New preprint from the lab, accepted at @neuripsconf.bsky.social NeurIPS 2025! lnkd.in/eKVNGVNU Most AI models treat all inputs the same. We show how to adaptively route each sample across modalities and tasks, boosting performance, interpretability, and clinical impact for heterogenous data.

"JAMA enthusiastically launches JAMA+ Women’s Health as a curated source of trusted clinical information to help clinicians advance health care for women worldwide." 🔗 Read this Editorial to learn more: ja.ma/46kacf7 📌 Explore JAMA+ Women’s Health: ja.ma/42ifhSF

Launched in 2023, Imaging Neuroscience is now firmly established, with full indexing (PubMed, etc.) and 700 papers to date. We're very happy to announce that we are able to reduce the APC to $1400. Huge thanks to all authors, reviewers, editorial team+board, and MIT Press.

🤖🧠 Paper out in Nature Communications! 🧠🤖 Bayesian models can learn rapidly. Neural networks can handle messy, naturalistic data. How can we combine these strengths? Our answer: Use meta-learning to distill Bayesian priors into a neural network! www.nature.com/articles/s41... 1/n

1/12 How do animals build an internal map? In our new paper, we tracked thousands of mouse CA1 neurons over days/weeks as they learned a VR navigation task. @nspruston.bsky.social & co-1st author @johanwinn.bsky.social Video: www.youtube.com/watch?v=yw_4... Paper: www.nature.com/articles/s41...

Now that the first Advisory Council meeting will be announced in the Federal Register tomorrow (National Institute for Complementary and Integrative Health) tomorrow, I want to clarify something that has been happening for at least the past 30 days. 1/n

Beyond Mechanism—Extending Our Concepts of Causation in Neuroscience onlinelibrary.wiley.com/doi/10.1111/... - really pleased that this paper with Henry Potter is now published in the European Journal of Neuroscience 😊

Re the Tononi paper: Both Tononi’s IIT (phi) and Friston’s FEP start from fundamental, axiomatic, and debatable assumptions. These assumptions are generally made without any humility. This logic allows them to make exceptionally broad claims. Which contributes to my unease about them.

Decoding the rhythmic representation and communication of visual contents www.cell.com/trends/neuro... #neuroscience

⚡️ Excited to introduce ZAPBench, our #ICLR2025 spotlight: The Zebrafish Activity Prediction Benchmark measures progress in predicting neural activity within an entire vertebrate brain (70k+ neurons!) Explore interactive visualizations, datasets, code + paper: google-research.github.io/zapbench 🧠🧪

Can we use brain scans to identify biologically distinct "autisms"? And If so, how? In our latest cross-species study 🐭👨 with @iitalk.bsky.social we dive into these questions! Find out more here 👇👇 www.biorxiv.org/content/10.1... Thread below 🧵 1/n

One of the NIMH PIs whose name is the termination list: Sooyhun Lee, whose lab just published this beautiful @nature.com paper (that has received too little press because comms at NIH are down). Read! Cite! www.nature.com/articles/s41...

Should you stick to your goal, try something else, or give up? Your median raphe nucleus in the brainstem knows and will decide for you 😉. First foray of my lab into foraging, behavioural strategies and exploration. Amazing work from the one and only Mehran Ahmadlou: www.nature.com/articles/s41...

The Society for Neuroscience selected the paper on this tool for an upcoming webinar on FlexRig and open-source tools in general. The webinar will be on March 25 from 11:45 am – 1 pm ET. Here is the registration page for this webinar: neuronline.sfn.org/scientific-r....

Nice story on fabulous new work from @dingliu.bsky.social in my lab, just out today!...

New work accepted to #AISTATS2025 ✨ We developed a new explainable AI method to better understand time-series data. We can map back inputs (neurons) to their latent space. Basically merging #CEBRA with measuring the individual neurons contribution over time arxiv.org/abs/2502.12977

Evo 2 Can Design Entire Genomes https://www.asimov.press/p/evo-2 🧬🖥️🧪

Can whole-brain imaging of cellular c-Fos signals be used to classify #psychedelics? 🧠🔬💊 Our study by @aboharbf.bsky.social and @pashadavoudian.bsky.social, developing an imaging and machine learning pipeline to test a panel of psychoactive drugs. Paper here 👇 www.nature.com/articles/s41...

Cells in the hippocampus encode not only information about a fellow bat’s position, but also its identity and distinct social features. By @claudia-lopez.bsky.social #neuroskyence www.thetransmitter.org/social-behav...

delighted to share our work revealing a new type of excitatory hippocampal neuron we call the "ovoid cell", which has really exquisite properties relative to adjacent pyramidal cells! superb work of first-author @adriennekinman.bsky.social and many others in the lab www.nature.com/articles/s41...

Two cool papers on the HPC out today in @nature.com: 1. Replay during non-REM sleep switches between prioritizing recent versus remote memories: www.nature.com/articles/s41... 2. All optical obs. of synaptic plasticity rules during place field formation: www.nature.com/articles/s41... 🧠📈 🧪

Congratulations @kaidikang.bsky.social et al on their paper in Nature. Also proud of the team at #MIDB in the support! @tervoclemmensb.bsky.social @bart-larsen.bsky.social and more! If you are interested in optimizing study designs, take a look, it's a great read! www.nature.com/articles/s41...

Together with some friends we built an optical Brain Machine Interface to see if cortical activity (PPC) can control behaviour in real time, without animals having to learn to use it. Here's what happened: www.biorxiv.org/content/10.1...

Stellar new work lead by the inimitable James Whittington in Neuron that develops a new theory unifying episodic and working memory and explains diverse hippocampal and prefrontal data: www.cell.com/neuron/fullt... w/Will Dorrell, @behrenstimb.bsky.social Mohamedy El Gaby

Exciting news from the lab! In collaboration with our #EU-Nanobrigth partners, we’ve developed and tested a new probe for vibrational fiber photometry, enabling minimally invasive Raman spectroscopy deep into the 🧠. www.nature.com/articles/s41... @naturemethods.bsky.social

Where do the dopamine hits does from? A new reward center in the brain is discovered (yes, single cell, spatial omics at work again) @science.org www.science.org/doi/10.1126/... www.science.org/doi/10.1126/...

Thanks to everyone's feedback, we have updated our calculator to optimize sample size N & scan time T for fMRI studies: leonoqr.github.io/ORSP_Calcula... The first new feature is that users can explore how different N & T leads to different accuracy, e.g., N=1000 & T=30min => 81% max accuracy. 🧵

Out now in Nature Human Behaviour: Our 68-country #survey on public attitudes to #science 📣 It shows: People still #trust scientists and support an active role of scientists in society and policy-making. #OpenAccess available here: www.nature.com/articles/s41... @natureportfolio.bsky.social (1/13)

Appreciate this generous write up on our recent preprint! www.thetransmitter.org/spectrum/hai...

Ten simple rules for good model-sharing practices https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1012702

Three distinct gamma oscillatory networks within cortical columns in macaque monkeys’ area V1 www.frontiersin.org/journals/neu... #neuroscience

New results for a new year! “Linking neural population formatting to function” describes our modern take on an old question: how can we understand the contribution of a brain area to behavior? www.biorxiv.org/content/10.1... 🧠👩🏻‍🔬🧪🧵 #neuroskyence 1/

Delighted to highlight our new paper in @natureportfolio.bsky.social entitled 'Synaptic basis of feature selectivity in hippocampal neurons' (thread below 👇): www.nature.com/articles/s41...

Today in "Things That Make C. elegans Disqualifying Weird But Also Somehow Make the Human Brain 'Special'" www.sciencealert.com/a-first-of-i...

Thanks to the WCM newsroom for this very nice writeup of the lab's papers at NeurIPS @neuripsconf.bsky.social medicalxpress.com/news/2024-12...

What measures in brain/mind research could benefit from standardization? In the era of big data, it helps to measure the same things in the same ways. Ex: 3D protein structure > AlphaFold. But standardization of depression measures did not happen until 2019: www.thelancet.com/journals/lan...

Slides for "Table Foundation Models" I explain why these models can strongly outperform tree-based models, what are the intuitions, hopefully pointing to ways forward for more improvement speakerdeck.com/gaelvaroquau...

The lab's third paper at NeurIPS @neuripsconf.bsky.social, first authored by Mason Hargrave, introduces a first benchmark for reinforcement learning in healthcare settings. neurips.cc/virtual/2024...

Our second paper at NeuIPS, first authored by Isaac Osafo Nkansah, properly extends CNNs to graph-structured data. Our QuantNets capture complex patterns in irregular structures—like molecular graphs, fluid simulations, and neuroimaging—just as CNNs do for images. neurips.cc/virtual/2024...

Very excited for the lab to be at NeurIPS this year with three projects. The first, by very talented postdoc @amandambuch.bsky.social introduces a simple, scalable approach to jointly cluster and embed high-dimensional biomedical data.

Continuing to go through my trove of cool videos... here is a classic from my work with Philipp Keller's team at @hhmijanelia.bsky.social A developing fly embryo imaged with a SiMView #lightsheet microscope augmented with our Autopilot self-alignment system. www.nature.com/articles/nbt... 🧪🔬🪰💻

Entropy is one of those formulas that many of us learn, swallow whole, and even use regularly without really understanding. (E.g., where does that “log” come from? Are there other possible formulas?) Yet there's an intuitive & almost inevitable way to arrive at this expression.

V. proud of this paper on "cell type dynamical systems", to appear at NeurIPS 2024 — led by the fearless @aditijh.bsky.social !!

I always found this paper kind of nice: www.science.org/doi/10.1126/... it shows that if the environment fluctuates then evolution will favor sensors and the mutual info between sensors and env increases fitness. If the env has memory, I suspect organismic memory will provably be selected for also

Well that's a spanner in the works. Transient CaMKII inhibition in the amygdala blocks the short term memory (1h) but the long term memory (>1 day) still forms!!!! Brilliant work from the Yasuda lab #neuroscience #biophysics www.nature.com/articles/s41...

🔊 Exciting AI Research Opportunity! my team at @czbiohub SF is hiring a Research Scientist to develop foundational AI models for BioImage analysis. Join us to work on cutting-edge ML/CV! 💻🔬🧪🧠 #AIJobs #ML job-boards.greenhouse.io/chanzuckerbe...