Well, this could mean that Rosetta, POEM, Predictor and any other prediction projects out there will not be needed to predict the structure of the proteins. But there would still be a need for simulations in the protein docking part, for the design of new drugs.
Not sure if i got this right, i don't have anything in common with molecular biology.
By my understanding, no, the other projects would still be needed.
Protein folding is a mystery. It happens in nannoseconds. Also, in the body, it happens correctly, nearly every time. In the lab the error rate is orders of magnitude higher. The process could not be viewed, and that is very important.
Also, 4nm is a very fine scale, but atoms are much smaller. This MRI would show "large" structures, but not fine detail. Further, charge density would not be visible, and that is very important for predicting chemical reactivity (such as docking).
While seeing the end result is very important (an din this case a major step forward in science), scientists also want to know what happened, how it happened, and if possible why.