Bio researcher questions protein folding modeling: claims Baker Lab has no in vivo translation

iskander · x · 2026-09-15

A bio researcher shares a series of hot takes: Baker Lab's protein design work has never translated in vivo (arguing VLPs were the real win), and modeling protein folding may not be that useful at all. Also: TCGA misses post-translational modifications, and phosphoproteomics data may be unreliable given minutes-long delays between resection and snap freezing. Unreviewed personal opinions, but a notable counterpoint to AI protein-structure hype.

Original post →

More from AGI Musings

AGI Musings channel →