Patrick Walters

cheminformatics, machine learning, drug discovery, opinions

[Patrick Walters @wpwalters]

An interesting piece about Axiom Bio's work combining cell painting and machine learning to predict toxicity

[Alex Kesin @alexkesin]

For a decade, techbio optimized the drug discovery "paper mill": more molecules, faster. But the hard part - knowing which ones will work safely in humans - hardly moved at all.

Late last year, I spent time with Axiom Bio, a company built around the idea that the real bottleneck is predicting clinical activity in molecules, *not* merely discovering them.

[Patrick Walters @wpwalters]

@andrewwhite01 Nice, you should ask kosmos to submit to the OpenADMET/ExpansionRx blind challenge.

[Patrick Walters @wpwalters]

@ddale8 @jaketapper Zz

[Patrick Walters @wpwalters]

New Practical Cheminformatics post, "Three Papers Demonstrating That Co-folding Still Has a Ways to Go”.
https://t.co/uTtjdcRuHC

[Patrick Walters @wpwalters]

@davidlmobley My favorite paper this year!

[Patrick Walters @wpwalters]

New Practical Cheminformatics Post, "Useful RDKit Utils - A Mötley Collection of Helpful Routines"
https://t.co/MaP7vjkQ39

[Patrick Walters @wpwalters]

The latest Practical Cheminformatics post, “The Trouble With Tautomers,” emerged from a discussion about the impact of tautomers on machine learning model predictions.

https://t.co/nylhStlG42

[pen(Taka) @iwatobipen]

Integration Chembl REST API and Claude with MCP #cheminformatics #mcp #ai https://t.co/TbPQ2j0Z81

[Patrick Walters @wpwalters]

The latest addition (#33) to the Practical Cheminformatics Tutorials series explores Bayesian optimization of reaction conditions.
https://t.co/NEii9bKEaN

[Patrick Walters @wpwalters]

In a new Practical Cheminformatics post titled "Even More Thoughts on ML Method Comparisons," I share several plots that I find valuable for comparing machine learning methods.
https://t.co/4JaGk3n8bD

[Patrick Walters @wpwalters]

@JCIM_JCTC @sgj30 @songyojung @Cambridge_Uni @JCIM_JCTC It's distressing that authors continue to publish papers using the highly flawed MoleculeNet dataset. Journals like @JCIM_JCTC must provide guidelines on what constitutes an acceptable validation. https://t.co/x8eEdOoLpU

[Patrick Walters @wpwalters]

@Azhar__Shaikh Given my chemistry background, I have my own biases, but I believe that domain knowledge is essential for anyone involved in ML for drug discovery.

[Patrick Walters @wpwalters]

Machine Learning in Drug Discovery Resources page updated for 2025. https://t.co/2qcCHtmcA2

[Patrick Walters @wpwalters]

ChEMBL 35 is out. Happy Holidays!
https://t.co/FCJOpVV3Gs

[Patrick Walters @wpwalters]

Excellent new paper (with code) by my former colleagues Steven Kearnes and Patrick Riley describing a procedure for associating confidence levels with regression model predictions in drug discovery. https://t.co/0iQbNcEOwc

[Patrick Walters @wpwalters]

I'm not happy with the way this all went down. I have tremendous respect for Gabriele and the rest of the DiffDock authors. Their work has broken new ground and helped advance machine learning in drug discovery. If I had to do it over again, I'd do things differently.

[Patrick Walters @wpwalters]

Many of you have seen Gabriele Corso's response to our recent preprint. If you haven't, it's here. https://t.co/ZVympXbzGz

[Patrick Walters @wpwalters]

@prof_ajay_jain's response is here. https://t.co/X3OTDRwy7u

[Patrick Walters @wpwalters]

@DdelAlamo @prof_ajay_jain Time splits with the pdb are a profoundly bad idea. Groups often deposit a novel structure and then deposit the structures of analogs a couple of years later. Efforts like Plinder are addressing this.

[Patrick Walters @wpwalters]

I'm thrilled to announce a new preprint describing collaborative work with @prof_ajay_jain and Ann Cleves Jain, "Deep-Learning Based Docking Methods: Fair Comparisons to Conventional Docking Workflows".

https://t.co/8WzcGYlQAQ