Back to Synthetic Challenge #119 (reddit.com/r/chemistry/co…). We start with Product A. (1/3)

We are excited that the Molecular Transformer publication (pubs.acs.org/doi/full/10.10…), the backbone of IBM RXN for Chemistry, has just reached 20k downloads and is still featured in the ACS Central Science most-read lists (bit.ly/3aslLnf, last 30 days/12 months)!

It’s time! Our #IOPPposter Twitter Conference is now live! Thank you to all who registered - we look forward to seeing your posters here today! Don’t forget to use the official #IOPPposter hashtag and your subject area hashtag! ow.ly/ZWV250ETTsk #STM #TwitterConference

Automated ligand design and synthesis in our Machine Learning: Science and Technology work. #PaccMannRL generates SARS-CoV-2 proteins' ligands and RXN for Chemistry plans the synthesis. Paper: ibm.biz/paccmann-mlst-… Code: ibm.biz/paccmann-mlst-… IBM Research IOP Publishing #IOPPposter #biomedical

Accelerating IP generation in materials science using #NLP for patent classification. See how adaptive fine-tuning on #transformers improves over SOTA. IBM Research  Queen's University Belfast 🎓 IOP Publishing #IOPPposter #materials

Check out my #IOPPposter poster on machine learning models for disconnection-aware synthesis planning. Let human experts decide where to cut bonds. (Thread) IBM Research RXN for Chemistry #rxn4chemistry IOP Publishing #Atomic #Molecular #WomenInSTEM #BlackInSTEM

Thanks to auto-encoders, it is possible to machine-learn the relationship between #materials' compositions with their structural properties and optimise them #IOPPposter IOP Publishing IBM Research

Active site is enough! In drug-target binding affinity prediction,DL models rely on SMILES & full proteins. But for kinases, active sites yield better performance. Consistently across models, datasets & data splits. IOP Publishing #molecular #biomedical #IOPPposter IBM Research

Curious to know how to get more diversity out of your retrosynthesis predictions? Check out the chemistry puppeteer! IBM Research #rxn4chemistry IOP Publishing #IOPPposter #Mathematics #Computation #Molecular #Chemistry

Combining the power of generative models and optimization to discover a new set of promising catalyst candidates. IBM Research IOP Publishing #IOPPposter #Molecular

Leveraging machine learning for more sustainable catalysis in chemical industries by supercharging IBM RXN @ForRXN with biocatalysed reaction planning.  IBM Research IOP Publishing #IOPPposter #Molecular

Tadaa! Check out my #IOPPposter and learn how chemical language models predict reaction yields - even in the low data regime (98 train samples)! #atomic #molecular #transformers #DeepLearning #compchem IOP Publishing IBM Research Machine Learning: Science and Technology paper: iopscience.iop.org/article/10.108…

Here is an example on how protein sequences could be used to perform more accurate synthesis planning IBM Research IOP Publishing #IOPPposter #Materials #Environment #GreenCatRXN #RXN4Chemistry

The #IOPPposter event is a fantastic opportunity to show you what we did so far in 2021 and where we are going. I am incredibly privileged of working with spectacular scientists at IBM Research. This thread is about them and our team contributions. Details in the thread. 1/n

Now out on ChemRxiv! doi.org/10.33774/chemr… We developed a method to quantify the cost of synthetic routes. This can be applied to manual (regular) and automated synthesis, as well as a combined approach that we call “subway map” synthesis. Read the intro to see why 🚶🚴🚍🚋🚇

I did not know about the BigScience Research Workshop, but they are actually doing some extremely cool stuff in NLP in an open and ethical manner! Very impressive!

#ASAP by Alan Aspuru-Guzik & team! They developed a way to quantify the cost of manual and automated synthetic routes and use it to find the best route or the cheapest molecule without sacrificing important physical properties: go.acs.org/8u

Happy to have co-organized the Applied Machine Learning Days track “AI & the Molecular World” w/Philippe Schwaller (he/him) @ilvailvail Bruno Correia Théophile Gaudin! A shout-out to the great speakers Charlotte Bunne Michael Bronstein Yasemin Bozkurt V. Jannis Born Alain Vaucher @ZHarteveld @MarkusOrsi Loïc Roch A Mollaysa, AF Bitbol👇