Alexandre Hofer

@hofalex.bsky.social

R&D bioorganic chemist in biotech. High affinity to nucleotides and nucleic acids. All views my own.

This report on the CACHE competition is worth a read. It’s about as close as we’ve gotten to some open benchmarking of AI impact to the drug discovery process. I’d suggest a title change to “When Can AI Deliver the Drugs?” though. A few more thoughts👇 1/

When can AI deliver the drug discovery hits?

The CACHE hit-finding competition highlights the potential of AI to identify small molecules that bind to hard-to-drug targets — and the long road ahead for these computational screening approaches. T...

nature.com

"Then they took the extra step, rarely realized in such papers [...], of actually synthesizing hundreds of the hit molecules and experimentally testing them against the real proteins. (Faint cheers can be heard echoing from the lab benches and fume hoods)." 😅

Derek Lowe@dereklowe.bsky.social · 3y ago

Using AlphaFold protein models to screen against has been looking like not-so-great idea. But it might work better than you'd have thought: 🧪 Chemsky www.science.org/content/blog...