[buster-discuss] Error for rms gradient in Buster or Gelly after sanity check

Nicolas Foos Nicolas.Foos at afmb.univ-mrs.fr
Wed Mar 3 18:47:53 CET 2010


Le 03/03/10 16:55, Thomas Womack a écrit :
> Dear buster users,
>
> The message
>
> ' *** ERROR rms gradient is NaN - not a number'
>
> is almost always caused by having two atoms in exactly the same place, often by inserting a water into a peak in coot twice.
>
> The next release will regard this as a sanity error, with message
>
>          1 dist=   0.00 idealD=   2.74 sigma=   0.20 abs(Delta/sigma)=   13.70
>            Between atoms A|121:O (WAT) A|150:O (WAT)
>            distance is exactly zero. So atoms are exactly coincident.
>            Advice: check this carefully
>
> Probably the easiest way to figure out which atoms are on top of each other is to use the ccp4 remark-500 tool: run 'r500 HKLIN my.pdb', which will produce a file 'rem500' containing a message of the form
>
> REMARK 500 THE FOLLOWING ATOMS ARE IN CLOSE CONTACT.
> REMARK 500
> REMARK 500  ATM1  RES C  SSEQI   ATM2  RES C  SSEQI           DISTANCE
> REMARK 500   O    HOH A   121  -  O    HOH A   150              0.00
>
> Hope this helps,
>
> Tom Womack (Global Phasing)
> _______________________________________________
> buster-discuss mailing list
> buster-discuss at globalphasing.com
> https://www.globalphasing.com/mailman/listinfo/buster-discuss
>
>    
Thank you for your fast answer.

I used r500 in order to Check distance between atoms. I find some Water 
which are exactly in the same place. I remove one on two of these water.
I have tried to run Buster.
It give me the same error message.
In order to fix it, i have check all atoms put in light by R500 for the 
shorter distance.
I remove some other Water and one Residue in C-ter.
  And after this Buster run very well.

I am suprised because, i have in the past used buster for refine model 
wich have some problem of clash. But Buster give error message about the 
sanity Check. When you have this message you can correct the problem in 
your model or force the run with an option in command line. But in my 
case the only one message is about the rms gradient.


Thank you.

-- 
Nicolas Foos, PhD Student
Molecular Transport&  Signalling
AFMB UMR 6098 CNRS/UI/UII Case 932
163 Avenue de Luminy
13288 Marseille cedex 9 (France)
Tel : +33 4 91 82 55 60
Fax : +33 4 91 26 67 20
e-mail : Nicolas.Foos at afmb.univ-mrs.fr
http://www.afmb.univ-mrs.fr/


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



More information about the buster-discuss mailing list