I just learnt about this paper, which I gather identifies what might be quite a common situation in drug development: the drug was devised to hit a particular target, and it works - but still does when the target is knocked out. Oh boy, I have questions.
science.org/doi/10.1126/sc…

Would you like a data set of 40,000 reactions for evaluating your reactivity prediction models?

We made it, so you don’t have to. 1/10

Ha brilliant

new paper with Bowen Deng. PES analysis to guide development of future ML force fields + how to data-efficiently fix force underpredictions in many existing ones (incl. MACE-MP) worth reading in full imo but i esp. like Section H! disclaimer: i'm biased :)

New open positions to work with me in the algorithms team at Xanadu. Apply with the link below!

xanadu.applytojob.com/apply/Pmd45HMe…

Job alert! I am looking for a talented PhD student to do research on generative diffusion and its connections to statistical physics and memory.

Feel free to send me a PM if you need more info!

You can apply here:
ru.nl/en/working-at/…

Otherwise, please retweet!

We supported Sam Holton to write a roadmap paper on atomistic simulation.

Techniques like AlphaFold have changed the way we design proteins, how else can computers help us design matter?

Once again Derek saved me the trouble of wading through overstated stats on clinical success rates of “AI-discovered” candidates — using that term in its loosest sense. The breakdown on “AI-discovered” targets, meanwhile, is nothing short of comical.
science.org/content/blog-p…

New preprint: Accelerating QM/MM simulations of electrochemical interfaces through machine learning of electronic charge densities arxiv.org/abs/2405.07370

Tired of applying AI/ML to computational databases with no connection to the real world? Want to use your models to autonomously operate state-of-the-art scientific equipment and advance science for humanity? We are hiring! 🚀

Consider applying for the 'Machine Learning for…

Can we use AlphaFold2 to discover how drugs bind in specific, non-native protein conformations?
Answer: no and yes.

Presenting revised preprint: arxiv.org/abs/2404.07102

Gleevec is a blockbuster anti-cancer drug, having saved an estimated 350,000 lives globally since its…

🔥

(yes, it's equivariant)

Another very impressive general purpose graph NN for molecular simulation. Feels like this field is really accelerating. arxiv.org/abs/2405.04967

New paper alert: introducing AccGeoLDM, an accelerated equivariant latent diffusion model.

We speed up generation of molecular conformations by ~8x with limited loss in quality!

Excited to present at ICLR 2024 GEMBio Workshop 2024 in Vienna. Links to paper & code below. 🧵⬇️

Ha cool it phase transitions to micelles.