[gmx-users] Generation of the Distance Restraints

[Mark Abraham]

On 28/03/2012 10:48 PM, James Starlight wrote:
> Mark,
>
> This sounds like I use very small forces but expect reasonable effect. 
> But I've applied different forces with step-by-step increasing of 
> force constants ( from very softest comparable with the thermal motion 
> ( 0.1 kj mol nm-2) to relatively hight (10).  As the consequence I've 
> observed effect of application of that harmonic constraints wich I've 
> defined in the r0 and r1 range but in some case ( where forses were 
> were hight I've seen perturbation of my structures) and when 
> constraints were low ( in accordance to my literature) I've not seen 
> desired effect like the selection of the constraints was wrong ( but 
> actually all restraints were applied on the correct possitions). This 
> was seen by measurement of the distances between atom pairs wich were 
> contrained. Eg If I define this distances in the 0.1<Rij<0.4 ( r0=0,1 
> r1=0,4 for this instance) range the real distance between i and j 
> atoms in the simulated structure was lower or higher of the defined 
> range.

What's reasonable depends on the objective. If you want to keep 
something very close to where it *already* is, and it's reasonably happy 
there already, then you don't generally need much in the way of 
restraints. If your starting configurations are different from what you 
wish to achieve, then you're going to have to metaphorically speak loud 
enough to be heard over the thermal commotion, and then loud enough to 
cross the relevant barriers. That can mean big restraint forces and tiny 
integration steps and lots of tweaking and praying. Or finding a new 
starting configuration that's more relevant for the objective.

For comparison, pdb2gmx generations position restraints with 1000 kJ mol 
nm-2 force constants for helping people not perturb their structures 
during equilibration. Your distance deviations are much larger than will 
usually occur with PR, so you don't want to go that large.

>
> By the way I've found that besides such harmonic restraining also I 
> can apply more rigid holo restraints from specified value. Could you 
> tell me where I could find information of the application of such 
> restraints in the topology of my protein? As I understood this could 
> be done by means of editing of the bond enty in topology but what 
> exactly specified type should I applied on the restried atoms?

Rigid constraints are not useful for you, because your initital 
conditions are a long way from your target conditions. Any kind of 
harmonic potential has the same or worse issues than you already have.

Mark

>
>
> Thank for help again,
>
> James
>
> 22 ????? 2012 ?. 17:32 ???????????? Mark Abraham 
> <Mark.Abraham at anu.edu.au <mailto:Mark.Abraham at anu.edu.au>> ???????:
>
>     If there's a car at the bottom of one valley in the Alps, and you
>     think it should be two valleys over, and you pull it with a
>     trained cat... it's not going to move much. How big an animal you
>     need depends on the geography. There need not even be a reasonable
>     route for you to take, if the target valley is effectively on Mars.
>
>     Mark
>
>     -- 
>     gmx-users mailing list gmx-users at gromacs.org
>     <mailto:gmx-users at gromacs.org>
>     http://lists.gromacs.org/mailman/listinfo/gmx-users
>     Please search the archive at
>     http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>     Please don't post (un)subscribe requests to the list. Use the www
>     interface or send it to gmx-users-request at gromacs.org
>     <mailto:gmx-users-request at gromacs.org>.
>     Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120328/37704d24/attachment.html>