Interpretation of experimental data

While the limited resolution is an experimental fact, that can directly be determined, the impact of inconsistencies in the crystal or the background noise elimination will only show up in the electron density.

The refinement of this density during the manual and automated fitting process for the atomic coordinates introduces new problems that can not be solved without reference to external information (bond length etc.). The force field part of this fitting procedure does not ensure convergence to a consistent geometry due to the simultaneous fit to the density.

Thus, a final structure still shows ambiguities that have to be resolved prior to a modeling study. We have learned that this step is not considered to be a part of the solution of the crystal structure although software is available (What If) that can handle some of the inconsistencies automatically.

Thus, the modeller has to prepare the raw protein structure himself.