bond angle an length rmsd is a completely context sensitive data item. It can be zero (for torsion angle refinement with fixed bonds and angles) up to the same rmsd values observed for small molecule targets.
Anything in between is fair game, with the general tendency to become larger with more relaxed restraints (i.e. more data). So it depends on your refinement protocol and restraint settings. Asking for certain hard rmsd values that match all refinements is old school, despite what some references might tell. Ian will probably expand more about the relation to maximum log free likelihood. Best check his Acta D papers on that subject. BR -----Original Message----- From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Simon Kolstoe Sent: Tuesday, December 16, 2008 9:51 AM To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] RMSD's authority Dear ccp4bb, Can anyone point me to a table/paper that gives "acceptable" RMSD's for bond lengths and angles against resolution? We have had two different referees for a paper contradict each other and I am not sure what to use as a good authority for this. Thanks, Simon
