Protein structure prediction is at the single-structure frontier
TJ's paper is out in Nature Methods!

Protein structure prediction is so good, it's hit the "single structure frontier". Next step: perturbations and distributions. Read more in the paper, available from Nature Methods.