Hi Simon I think I can say without fear of contradiction that this is still a subject under debate, so I'm not at all surprised that the referees can't agree between themselves! I think there are actually two answers to your question: (1) what is the formally correct answer, and (2) what is the answer that should be acceptable to a referee who is open to reasoned argument. In many ways it's easier to give answer (2), since answer (1) is not yet settled in a way that is easily usable.
The first thing I should point out is that you definitely won't find a table/paper that gives acceptable RMSDs for bond lengths and angles against resolution, so don't bother looking for one, sorry! If you want the 'CCP4 community view' on this I think I can speak for the majority and say that a reasonable compromise is to take the median value of RMSD(bonds) (originally suggested by Clemens Vonrhein I think - as least that's the value I use if I have to choose one) to use as a target in refinement at 'typical' resolutions (say between 2.5 and 1.5 Ang). I think it is generally accepted now that the RMS Z-score (Z=deviation/SD)is a more statistically meaningful value to quote (printed by the latest versions of Refmac), so since the range of RMSZ(bonds) values is by definition 0 to 1, the median is 0.5, which corresponds to a RMSD(bonds) of about 0.012 Ang. For bond angles the corresponding values are RMSZ(angles) ~ 0.75, RMSD(angles) ~ 1.5 deg. Having said that, my view (as set forth in Acta Cryst., 2007, D63, 1274-81) is that the correct target for optimisation of weights is not RMSZ/RMSD(bonds) at all but Rfree (as originally suggested by Brunger), or better the log(free likelihood) (suggested by Bricogne). If you do that (properly) the evidence is that do you get resolution-dependent values of RMSD/RMSZ, i.e. lower values at lower resolution, and higher values at higher resolution than the 'typical' resolution range indicated above. For example CNS/CNX and phenix.refine (I believe, though I can't vouch for it as I don't use it), which do implement weight optimisation 'properly' using Rfree as the target, often gives RMSD(bonds) ~ 0.006 (RMSZ ~ 0.25) for resolutions around 2.5 Ang or lower. Also Shel-X seems to give sensible resolution-dependent RMSD/RMSZ(bonds) values (though not for angles, since it doesn't parameterise the uncertainties using the Engh & Huber values). Sorry if this doesn't give the definitive answer you were looking for, but I hope it helps to convince the wayward referee! Cheers -- Ian > -----Original Message----- > From: owner-ccp...@jiscmail.ac.uk > [mailto:owner-ccp...@jiscmail.ac.uk] On Behalf Of Simon Kolstoe > Sent: 16 December 2008 17:51 > To: CCP4BB@JISCMAIL.AC.UK > Subject: 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 > > Disclaimer This communication is confidential and may contain privileged information intended solely for the named addressee(s). It may not be used or disclosed except for the purpose for which it has been sent. If you are not the intended recipient you must not review, use, disclose, copy, distribute or take any action in reliance upon it. If you have received this communication in error, please notify Astex Therapeutics Ltd by emailing i.tic...@astex-therapeutics.com and destroy all copies of the message and any attached documents. Astex Therapeutics Ltd monitors, controls and protects all its messaging traffic in compliance with its corporate email policy. The Company accepts no liability or responsibility for any onward transmission or use of emails and attachments having left the Astex Therapeutics domain. Unless expressly stated, opinions in this message are those of the individual sender and not of Astex Therapeutics Ltd. The recipient should check this email and any attachments for the presence of computer viruses. Astex Therapeutics Ltd accepts no liability for damage caused by any virus transmitted by this email. E-mail is susceptible to data corruption, interception, unauthorized amendment, and tampering, Astex Therapeutics Ltd only send and receive e-mails on the basis that the Company is not liable for any such alteration or any consequences thereof. Astex Therapeutics Ltd., Registered in England at 436 Cambridge Science Park, Cambridge CB4 0QA under number 3751674
