From: Yogesh Aher (aher.yogesh_at_gmail.com)
Date: Wed Jul 15 2009 - 09:49:33 CDT
Dear Chris,
Thank you very much for your reply and nice suggestions.
1. I tried to increase cutoff values till 30 A, but the problem didn't show
any improvement.
2. Yeah, Thanks for this.. I checked with dcd and found 2 residues movement.
Then again with little googling, found something about checking some long
bond. After checking .psf and .pdb files, came to know that my (rather my
colleague's) protein is not continuous. i.e. It has 2 chains, first ends at
residue number 630 and next starts at 702. This empty region is the loop
region and we didn't give it much importance as it's not involved in
binding.
The same error is there, but now I am wondering how I can proceed with these
2 chains. As if I join them with bond, the whole geometry will be disturbed.
Any idea about this?
Thanking you for your kind consideration.
Sincerely,
Yogesh
On Mon, Jul 13, 2009 at 4:54 PM, Chris Harrison <charris5_at_gmail.com> wrote:
> Yogesh,
>
> Two general, but not foolproof, approaches to resolving this problem
> include:
>
> 1) increasing your cutoff and restarting (not "redo-ing") the simulation.
> This is actually more of a "post-injury bandage" than cure for the
> underlying issue that caused the problem and assumes whatever caused the
> problem is over and will not continue or recur.
>
> 2) Decrease the output frequency of your dcd output to 10 or even 1 and
> redo the simulation. When it crashes the output may dump some atom indices
> that are exceeding velocities limits. Visualize the trajectory with VMD and
> locate these atom's by creating a new rep (selection: serial "atom indices
> from NAMD" ) using VDW. If no atom indices are dumped by NAMD, you will
> still need to visualize the system with VMD to determine if your system is
> exploding, imploding, has atoms or residues rapidly deviated from expected
> behavior, etc.
>
>
> Chris
>
>
> --
> Chris Harrison, Ph.D.
> Theoretical and Computational Biophysics Group
> NIH Resource for Macromolecular Modeling and Bioinformatics
> Beckman Institute for Advanced Science and Technology
> University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801
>
> char_at_ks.uiuc.edu Voice: 217-244-1733
> http://www.ks.uiuc.edu/~char <http://www.ks.uiuc.edu/%7Echar>
> Fax: 217-244-6078
>
>
>
>
> On Mon, Jul 13, 2009 at 5:57 AM, Yogesh Aher <aher.yogesh_at_gmail.com>wrote:
>
>> Dear NAMD- users!
>>
>> I was trying to run simple namd run for my homology-modelled protein which
>> is quite big in size.
>> It's PBC dimensions are like this (after solvating with option -t 5):
>>
>> # Periodic Boundary Conditions
>> cellBasisVector1 105. 0. 0.
>> cellBasisVector2 0. 121. 0.
>> cellBasisVector3 0. 0. 125.
>> cellOrigin 21. 73. 27.
>> PME GRID DIMENSIONS 108 128 128
>>
>> But I get the following error:
>> Warning: Bad global exclusion count, possible error!
>> Warning: Increasing cutoff during minimization may avoid this.
>>
>> I increased cutoff values along with the pairlist-distance, but no impact
>> of them on avoiding this error. (Is there any rule of thumb to calculate
>> cutoff values depending on the PBC and PME dimensions?)
>>
>> I read other posts with the same subject, but couldn't get the relevant
>> solution, so thought to raise this question again.
>>
>> Looking forward for your valuable suggestions..
>>
>> Thanking you,
>> Y.
>>
>
>
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:02 CST