Ella Rajaonson

PhD student in the Matter and ChemCognition labs @UofT, @EPFL alumni 👩🏽‍🔬 • AI enthusiast @VectorInst 👩🏽‍💻

[PolymathicAI @PolymathicAI]

Are chemists done & is NMR solved? Not so fast! 🤔
Outperforming software is not the goal, to be truly useful, AI needs to operate like a human chemist under real lab conditions.

✨Check our #ICML2026 AI4Science Spotlight!
🔗OpenReview: https://t.co/sgh9djOqUG
Thread 🧵

(1/7)

[Marta Skreta @martoskreto]

Feynman-Kac Correctors back in action 😎 -- this time for steering protein generation towards (1) multimodal rewards that require both sequence and structure information and (2) specificity (predicted binding to an on-target but not an off-target).

This is only one small mirror on the DISCO 🪩 ball -- our new model to co-design generate protein sequences and structures around arbitrary biomolecules (no inverse folding needed!).

Check out the thread below to see everything you can generate with our model, including tips n tricks for inference (turns out these make a huge difference), and, most importantly, super cool wet lab results!!! 🧪

Super proud of the whole team (special shoutout to @jarridrb, @ChengHaoLiu1, @k_neklyudov, @AlexanderTong7), this was a huge effort. 🫶

[Phil Fradkin @phil_fradkin]

The news is out! We're starting Blank Bio to build a computational toolkit assisted with RNA foundation models. If you want to see my flip between being eerily still and overly animated check out the video below!

The core hypothesis is that RNA is the most customizable molecule in the world. It's fast to produce, inherently programmable, and highly adaptable to diverse use cases. It's an open secret amongst folks who've worked at Deep Genomics, and that's why almost everyone is still working on RNA applications.

We're building out the next generation of tooling enabled by our foundation models and partnering with companies to accelerate their work. If you have a use-case big or small we'd love to chat, please reach out.

The foundation models we're building stem from intuition acquired through years of research from our PhDs. We learned that effective biological models require deep understanding of data, specific architectures and loss functions that reflect the molecular processes shaping the data. Genomics metrics like conservation and constraint inform model development.

One of the best parts about this is that I get to work with such talanted, kind, ambitious people. @hsu_jonny has spent the last five years defining what great ML+Bio products look like and consistently pushes us to adapt to customer needs with mission oriented clarity. @ianshi3 is the most thoughtful and creative scientist I've ever worked with, bringing a great sense of design to everything he does.

We're on a mission to build something great and useful for RNA scientists around the world.

blank dot bio

[Marta Skreta @martoskreto]

AI4Mat is back for NeurIPS! time to crystallize those ideas and make a solid-state submission by august 22, 2025 💪

new this year: opt-in your work for our Research Learning from Speaker Feedback program -- a new structured discussion format where spotlight presenters receive curated, in-depth feedback from invited discussants 🧐

check out our website for more:
https://t.co/Pdw9OFnuc0

brought to you by a 🔥organizing committee: @MiretSantiago @anoopnm007 @SteMartiniani @kmjablonka @ADuvalinho Emily Jin Rocío Mercado Oropeza

[Yakavets Ilya @IYakavets]

I'm so excited to finally share our paper, just published in Science Advances! Kudos to @kheiri_sina and all co-authors!
#Microfluidics, #PatientDerivedOrganoids, and #BayesianOptimization #ML https://t.co/n9OOY55nmc

[Stefan Schmid (Neurips 2024) @Stefan_P_Schmid]

I will be at #NeurIPS next week, presenting work on Bayesian Optimization over curried functions at the AI4Mat workshop. We use it to identify general reaction conditions https://t.co/e5VSumKOnY

If you are around and interested in talking #BayesOpt or #AI4Science, let’s meet

[Chris J. Maddison @cjmaddison]

Is there enough public data in AI4bio to repeat the success of LLMs in NLP? Not clear. What is clear is that we aren’t even using all the available tokens. Our new preprint: often-ignored text tokens can improve protein LMs.
🤗https://t.co/kxCp88l9JF
📝https://t.co/2OoRTwN4bu

[Ella Rajaonson @ERajaonson]

@irina_espejo @KyleCranmer @iris_hep @GoogleDeepMind @FulbrightPrgrm @NYUDataScience Congratulations Irina ! So proud 💜 Hope you are having fun in Zurich !

[Abdulrahman Dossary @an_dossary]

🚀🚀 New Publication Alert! https://t.co/Bn62sTosrm

Discover how AI and ML are revolutionizing computational chemistry by enhancing scalability, speed, and accuracy in predicting chemical properties and solving the Schrödinger equation. This review highlights key advancements, challenges, and a roadmap for integrating AI with traditional methods. Perfect for researchers in materials science and AI! 🔬🤖🧪 #AI #Chemistry #MachineLearning #compchem #Research @AdvSciNews @wileyinresearch

Great job team! @A_Aspuru_Guzik @spgarcica @leongshixuan @JorgeACamGA @ERajaonson @mistergtom @wavyparticle Luca Theide and Davide Avagliano! #matterlab

[Felix Strieth-Kalthoff @felix_s_k]

More than 2.5 years after joining @A_Aspuru_Guzik's lab at @UofT, our decentralized self-driving lab has eventually found its home in @ScienceMagazine. It's been the biggest, most intense (and craziest) research project I've ever been involved ... [🧵]

https://t.co/xImpD1DO2L

[Stefan Schmid (Neurips 2024) @Stefan_P_Schmid]

Excited to share that our Review on Self-Driving Laboratories (SDLs) is now published! 🧪 If you are interested in AI for Science and/or lab automation, check it out! 💻🦾

Congratulations to the entire team!!

[E. Kumacheva Group @EKumachevaGroup]

Halloween group presentation at @EKumachevaGroup @IYakavets gave a talk about integration of #AI with #microfluidics for #cancer research in collaboration with @A_Aspuru_Guzik and @PMResearch_UHN #Halloween #Science

[Daniel Flam-Shepherd @FlamShepherd]

We built a framework for fragment-based 3D molecular design using reinforcement learning. Scales to more than 100 atoms.

Preprint: https://t.co/VgXt8Aozsu
w/ @alex_zhigalin, @A_Aspuru_Guzik

Watch the agent design a molecule by placing fragments from Imatinib:

[Ella Rajaonson @ERajaonson]

Super excited to have such a lab at EPFL, I wish I was still there to feel the excitement 🥺 ! I hope this will motivate more EPFL chemistry students to explore the amazing world of AI for Chemistry, can’t wait to see how things go !

[LIAC at EPFL @SchwallerGroup]

Today's our first day. We are excited to start this new adventure at @EPFL_CHEM_Tweet/@EPFL_en.

What's your best advice for a #NewPI (-> @pschwllr)?
#compchem #chemtwitter

[MIT Technology Review @techreview]

.@A_Aspuru_Guzik is combining artificial intelligence and robotics in an attempt to accelerate materials discovery—with the ultimate aim of solving really thorny problems like climate change. https://t.co/0iTNGsRRHi

[Ella Rajaonson @ERajaonson]

Definitely making big leaps in my life these days… Happy to see myself growing, lucky to be part of an amazing working environment, grateful to have people supporting me ! 🤸🏽‍♀️

[Alan Aspuru-Guzik @A_Aspuru_Guzik]

Today is a great day for #ai for #chemicals and #materials discovery.

Rick Mullin @cenmag wrote a great article about #selfdriving labs where many in the field are quoted including yours truly. https://t.co/TU28aBfAz8

This sets the stage for an announcement from @UofT 1/8

[Camila Lenis @ladivaloper]

Don't spend 6 minutes doing something by hand when you can spend 6 hours failing to automate it.

[Maerkl Lab @MaerklLab]

@EPFL_iGEM 2020 iGEM project wiki describing their project this (crazy) year. Clean water is a human right, but water quality is declining! The iGEMers developed sentinel yeast strains and a low-cost hardware platform to enable water testing! https://t.co/gjXcldnNs9 @EPFL_en

[Ella Rajaonson @ERajaonson]

Sometimes you’ve gotta hack your room to provide some content for the #iGEM2020 Wiki ! Here is a lamp turned into an overhead tripod ! @SebastianMaerkl @EPFL_iGEM