Tweets

2022-10-20T14:05:39+00:00 | 🔗

@OlorenAI #AI #Chemoinformatics #Cheminformatics https://t.co/jIYx7yskVO

2022-10-20T14:04:27+00:00 | 🔗

RT @OlorenChemEng: "A Unified System for Molecular Property Predictions: Oloren ChemEngine and its Applications" Whitepaper for Oloren Chem…

2022-10-19T23:05:35+00:00 | 🔗

RT @OlorenChemEng: A demonstration edition of Oloren ChemEngine is now pip installable (virtually) without dependencies!!! pip install olo…

2022-10-18T13:46:07+00:00 | 🔗

RT @moyix: Proposal: PhD-level course in "how to get stuff to build and run". The final exam assigns you a random GitHub repo for a paper f…

2022-10-16T21:33:02+00:00 | 🔗

@lab_COSMO @marceldotsci @MicheleCeriotti Wow, the world was just hiding this from me, super cool!!

2022-10-16T00:03:17+00:00 | 🔗

Does anyone know why in molecule GNNs we encode atoms with their atomic number instead of their position on the Periodic Table? Seems like the latter would build in some useful priors.

2022-10-14T20:39:44+00:00 | 🔗

likely of differing chemical spaces. That agreement was measured by measuring the correlation of the model's predictions with the data on the other dataset. These models (based on published results) outperform other models on these datasets.

2022-10-14T20:39:44+00:00 | 🔗

There are two there because each one is built on a different dataset, either PhysProp (https://t.co/SIbRPgDmoS) or an AZ dataset (https://t.co/CR9QxN40dG). The models based on the two datasets only moderately agree with each other, which could be a by-product of bias or more...

2022-10-14T20:39:43+00:00 | 🔗

Went on an adventure making LogP calculators... Making them public on the links below (need to make free account) LogP calculator (PhysProp): https://t.co/mxDWvd9EUE LogP calculator (AZ_dataset):https://t.co/mxDWvd9EUE https://t.co/KzqovlHQVP

2022-10-11T00:16:41+00:00 | 🔗

Some structure-based methods coming to OCE soon (but probably not that soon)!