Understanding where genomics models work and fail will be critical to their successful deployment. Meta models can tell us this, and give some insight into failure modes. 3/3 for @muntakimrafi.bsky.social this week. CSHL AIxBIO to boot (sorry to miss!) What a week!

Which active learning methods are best for genomics models? See Rafi's thread on our recent preprint, below. www.biorxiv.org/content/10.6... Rafi is having a good week - he's got one more preprint to post!

There is widespread homology across our genomes. How does this impact genomics models? Major update to our preprint! 🎉 We tested more models and more tasks; the overall findings were robust, but one conclusion on how future models should be trained differs⚠️ www.biorxiv.org/content/10.1...

gRely: Relyability for genome trained sequence-to-expression models https://www.biorxiv.org/content/10.64898/2026.05.23.727431v1

Our Decima paper is now published in Nature Methods! Full text: rdcu.be/fkASS. Decima can predict gene expression in specific cell types and cell states from DNA sequence. I would say more but @drkbio.bsky.social and Johannes Linder wrote this great news & views: www.nature.com/articles/s41...

CIHR has launched a trial ballon that suggests shifting to one competition a year. This is the worst idea I have heard in a long time. The only benefit is making top-line success rates look better than they are. Keep it as is, do in-person reviews, and keep the pain in the open.

Why can't we explain enhancer action despite 2 decades of chromosome conformation technologies? 😬 Our new study spearheaded by Leonid Mirny's group points to a flaw in our assumptions, and to a solution from physical principles By @timothyfoldes.bsky.social 💻& @karissalhansen.bsky.social 🧪 🧵👇

I think the LLMs finally found me. Got like 5 review requests this weekend.

🚨 New preprint! How does noncoding variation drive complex traits? We built a massive atlas of cattle regulatory variants (>150k emVars!) and directly tested them in both bovine AND human cells. 🐄🧬🚶‍♂️ Read it here: www.biorxiv.org/content/10.6...

I think both journals and authors vastly underestimate the reputational harm that comes from sending bad papers out for review. Demonstrates disrespect for our time, and poor judgement. I remember it next time I am asked to review, and will rethink submitting to that journal.

It's been a long time coming to finally see this in print! While not the first paper from my lab, it is one of the very first projects we started working on from day 1 links.springernature.com/f/a/1IP7gCfA...

"In this study, we present a genome-scale map of genetic interactions in the human haploid cell line HAP1, based on CRISPR-based perturbation of ∼4 million gene pairs. The resulting network comprises ∼89,000 high-confidence gene-gene interactions" doi.org/10.1016/j.ce...

Thrilled to have my first PhD paper with @mmdesai.bsky.social out! @science.org Link: www.science.org/doi/10.1126/... Thread below: #mevosky #evosky #evobio #popgen #evolsky #ecoevo #molevol

New preprint @cxqiu.bsky.social @jshendure.bsky.social ! Can we learn regulatory grammars of human cell types — by training on mouse development and transferring across 241 mammalian genomes? Introducing STEAM & a whole-organism scATAC-seq atlas from E10 to birth. www.biorxiv.org/content/10.6...

Genetic background shapes AI-predicted variant effects [new] ...uncovering heterogeneous predictions where variants shift between pathogenic and benign outcomes across diverse genomic contexts, crucial for interpretation.

It is with great sadness that I'm sharing the news that Dr. Don Moerman passed away this weekend. He was such a legend in the #Celegans and Canadian science community.

The @steinaerts.bsky.social lab published CREsted, an end-to-end modeling framework to 🧬 Train sequence-based enhancer models on large sc datasets 🔍 Decode enhancer logic with nucleotide-level interpretability ⚙️ Design synthetic enhancers with cell-type specificity https://tinyurl.com/ypurmrw5

“My patient is refusing a drug studied in 170,000 people because of side effects that a 124,000-person analysis just confirmed do not exist — while injecting a compound studied in 14 humans, from unregulated sources, based on the recommendation of someone who profits from selling it.”

Stoked to announce that I’m starting a lab as an Assistant Professor in the School of Biomedical Engineering and Centre for Brain Health at UBC! mcdiarmid-lab.sbme.ubc.ca Recruiting at all levels - please share the word! Cheers and happy long weekend all!

AITHYRA is recruiting an AI/ML STARTING PI Generative Models, Large Language Models, Agentic AI, Active & Reinforcement Learning, AI-driven Scientific Discovery & Lab Automation, ML-driven molecular simulations... Job Ad: aithyra.at/fileadmin/do... Apply by 30.4.2026 #StartingPI #AI/ML #AITHYRA

AITHYRA is recruiting a LIFE SCIENCE PI We are interested in research questions related to one of our four core disease areas: cancer research, immunological disorders, infectious diseases, neurodevelopment disorders. Job Ad: aithyra.at/fileadmin/do... Apply by 30.4.2026 #StartingPI #LS #AITHYRA

"Beyond the baseline: mapping the context-specific regulatory landscape of disease" by Yoav Gilad and Alexis Battle authors.elsevier.com/sd/article/S...

oh my god I cannot stop laughing x.com/om_patel5/st...

New preprint from our lab where we learn some interesting things about the context specificity of TFs in their regulation of gene expression by dissecting MPRA-trained sequence-expression models. Surprising guest appearance from RNA Pol III(!) Thread from 1st author @sambean12.bsky.social 👇

Interallelic cis-regulatory dominance promotes robustness and evolutionary innovation https://www.biorxiv.org/content/10.64898/2026.03.17.712157v1

The Programmable Genomics Lab at @umasschan.bsky.social just officially launched our site! We have a "simple" goal: to develop synthetic regulatory elements that target every cell type in every tissue, and control payload dosage/duration Check it out! programmable-genomics.github.io

New paper showing that much of the apparent success of protein language models in predicting mutational effects is a mirage: These models mostly memorize sites. 1/ www.biorxiv.org/content/10.6...

How do chromatin mark control TF activity and gene regulation? I will unfortunately not be at the @cshlmeetings.bsky.social Systems Biology this year, but very excited that @mpalamin.bsky.social will present her work @embl.org. Spoiler it involves a novel multiplexed single molecule genomics assays!

Abstract deadline approaching soon (March 6)! We welcome both short talks and posters. Travel fellowships up to $500 available for students & postdocs.

We wrote a perspective "How to build the regulatory genome: a constructionist guide to the cis-regulatory code", out in Development yesterday. Title says it all. Find it here: journals.biologists.com/dev/article/...

Congrats to the USA on this milestone www.theglobeandmail.com/gift/2109fef...

Exited to share Current Opinion review on how chromatin hubs involving multiple enhancers and promoters are formed, and their potential roles in gene regulation: www.sciencedirect.com/science/arti...

I used Claude Opus 4.5/4.6 (and a bit of Codex GPT-5.3) to port edgeR to Python. See edgePython github.com/pachterlab/e... This allowed me to develop a single-cell DE method that extends NEBULA with edgeR Empirical Bayes. All in one week. Details in doi.org/10.64898/202...

Registration now open for SynBio 8.0, June 15-17 at the University of Waterloo. www.synbiocanada.org/synbio8 Join us at Canada’s premier synthetic biology conference, bringing the community together to showcase advances from across the country and across the full breadth of the field! #synbio8 🍁🧬

Check out our new protocol paper on molecular recording of transcriptional events in genomic DNA! Super fun pulling this together with @jennynathans.bsky.social @chenomics.bsky.social and @jshendure.bsky.social

🧵 1/ 🎉New paper alert! Pretrained protein language models (#pLMs) are all the hype, but are they really helping us predict protein- protein interactions? 🤔Dive into our thread to see why you should read the full study @natmachintell.nature.com. ⬇️ 🔗 rdcu.be/e3PGD

The @steinaerts.bsky.social lab is looking for a postdoctoral researcher to develop next-generation sequence-to-function models for glioblastoma, one of the most aggressive brain cancers. More info & how to apply 👉 https://vib.ai/en/opportunities#/job-description/130090

Polygenic scores are useful for early detection, but also early prevention! New paper shows people with high genetic risk of T2D have 3x more risk of developing it by age 35. Early detection could trigger lifestyle changes, metformin prescription... doi.org/10.1038/s414...

We are hiring a staff member to support various projects in our lab. Please reach out with questions! wd5.myworkdaysite.com/recruiting/u...

AlphaGenome is out in @nature.com today along with model weights! 🧬 📄 Paper: www.nature.com/articles/s41... 💻 Weights: github.com/google-deepm... Getting here wasn’t a straight path. We discussed the story behind the model, paper & API in the following roundtable: youtu.be/V8lhUqKqzUc

Our paper on the evolution of XIST regulatory network in primates is now published in Science Advances! Check out the paper www.science.org/doi/10.1126/... or a digest of our findings emmanuelczt.github.io/posts/2026/0... A short 🧵 of our main findings 👇

We're testing AlphaGenome. It's a very valuable step forward towards the goal of identifying functional non-coding variants (FNCVs) causing human diseases. Congrats to @avsecz.bsky.social and the team for this landmark publication. www.nature.com/articles/s41...

Happy to share that our review on mammalian genome writing with Jack Atwater, Ran Brosh, Jef Boeke, @jshendure.bsky.social and Matt Maurano is now out in @cellpress.bsky.social! www.sciencedirect.com/science/arti...

Deadline TODAY January 23 - please join the Systems Biology: Global Regulation of Gene Expression meeting in beautiful @cshlnews.bsky.social (March 11-14). Register and submit your abstract here: meetings.cshl.edu/meetings.asp.... Please Repost!

Metabolites can be quantified using a combination of aptamers and DNA barcodes go.nature.com/42BbXTQ rdcu.be/eZ8BP

TF-MINDI is out! A new method to learn cis-regulatory codes through rich embeddings of TF binding sites. TF-MINDI decomposes motif neighbourhoods, and works downstream of any sequence-to-function deep learning model. We deeply study the enhancer code in human neural development, check out the thread

Dear American Friends, I’m here with an uncomfortable message: watching in horror is not good enough. Sitting idly by makes you complicit. Things will keep getting worse until Americans act.

1/ We found that deep sequence models memorize atomic facts "geometrically" -- not as an associative lookup table as often imagined. This opens up practical questions on reasoning/memory/discovery, and also poses a theoretical "memorization puzzle."

Vaccines, the most impressive public health intervention in medical history, and where we could be headed if there was not efforts to negate truth, facts, and evidence A great, open-access, review and perspective by @scientificdiscovery.dev

Looking to start your lab in generative biology / AI? Come join us at the @sangerinstitute.bsky.social Sanger is core-funded so you can generate data at scale to train the next generation of models and understanding. Design/Engineering/Chemistry/Proteins/Pathways! pls RT tinyurl.com/GenGenFaculty