3 ms·
You can't get 100% accuracy on something for which you don't or can't know the ground truth.
by gfodor 6y ago
You can't get 100% accuracy on something for which you don't or can't know the ground truth.
- dekhn 6y agoThe protein folding problem is predicated on the idea that there is a ground truth (a single static set of atomic coordinates with positional variances). If your point is that even experimental methods can't truly reach 100% (due either to underlying motion in the protein, or can't determine the structure), that's more or less what Moult is saying (they more or less arbitrarily define ~1A resoution and GDT of 90 as the "threshold at which the problem is solved").