Improving published models

From Proteopedia

Jump to: navigation, search

This page summarizes information about free servers that are capable of improving many published empirical macromolecular models (PDB files).


The MolProbity Server adds hydrogen atoms (which are missing in most crystallographic models) and then looks for atomic clashes and analyses the quality of the geometry. Sidechains of Asn, Gln and His often need to be flipped 180 degrees for optimal fit into their immediate environments. Because the electron densities of oxygen and nitrogen are very similar, crystallographers often leave these sidechains in suboptimal positions in published models. MolProbity recommends which sidechains have strong evidence for being in the wrong positions, and offers to flip them to correct the model.

Molprobity's analyses of atomic clashes and geometry are useful in assessing the quality of a model, overall, and in specific regions of interest, where clashes may be visualized in KiNG.


PDB_REDO does automated re-refinement of published models, and improves about two-thirds of them, according to improvements in Free R. See the publications linked to the site. Sanderson's succinct overview in Nature, 2009 is a good place to start.

See Also

Proteopedia Page Contributors and Editors (what is this?)

Eric Martz, Jaime Prilusky

Personal tools