We're excited to share our new publication, "A unifying Bayesian framework for merging X-ray diffraction data," live now in Nature Comm! In this paper, Kevin Dalton and Jack Greisman present "careless", a python package for Bayesian scaling and merging nature.com/articles/s4146…

careless is part of the "Reciprocal Space Station" family of open-source software for analysis of diffraction data. Learn more about all of the RS-Station packages here: rs-station.github.io

Our #Atmo Bay Area Forecast is accurately reflecting heavy rain, wind, & swell accompanied by flooding and power outages in affected areas. Be sure to watch the #bayarea storm forecast at: sf.atmo.ai/precipitation-… Alex Levy Johan Mathe #cawx #bombcyclone #sfbay

Lights on(!) coincides with our first #ultrafast #LaserPhasePlate for #cryoET paper by Daniel X. Du Cell Reports Methods #OpenAccess: bit.ly/fitzpatricklab… #femtosecond #laser #optics #UEM #SEM #cryoEM experiments made possible by generous support from CZI Science #ImagingTheFuture

A new manuscript from our group is just online: "Slicing and Dicing: Optimal Coarse-Grained Representation to Preserve Molecular Kinetics" pubs.acs.org/doi/10.1021/ac… We propose a bottom-up coarse-graining scheme that correctly preserves the relevant slow degrees of freedom.

Excited to present our preprint on the structural basis of complement receptor (C3aR) activation by an endogenous and synthetic ligand. Marvelous collaboration with the one and only Arun K. Shukla & Co IIT Kanpur. Stay tuned! Ravi Yadav USC Dornsife College of Letters, Arts & Sciences USC Bridge Institute biorxiv.org/content/10.110…

How can we use insights from industry to improve error culture in structural biology? Read in our newest paper! scripts.iucr.org/cgi-bin/paper?…

Tune into our next #SLACPublicLecture on 4/6, when Axel Levy will share how AI can help us visualize complex sequences of motion within cells. Event page: stanford.io/40DngXa

‼️Tomorrow (4/6)‼️ How can artificial intelligence construct 3D models of molecules from millions of enigmatic images? Join the #SLACPublicLecture, where Axel Levy will talk about this rapidly developing technology. Event link: stanford.io/3GdHz4s

Check out Axel Levy's SLAC National Accelerator Laboratory public lecture on our work for an accessible introduction to cryo-EM, AI, and structural biology! 👏👏👏 "Capturing molecular motion using artificial intelligence" youtube.com/watch?v=cuSC7L… Frédéric Poitevin Gordon Wetzstein

Honored to be named Hellman Fellow! Esp grateful to extraordinary lab members Mathilde Papillon🦋 mathildepapillon .bsky .social Adele Myers Francisco Acosta Bongjin Koo Sophia Sanborn David Klindt Christian Shewmake who made it happen! 👉We create cutting-edge AI to reveal the geometries of intelligent life🌐🍩

How do 3D reconstruction algorithms perform under very low SNR?🧐 At 11.40am tomorrow at the CCD workshop #CVPR23: I'll compare inference methods in physics-inspired deep generative models for 3D reconstruction of nanoscopic structures from 2D #cryoEM images🧬 See you there?☺️

Congratulations Tatyana Bodrug! Check out our work on reconstructing time-resolved conformational landscapes of the APC/C complex with cryoDRGN ❄️🐉. Joint with Brown lab UNC Pharmacology and Haselbach lab, now published in NatureStructMolBiol nature.com/articles/s4159…

Preprint alert: Superposition in CNNs and Brains Context: Recent work on LLMs showed how sparse coding can recover interpretable representations in transformers Anthropic TL;DR: Concurrently, we have been doing similar analyses in vision models and brains.

America is home to the brightest, fastest, most powerful X-ray laser in the world thanks to SLAC. Yesterday, we celebrated the upgraded Linac Coherent Light Source (LCLS-II) and a new wave of scientific development from quantum materials to clean energy tech and medicine. 1/

Happy to share our latest preprint Towards Interpretable CryoEM: Disentangling Latent Spaces of Molecular Conformations Thanks to my amazing collaborators Frédéric Poitevin Nina Miolane 🦋 @ninamiolane.bsky.social Axel Levy and Aapo Hyvärinen 🎉 biorxiv.org/content/10.110…

Congrats David! This is a really cool endeavor 👏

We present DRGN-AI for fast, ab initio cryo-EM reconstruction! * learns a neural field from unposed images, * designed for single-shot reconstruction of unfiltered datasets, * finds new states missed by prior approaches! Teamwork led by Ellen Zhong drgnai.cs.princeton.edu 1/

Protein generative models have learned patterns across 3D structural space. Can we leverage this capability for inverse problems? With ADP-3D (Atomic Denoising Prior), we solve inverse problems in protein space using a diffusion model as a prior! axel-levy.github.io/adp-3d/