So excited about our collaboration PenteluteLab building a new paradigm of thinking through the AI-driven design of molecules 💊🧪🩺🏥

Congratulations PenteluteLab! Thanks for the shoutout of Therapeutics Data Commons

Nature review: Scientific discovery in the age of artificial intelligence go.nature.com/45fUazH

📢📢 New release Therapeutics Data Commons - Clinical drug development 💊🏥

New tasks & datasets with 17,000 clinical trials that include validated trial outcomes and other data, including disease and drug information, and trial eligibility criteria

Thanks 符天凡

tdcommons.ai/multi_pred_tas…

9/ The afternoon sessions started with Marinka Zitnik who highlighted the many applications of geometric machine learning for molecular medicine. 💊

From Therapeutics Data Commons, to knowledge graphs, contextual AI models, and more.

Watch the recording here: youtu.be/GQKgCrfcYwo?li…

Thank you Amazon Science for highlighting our efforts in Therapeutics Commons Therapeutics Data Commons to advance safe and effective treatments through AI

amazon.science/research-award…

Excited to share TxGNN! We study in-depth how to make a knowledge graph system for therapeutic use predictions actionable and practical.

medrxiv.org/content/10.110…

🧵1/7

Meet TxGNN, a model that utilizes geometric deep learning and human-centered AI to make zero-shot predictions of therapeutic use across a vast range of 17,080 diseases DBMI at Harvard Med Icahn School of Medicine at Mount Sinai Stanford University Harvard Data Science Initiative Harvard Medical School MIT CSAIL 1/9

txgnn.org

This!

We at Therapeutics Data Commons Therapeutics Data Commons are working towards open models, datasets, and education programs to provide better scientific and medical hypotheses #AI4Science #biomedicalAI

Brand new release of Therapeutics Commons

Excited to announce the release of Therapeutics Data Commons 0.3.9. Find out more about using AI to advance therapies tdcommons.ai/news

📢 Nine high-throughput datasets across several classes of protein targets

📢 New toxicity dataset

In our first community blog of 2023, Kexin Huang and the Therapeutics Data Commons team created a tutorial on how to access more than 50 ML-ready datasets across the entire drug discovery pipeline through a few lines of code.

Read more here: m2d2.peek.link/2Y7d

Thanks for elevating the work of Therapeutic Data Commons

Check out our recent conversations with Harvard Medicine Harvard Medical School News

We discuss Therapeutics Data Commons, its current state as well as long-term goals to expand the use of AI for various therapeutic tasks

hms.harvard.edu/news/can-ai-tr…

📢 📢 Release 0.3.8 Therapeutics Data Commons is available. New datasets and tasks to support structure-based drug design and membrane permeability assays

tdcommons.ai/news

#AI 4Science #therapeutics #AI #drugdiscovery

Excited to share MolE, a molecular foundation model created Recursion by Oscar Méndez Lucio et al. Uses DeBERTa for disentangled atom content/position, pretrained on structures & properties. Achieves SOTA performance on 9 of 22 Therapeutics Data Commons ADMET tasks. arxiv.org/abs/2211.02657

Oloren (Andrew Li) new set of results published on Therapeutics Data Commons!

It's remarkable how far gradient-boosting+stacking diverse predictors together gives performance gains across the board!!!

Best-on-leaderboard performance on:

HIA_Hou (Human Intestinal Absorption)
1/n

Artificial intelligence is poised to transform therapeutic science. Therapeutics Data Commons is an initiative to access and evaluate #AI capability across therapeutic modalities and stages of discovery, establishing a foundation for understanding which AI methods are most suitable and why.