@moleculemaker.bsky.socialNew in Nucleic Acids Research: CLEAN Database, with 55M+ AI-predicted enzyme functions across ~1.1M strains, plus an ESM-2 mutation-effect explorer with heatmaps and 3D views. Free, no login.
Congratulations, Le Yuan, Huimin Zhao and team at Illinois! doi.org/10.1093/nar/... #AIforScience #synbio
@xtalpi.bsky.socialWe've made lcms2yield available on GitHub - along with the code, model weights, and datasets. We hope others working on automated chemistry find it useful and build on it.
GitHub: github.com/xtalpi-inc/...
Paper: pubs.acs.org/doi/10.1021...
@openbind.bsky.socialThe #OpenBind Blind Challenge is now LIVE!
Can your model predict ligand binding poses for 400 compounds against the Zika virus NS2B-NS3 protease?
Join the challenge, submit predictions and see how your method performs. https://loom.ly/1aQdU3U
@ishiharaksk.bsky.socialSo excited to build this with @mweber.bsky.social and @hattaca.bsky.social ! Together weโre combining geometric AI, multi-species genomics, and human cardiac organoids to ask why some animals regenerate their hearts and we canโt. Grateful to #Googleorg for the support. #AIforScience
@aidailypost.comFrom climate modeling to fusion breakthroughs, 17 orgs are supercharging research with AI. See how Google.orgโbacked labs and AI agents are reshaping scienceโand why it matters for the next big discovery. #AIforScience #MachineLearning #MaterialsDiscovery
@byteandpieces.bsky.social๐ฃ New Podcast! "AI Just Made Science Obsolete: The Truth Behind ASI & The Eureka Machine" on @Spreaker #agi #aiforscience #aihype #aipodcast #aitruth #artificialintelligence #asi #deeplearning #fullstackai #futureofscience #innovation #machinelearning #peterdiamandis #recursiveselfimprovement
@antifocal.bsky.socialNetworks trained on simulated images needed retraining on labelled real ones, about 30 min on one GPU. People set the drift limit, steered the models and ended each run.
Roccapriore et al., Science Advances 10, eadn5899 (2024)
Values are representative.
@antifocal.bsky.socialErrors from X and Y coils, never varied, were corrected least well. Very large distortions stay partly uncorrected. People ran the calibration and remove big solvent peaks first.
Jopa et al., J. Phys. Chem. B 129, 6078 (2025)
Values are representative.
@berkeleycdss.bsky.socialUC Berkeley researchers have been awarded $1.7M from Google.org to advance medical discovery using AI.
๐ Read more: https://bit.ly/4hUBn4H
@openbind.bsky.socialThe #OpenBind Blind Challenge is now LIVE!
Can your model predict ligand binding poses for 414 compounds against the Zika virus NS2B-NS3 protease?
Join the challenge, submit predictions and see how your method performs. https://loom.ly/1aQdU3U
@plasmaphysik.bsky.socialThe team led by Prof. Dr Josefine Proll and colleagues from the IPP and Eindhoven University of Technology (@tue.nl) has been selected for the #Googleorg Impact Challenge: #AIforScience. For the STELLAR-AI project, they will receive funding of 1.25 million US dollars.
@fjhheras.bsky.socialRead more about Gemini Argon and our results in the blog post: goo.gle/3W85sXj
#QuantumComputing #QuantumErrorCorrection #AlphaQubit #AIforScience #GoogleDeepMind
@thdept-fhi-mpg.bsky.social๐จCheck out the new publication from our group on npj Computational Materials @nature.com. ๐๐ ๐ผ๐ฝ๐ฒ๐ป๐ ๐ป๐ฒ๐ ๐ฝ๐ผ๐๐๐ถ๐ฏ๐ถ๐น๐ถ๐๐ถ๐ฒ๐ ๐ณ๐ผ๐ฟ ๐บ๐ฎ๐๐ฒ๐ฟ๐ถ๐ฎ๐น๐ ๐ฑ๐ถ๐๐ฐ๐ผ๐๐ฒ๐ฟ๐!
@fhi-mpg.bsky.social๐งช ๐๐ฎ๐ป ๐ด๐ฒ๐ป๐ฒ๐ฟ๐ฎ๐น-๐ฝ๐๐ฟ๐ฝ๐ผ๐๐ฒ ๐๐ ๐ฎ๐ฐ๐ฐ๐ฒ๐น๐ฒ๐ฟ๐ฎ๐๐ฒ ๐บ๐ฎ๐๐ฒ๐ฟ๐ถ๐ฎ๐น๐ ๐ฑ๐ถ๐๐ฐ๐ผ๐๐ฒ๐ฟ๐?
Researchers from our @thdept-fhi-mpg.bsky.social found that large language models can outperform specialised AI models in identifying low-energy materials โ without additional task-specific training.
@artificialscience.bsky.socialCan AI write enzymes nature never made? A filtered model output hit 97% activity in a study. Read the evidence:
artificialscience.org/2026/10/prot...
#ProteinDesign #SynBio #AIforScience #APNovo #DeepMind #AlphaProteinNovo
@science-pubmed.bsky.social๐ค Free master prompt: paste into ChatGPT / Claude / Gemini โ your AI becomes a PubMed paper-chat assistant. Grounded strictly in the abstract you paste. Grad students, postdocs, PIs welcome. Unlimited use in your own subscription. https://sciencepubmed.net/en/lab/toolkit/ #AIforScience #PhDchat
@antifocal.bsky.socialErrors grew where training materials were sparse, worst near a refractive index of 1.58. Noise and real measurement variation are not yet built in. People set the parameter ranges.
Jung et al., Nanophotonics 14, 471 (2025)
Values are representative.
@moleculemaker.bsky.socialOur NSF MMLI Director, Prof. Huimin Zhao and his team at Illinois just released the #iCloudBioFoundry platform for autonomous biochemical research. Check out the preprint doi.org/10.64898/202... and the platform (and its associated pubs) here: icloudbiofoundry.illinois.edu
@softwarecampus.bsky.socialDeeBOP: AI for biological sequence design. ๐งฌ Software Campus participant Erik Kubaczka at the #TUDarmstadt works with #Merck on Bayesian optimisation for high-throughput biology. ๐ฌ๐ค Read: softwarecampus.de/en/projekt/d...
@artificialscience.bsky.socialAI 'lab teams' now propose, debate and test hypotheses. Some of it has held up at the bench; some has not. Here is how multi-agent AI labs really work. Read the breakdown. #AIforScience #MultiAgentAI #Biotech #AgenticAI