Colby T. Ford, Ph.D.·Sep 21Zero-Shot Classification of Protein Binders with JevJev, a zero-shot classification model by TypeSafe AI, popped up last week and it has taken over my LinkedIn feed. People have already been…
Late Starter AI·Aug 26“A Decade of Drug Discovery, Done in Three Weeks”… The Upheaval of AI Protein Design PlatformsSomewhere in London, a small AI lab backed by Alphabet is trying to compress a decade of pharmaceutical research into a matter of weeks. I…
InTowards AIbyDecoding AI·Aug 25Claude’s Protein Design Hit Rate Was 26.8%. One Target Returned 0 for 90.Anthropic’s August 18, 2026 campaign produced 354 binders from 1,320 designs. But 26.8% is a portfolio average across sixteen targets, not…
Saritarawat·Aug 22Claude Just Designed New Proteins – And a Lab Confirmed They WorkAn AI Agent Takes On Drug Discovery’s Slowest Step
Dipayan Sarkar·Aug 20Pre-train Your First Autoregressive Mixture-of-Expert Based Protein Language ModelProtein language models are still dense while the rest of the field went sparse. Here is how to build and pre-train a sparse one.
Engin Yapici·Jun 16PPIFlow: Picomolar Binder Design and the Limits of the BLI DataChangping Laboratory’s flow-matching pipeline reports picomolar affinities for six of seven mini-binder targets, all measured by BLI.
InDataDrivenInvestorbyKhang Pham·May 3Generative AI Protein Design After the Nobel PrizeIntroductionA response icon1A response icon1
Engin Yapici·Apr 21I Stress-Tested the AWS-MSK Nanobody Paper. These Are the Gaps.Technical analysis of an agent-guided de novo nanobody design campaign
Engin Yapici·Feb 24Germinal: Open-Source Nanobody Design With Epitope Evidence and Some Unresolved GapsStanford and Arc Institute’s pipeline designs nanobodies against specified epitopes, testing only 43–101 candidates per target.
Justinpezulu·Feb 15Designing Proteins Backwards: Reimagining the Arp2/3 Complex in an African Crop Using AI.What if we could start with a protein’s structure and design the sequence that creates it?
