomes, can persist for long periods of time in cells. Could molecular aging of ribosomes shape how proteins are made? In our new preprint we track ribosomes as they age in cells and uncover unexpected effects on translation (1/10) www.biorxiv.org/content/10.6...

New online! Gene regulatory networks: from correlative models to causal explanations

www.science.org/doi/10.1126/... This is clever, very clever