2 ms·
Refreshingly good. Bakers "early" (not really early, but earlier than AlphaFold) work (having humans with no background solve folds) really laid the groundwork
by throwawaymaths 2y ago
Refreshingly good. Bakers "early" (not really early, but earlier than AlphaFold) work (having humans with no background solve folds) really laid the groundwork to proving that heuristic methods were likely to outperform physical forcefield and ab initio/DFT methods for structure prediction. And AI structure prediction if nothing else is heuristic protein folding.