-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hello Xiaohu-- > I got a bunch of structures from Xplor-NIH calculation, out of NMR distance > restraints in a protein sample. > Is there any standard criterion to select "good" or "acceptable" structures? > Or put it in another word, what should be considered real high-resolution > protein structures? you shouldn't have any (or very small number of) covalent (BOND, ANGL, IMPR) violations. For the VDW violation metric reported, we seem to see about one violation per residue. For NMR terms, the answer to your question is more difficult, as it depends on how you set things up with your restraints. > > This is a very general and broad question. Apparently I should start with > energy terms, e.g., VDW, bond, angle, etc. But then the question boils down > to the boundary limit for each term, say, how much violation is acceptable > in angle under consideration, 1 degree, 5 degree, or 50 degree? Or they are > interacting with each other, reflecting each other in each term, if the > energy is high in bond, most probably the angle energy is also high, things > like that. For covalent terms, you can use the default threshold values as defined in python/xplorPotTools.py hope this helps-- Charles -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/> iD8DBQFGYCNxPK2zrJwS/lYRAtXkAJ94eJg86xVgR0n25dEQ8s5tuUXsjgCfTe+Y RwFqJMhPxWQEV2KD88IMScA= =JkOn -----END PGP SIGNATURE-----
