Crystallography‎ > ‎Refinement‎ > ‎


From Wikipedia:
"The resolvability in the electron density map of a molecule."
"The highest resolvable peak in the diffraction pattern."

Comments on high-resolution refinement:

- To minimize sequence bias use poly-serine (or poly-alanine) model for MR
- Rigid Body by CNS, beyond 4Å
- Simulated annealing by CNS
- Model Rebuilding, only major problematic places
- SHELXL, conjugate gradient least-squares methods against structure factor
- Model Rebuilding, multiple conformations using 1/1.5σ 1mFo-DFc and 3σ mFo-DFc
- Further refinement, anisotropic displacement parameters for protein and active site (SHELXL)
- Anisotropic displacement parameters of atoms of outer ligand and solvent molecules
- Hydrogen atoms addition according to stereochemistry
- Blocked full-matrix least-squares refinement using all reflections (including those previously used for Rfree

The "best" or recommended refinement program for atomic protein structures (resolution at 1.0 Å or better)?
SHELX is excellent - REFMAC will do a decent job but not such sophisticated hydrogen positioning. 
But what about running ACORN to get a bias free set of phases and using those phases as restraints?