Sam Blau

Sam Blau

@samblau.bsky.social

Research scientist & computational chemist at Berkeley Lab using HT DFT workflows, machine learning, and reaction networks to model complex reactivity.

2.2K followers340 following36 posts

Pinned

Sam Blau
@samblau.bsky.social

The Open Molecules 2025 dataset is out! With >100M gold-standard ωB97M-V/def2-TZVPD calcs of biomolecules, electrolytes, metal complexes, and small molecules, OMol is by far the largest, most diverse, and highest quality molecular DFT dataset for training MLIPs ever made 1/N

Sam Blau reposted

Emory Chan
@emorychannano.bsky.social

For a more digestible summary of our recent @natcomputsci.nature.com article, here's the research brief that @samblau.bsky.social and I wrote... complete with a #BehindThePaper tidbit that thankfully has less drama than a VH1 Behind the Music episode (free link: rdcu.be/eUVIw)

rdcu.beHomeNeed literature management for your research driven company? ReadCube helps your company discover, organize, read, annotate, share and cite.
Nature Computational Science@natcomputsci.nature.com

An accompanying Research Briefing is now available for this paper! www.nature.com/articles/s43... 🔓 rdcu.be/eUVIw

nature.comDeep learning accelerates discovery of complex nanomaterials - Nature Computational ScienceA physics-infused heterogeneous graph neural network has been developed to address challenges in designing complex nanomaterials with spatially varying compositions. This fully differentiable model en...

Sam Blau reposted

Emory Chan
@emorychannano.bsky.social

Our manuscript on deep learning of upconverting nanoparticle heterostructures w/ @samblau.bsky.social is finally out in @natcomputsci.nature.com! See 🧵for links to free access via Readcube and chemrxiv. @berkeleylab.lbl.gov @molecularfoundry.lbl.gov @mitcheme.bsky.social

Nature Computational Science@natcomputsci.nature.com

📢 @emorychannano.bsky.social, @samblau.bsky.social and colleagues develop a DL approach for optimizing the nonlinear optical properties of core-shell upconverting nanoparticles, uncovering photophysical design rules and a roadmap for DL in nanoscience. www.nature.com/articles/s43... 🔓 rdcu.be/eTH72

nature.comGradient-based optimization of complex nanoparticle heterostructures enabled by deep learning on heterogeneous graphs - Nature Computational ScienceGraph neural networks built on physically motivated representations enable gradient-based optimization of complex upconverting nanoparticle heterostructures, revealing photophysical design rules and a...
Sam Blau
@samblau.bsky.social

Come work with @emorychannano.bsky.social and me! #robotics #nanochemistry #machinelearning #UCNPs

Emory Chan@emorychannano.bsky.social

We have a postdoc opening w/ @samblau.bsky.social on the autonomous synthesis of colloidal upconverting nanoparticles. If you're looking for an exciting postdoc combining #robotics, #nanochemistry, #machinelearning, #UCNPs & simulations, see the link below! combinano.lbl.gov/openings

combinano.lbl.govChan Group @ Molecular Foundry - openingsThe Chan group welcomes inquiries for motivated, creative, and independent researchers at all levels (postdoctoral fellow, graduate students, undergraduates, and visitors), even in the absence of post...
Sam Blau
@samblau.bsky.social

OMol25 was calculated with ORCA. I want to acknowledge the work of the ORCA team to improve the quality of the gradient + the robustness of SCF convergence for complicated systems as part of the OMol effort - it was much appreciated and critical to ensuring that we're releasing high quality data!

FACCTs@faccts.de

“Built with the high-performance quantum chemistry program package ORCA (Version 6.0.1), OMol25 contains simulations of large atomic systems that, until now, have been out of reach.” - Meta #ORCAqc #ORCA6 #CompChem #QuantumChem #ML #Meta ai.meta.com/blog/meta-fa... arxiv.org/abs/2505.08762

ai.meta.comSharing new breakthroughs and artifacts supporting molecular property prediction, language processing, and neuroscienceMeta FAIR is sharing new research artifacts that highlight our commitment to advanced machine intelligence (AMI) through focused scientific and academic progress.

Sam Blau reposted

Berkeley Lab
@berkeleylab.lbl.gov

🚨 Just dropped: Open Molecules 2025 — a record-breaking dataset co-led by Berkeley Lab + Meta FAIR. 100M+ DFT snapshots. Built to train #AI for real-world chemistry 🧪. Could reshape discovery in batteries, drug discovery & much more! @cs.lbl.gov ⬇️

newscenter.lbl.govComputational Chemistry Unlocked: A Record-Breaking Dataset to Train AI Models has Launched - Berkeley LabScientists will finally be able to simulate the chemistry that drives our bodies, our environment, and our technologies.
Sam Blau
@samblau.bsky.social

The Open Molecules 2025 dataset is out! With >100M gold-standard ωB97M-V/def2-TZVPD calcs of biomolecules, electrolytes, metal complexes, and small molecules, OMol is by far the largest, most diverse, and highest quality molecular DFT dataset for training MLIPs ever made 1/N

View full thread
Sam Blau
@samblau.bsky.social

We can't wait to see what the community does with OMol! Don't hesitate to reach out with feedback on the data, models, or paper - we aren't going to submit to a journal until the leaderboard goes up, which means we have time to incorporate community feedback (within reason) 10/10

Sam Blau
@samblau.bsky.social

The Open Molecules 2025 dataset is out! With >100M gold-standard ωB97M-V/def2-TZVPD calcs of biomolecules, electrolytes, metal complexes, and small molecules, OMol is by far the largest, most diverse, and highest quality molecular DFT dataset for training MLIPs ever made 1/N

View full thread
Sam Blau
@samblau.bsky.social

Right now, OMol data has energy, forces, partial charges, partial spins, and HOMO/LUMO. But we have far more info that we still need to parse and hope to do a battery of GBW postprocessing. Plus we have 10 petabytes of electron densities. Lots more to come! 7/N

Sam Blau
@samblau.bsky.social

Co-leading OMol with Brandon and Larry was a joy and an honor - as was assembling a world-leading team of scientists from 2 companies, 2 national labs, and 6 universities who were excited to help build an open-source, revolutionary molecular DFT dataset to push science forward 8/N

Sam Blau
@samblau.bsky.social

The Open Molecules 2025 dataset is out! With >100M gold-standard ωB97M-V/def2-TZVPD calcs of biomolecules, electrolytes, metal complexes, and small molecules, OMol is by far the largest, most diverse, and highest quality molecular DFT dataset for training MLIPs ever made 1/N

View full thread
Sam Blau
@samblau.bsky.social

The Open Molecules 2025 dataset is out! With >100M gold-standard ωB97M-V/def2-TZVPD calcs of biomolecules, electrolytes, metal complexes, and small molecules, OMol is by far the largest, most diverse, and highest quality molecular DFT dataset for training MLIPs ever made 1/N

View full thread
Sam Blau
@samblau.bsky.social

OMol was constructed via an unprecedented diversity of methods: MD, ML-MD, RPMD, rattling, Architector, rxn path interpolation, AFIR, optimization, and scaled separation. We also recalculated some previous datasets and did additional sampling/structure generation atop others 3/N

Sam Blau
@samblau.bsky.social

We're also releasing baseline models trained on OMol. To guide future MLIP development, we built novel evaluations on intermolecular interactions, conformers, and charge/spin. We hope to include frequency, ΔG, and TSopt tasks when we put up a public leaderboard in the summer 4/N

Sam Blau
@samblau.bsky.social

The Open Molecules 2025 dataset is out! With >100M gold-standard ωB97M-V/def2-TZVPD calcs of biomolecules, electrolytes, metal complexes, and small molecules, OMol is by far the largest, most diverse, and highest quality molecular DFT dataset for training MLIPs ever made 1/N

Sam Blau
@samblau.bsky.social

OMol covers 83 elements, a wide range of intra and intermolecular interactions, explicit solvation, reactive structures, conformers, charges -10 to 10, 0-10 unpaired electrons, and 2-350 atoms per snapshot. It required >6B CPU hrs, 10x more than any prev MLIP training dataset 2/N

Sam Blau
@samblau.bsky.social

Final day to submit abstracts for ACS Fall 2025! Reminder that @ewcspottesmith.bsky.social , Brett Savoie (Notre Dame), and I are organizing a symposium on "Chemical Reaction Networks, Retrosynthesis, and Reaction Prediction". Will be a mix of invited and contributed talks - please submit! #CompChem

Sam Blau reposted

Gabe Gomes
@gabegomes.bsky.social

the @gpggrp.bsky.social is at the ACS Spring 2025! come check out the works of Daniil Boiko and Rob MacKnight at the "ML + AI in Organic Chemistry" Symposium (Hall B-1, Room 4) today! extreme scaling of experimental chemical reactions via MS and an OS for autonomous comp chem!

Sam Blau
@samblau.bsky.social

Looking forward to speaking at ACS on Sunday at 5:30! Come learn about "Popcornn" - a new method for double-ended transition state optimization atop machine learned interatomic potentials that is substantially better than NEB or GSM.

Sam Blau
@samblau.bsky.social

Fantastic new work from Aditi & co that shows how to leverage the expressivity + accuracy of massive pre-trained MLIPs to distill smaller, much faster models that are still extremely accurate to drive downstream simulations - no need to compromise on speed vs accuracy!

Aditi Krishnapriyan@ask1729.bsky.social

1/ Machine learning force fields are hot right now 🔥: models are getting bigger + being trained on more data. But how do we balance size, speed, and specificity? We introduce a method for doing model distillation on large-scale MLFFs into fast, specialized MLFFs! More details below: #ICLR2025

Sam Blau
@samblau.bsky.social

Applications closing in one week! If you’re interested in a prestigious postdoc at the intersection of AI/ML and nuclear nonproliferation, don’t hesitate to apply - come work with me on fascinating f-block chemistry and computational/ML methods! (Must be a US citizen)

Sam Blau
@samblau.bsky.social

Very proud of this work, going all the way from implementing the kMC in C++ to building datasets w/ high-throughput workflows to designing the novel graph representation to training the custom hetero-GNN w/ on-the-fly augmentation to inverse design of novel nanoparticles with GNN-based optimization!

Evan Spotte-Smith (they/them)@ewcss.info

Preprint! "Inverse Design of Complex Nanoparticle Heterostructures via Deep Learning on Heterogeneous Graphs" is now on ChemRxiv. In this work, we consider the problem of applying deep learning to heterogeneous nanostructures. (1/10) #ChemSky 🧪 #CompChem #physics #optics #nanoscience

doi.orgInverse Design of Complex Nanoparticle Heterostructures via Deep Learning on Heterogeneous GraphsApplications of deep learning (DL) to design nanomaterials are hampered by a lack of suitable data representations and training data. We report efforts to overcome these limitations and leverage DL to...

Sam Blau reposted

Bingqing Cheng
@chengbingqing.bsky.social

Long-range machine learning potentials strike again! 🚀 We benchmarked the Latent Ewald Summation method on diverse systems—molecules, solutions, interfaces. Learning just from energy & forces, it delivers the most accurate potential energy surfaces, physical charges, dipoles, and quadrupoles!

arxiv.orgLearning charges and long-range interactions from energies and forcesAccurate modeling of long-range forces is critical in atomistic simulations, as they play a central role in determining the properties of materials and chemical systems. However, standard machine lear...

Sam Blau reposted

Evan Spotte-Smith (they/them)
@ewcss.info

This week, "RNMC: kinetic Monte Carlo implementations for complex reaction networks" was published in @joss-openjournals.bsky.social. Work on RNMC started back in 2021, when brilliant mathematician Daniel Barter suggested using stochastic methods to study #reactionnetworks. #CompChem #ChemSky 🧪 1/6

joss.theoj.orgRNMC: kinetic Monte Carlo implementations for complex reaction networksZichi et al., (2024). RNMC: kinetic Monte Carlo implementations for complex reaction networks. Journal of Open Source Software, 9(104), 7244, https://doi.org/10.21105/joss.07244

Sam Blau reposted

Evan Spotte-Smith (they/them)
@ewcss.info

I've been slacking on research updates here! A few weeks ago, a preprint for "HEPOM: Using Graph Neural Networks for the accelerated predictions of Hydrolysis Free Energies in different pH conditions" dropped on @chemrxiv.bsky.social. 1/5

chemrxiv.orgHEPOM: Using Graph Neural Networks for the accelerated predictions of Hydrolysis Free Energies in different pH conditions.Hydrolysis is a fundamental family of chemical reactions where water facilitates the cleavage of bonds. The process is ubiquitous in biological and chemical systems, owing to water's remarkable versat...
Sam Blau
@samblau.bsky.social

Example nanoparticle heterostructure optimization, driven by gradients of UV emission with respect to layer thicknesses and dopant concentrations from our hetero-GNN (not accessible from kMC) and sub-second inference (vs days from kMC) #F24MRS

Sam Blau@samblau.bsky.social

Excited to speak at #F24MRS Thurs 1:30 - 1st talk of my career w/o any DFT connection - we design a hetero-GNN for learning core-shell nanoparticle properties, train on first ever large-scale NP kMC dataset, and use autodiff to optimize -> discover far OOD heterostructures with >6x enhanced emission

Sam Blau
@samblau.bsky.social

Excited to speak at #F24MRS Thurs 1:30 - 1st talk of my career w/o any DFT connection - we design a hetero-GNN for learning core-shell nanoparticle properties, train on first ever large-scale NP kMC dataset, and use autodiff to optimize -> discover far OOD heterostructures with >6x enhanced emission

You are up to date.