Varinia Bernales

Check out our latest preprint: Efficient Generation of Metal-Organic Frameworks Using a Hybrid Diffusion-Transformer Architecture By Seunghee Han, Yeonghun Kang, Taeun Bae,
@variniabernales.bsky.social, @aspuru.bsky.social, and Jihan Kim.

📜: arxiv.org/abs/2511.03122
[1/6]

Read the great work of @nicolasn.bsky.social who came to #Toronto and gave us an encyclopedic talk about his work on corrols. Looking forward to our collaboration @variniabernales.bsky.social @thematterlab.bsky.social @uoftcompsci.bsky.social

The Matter Blotter | Alan Aspuru-Guzik | Substack accelerating science, chemistry and materials, quantum computing, artificial intelligence. Click to read The Matter Blotter, by Alan Aspuru-Guzik, a Substack publication with hundreds of subscribers.

We’re relaunching The Matter Blotter — our Substack newsletter where the Matter Lab members will be writing about their research in greater detail, going beyond what’s possible on social media (and way past the Bluesky character limit!).

Subscribe here: aspuru.substack.com

Alán Aspuru-Guzik and Sheila McIlraith awarded AI2050 fellowships to advance research on beneficial AI The U of T computer science faculty members have been named AI2050 Senior Fellows by Schmidt Sciences.

Our director, @aspuru.bsky.social, has been named alongside
Sheila McIlraith as AI2050 Senior Fellows by Schmidt Sciences.

The prestigious fellowship honours researchers advancing responsible innovation and the development of AI that benefits humanity.

Learn more: uoft.me/bYA

Stereochemistry-aware string-based molecular generation Abstract. This study investigates the impact of incorporating stereochemical information, a crucial aspect of computational drug discovery and materials de

New paper out in PNAS Nexus!

Molecules aren’t flat — so why should our models be?🧬

Our latest work "Stereochemistry-aware string-based molecular generation,” explores how considering stereochemistry changes the way machine learning models design molecules.
Link: doi.org/10.1093/pnas...
[1/4]

What do transition state search, geometry relaxation, zero point energy corrections, and extrema classification have in common? They all require Hessians!

The problem is, accurate Hessians are really expensive, even with MLIPs. We say, just shoot 'em from the HIP! 🤠
[1/4]

Excited to share our new article: “Quantum algorithm for vibronic dynamics: case study on singlet fission solar cell design”, now published in Quantum Science and Technology!

📜 Paper: doi.org/10.1088/2058...
📝 Blog post overview: pennylane.ai/blog/2025/02...

[1/8]

We're extremely happy to share that our paper, ⭐ELECTRA: A Cartesian Network for 3D Charge Density Prediction with Floating Orbitals⭐, was accepted to NeurIPS 2025 as a spotlight paper.

Link to preprint: arxiv.org/abs/2503.08305
[1/6]

🎉Excited to share our new perspective: “Connecting the concepts of quantum state tomography and molecular representations for machine learning” — now on ChemRxiv!

Link: doi.org/10.26434/che...

[1/5]

As a multicultural group, we are proud to celebrate #HispanicHeritageMonth by highlighting our two directors, @aspuru.bsky.social and @variniabernales.bsky.social, along with former Matter Lab member Joel Yuen-Zhou, who have been recognized as Latin American Trailblazers by C&EN!
[1/3]

C&EN's 2024 Trailblazers

To celebrate #HispanicHeritageMonth, we are highlighting stories of chemists with Latin American roots. Read more about C&EN’s Trailblazers: cen.acs.org/careers/dive... #HHM #chemsky 🧪

These scientists left the US in Trump’s first term: their tips on taking the leap Nature talks to researchers about why they moved and how they relocated successfully.

Our lab director, @aspuru.bsky.social, talking about the University of Toronto: “I am the happiest guy here.”

This Nature article shares his story, the experiences of other scientists who made international faculty moves, and advice for those considering such a leap.

www.nature.com/articles/d41...

No worries, and thanks for highlighting our perspective!

And I just found this related Perspective by @aspuru.bsky.social and Varinia Bernales in Polyhedron where they argue that in #CompChem "we are at a crossroads where artificial intelligence will further democratize our discipline"

doi.org/10.1016/j.po...

Code for Quetzal is now available: github.com/aspuru-guzik...

Try it out now in Colab: colab.research.google.com/github/aspur...

Congratulations to our lab director, Alán Aspuru-Guzik (@aspuru.bsky.social), on being named a Fellow of the Royal Society of Canada (RSC)! 🎊

Read all about it here: web.cs.toronto.edu/news-events/...
[1/7]

Ever wondered how to get organic chemists to synthesize the molecules you designed with your newest generative AI model?

Please check out our latest article: “The elephant in the lab: Synthesizability in generative small-molecule design”.

🐘 Link to preprint: chemrxiv.org/engage/chemr...
[1/4]

With all the recent interest in physical machine learning, @realmantilla.bsky.social and I wondered what would be a mechanical machine learning system? We devised a simple springs and sticks model that leads to very interesting thermodynamics learning limits.
#chemsky
#physicssky #machinelearning

Excited to share our most recent work: “Learning with springs and sticks” — now on arXiv!

Paper link: arxiv.org/abs/2508.19015
Project page (with animations & code): bestquark.github.io/springs-and-...

Authors: @realmantilla.bsky.social, @aspuru.bsky.social
[1/4]

Have you ever wanted to use computer vision to speed up your materials research, but weren’t sure where to start?

High-throughput synthesis can generate huge libraries of new materials. However, characterization often lags behind, becoming the real bottleneck in discovery.
[1/6]

A hands-on tutorial on computer vision for chemistry coming out of @thematterlab.bsky.social in collaboration with @omarfarha.bsky.social ‘s lab led by Matter lab’s director of research @variniabernales.bsky.social

@accelerationc.bsky.social
#chemsky
#compchemsky #computervision #selfdrivinglabs

We're excited to present our latest article in Nature Machine Intelligence - Boosting the predictive power of protein representations with a corpus of text annotations.

Link: www.nature.com/articles/s42...
[1/4]

What if, instead of trying to predict properties of every molecule, we focus on simply ranking them? After all, when running Bayesian optimization (BO) for drug/materials discovery, what matters is picking the best candidates first.

Paper: doi.org/10.1063/5.02...
Code: github.com/gkwt/rbo
[1/5]

Quantum computing meets AI to fight cancer The system generated thousands of targets for KRas a difficult-to-drug protein involved in cancers

Quantum computing meets AI to fight cancer

The system generated thousands of targets for KRas a difficult-to-drug protein involved in cancers. cen.acs.org/acs-news/acs... #chemsky 🧪

Congratulations to @aspuru.bsky.social and all the members of @thematterlab.bsky.social for this exceptional piece 🥳👏🧪🤖 (We need a new emoji for #SDLs)

#chemsky #compchemsky

We are delighted to announce that our perspective article, “Steering towards safe self-driving laboratories (SDLs),” has been accepted for publication in Nature Reviews Chemistry.

Link: www.nature.com/articles/s41...

❤️ Hear some of our latest research live at ACS Fall 2025! The Matter Lab is heading to Washington, DC next week to share about chemistry digitization and self-driving labs. #ACSFall2025

Speakers:
Shi Xuan Leong, on 18 Aug, 4:15 pm
Sean Park, on 20 Aug, 4:45 pm

Thank you for the shoutout! Happy to see a strong interest in safety among other leading SDL researchers.

Headed to Accelerate 2025? You cannot miss the presentations from The Matter Lab. We've got your back--here's your ultimate cheat sheet so you don't miss a thing.

#Accelerate2025 #AIinScience
[1/2]