I would fix the issue before minimization and not rely on MD to appear
not to have a problem. If you start an MD in a bad conformation, it
might do even worse things to resolve the problem -- change ring
structure, flip chirality, etc. Simulated annealing can also cause
unnatural geometric conversions. Of course, you might get lucky in
some simulations, but I strongly recommend using the impose command to
fix the issue at the start.
On Tue, Feb 26, 2013 at 3:37 PM, Jio M <jiomm.yahoo.com> wrote:
> (Sorry mistake Jason I think I replied to your email rather replying to mailing list so writing here again)
>
>>> You could try some simulated annealing or something like that...
>
> Yes thanks this is pretty nice suggestion. I think I should locally anneal the rings only at high temperatures by fixing the other residues. This might help I guess.
>
> regards,
>
>
> ________________________________
> From: Jason Swails <jason.swails.gmail.com>
> To: Jio M <jiomm.yahoo.com>; AMBER Mailing List <amber.ambermd.org>
> Sent: Tuesday, February 26, 2013 10:25 PM
> Subject: Re: [AMBER] adding rings using sequence command
>
> The barrier to getting "unstuck" is too high, I guess. You could try some
> simulated annealing or something like that...
>
> But fixing the initial structure is still the better approach, IMO.
>
> Good luck,
> Jason
>
> On Tue, Feb 26, 2013 at 3:19 PM, Jio M <jiomm.yahoo.com> wrote:
>
>> Dear Lachele and Jason;
>>
>> Thanks for explanation and suggesting impose command (Jason also
>> suggested).
>> Just for knowledge sake I was curious why MD run couldn't resolve this
>> issue (if fragments remain stuck inside rings even after minimization they
>> keep stuck during MD run!)
>>
>> thanks,
>>
>>
>>
>>
>> ________________________________
>> From: Lachele Foley <lf.list.gmail.com>
>> To: Jio M <jiomm.yahoo.com>; AMBER Mailing List <amber.ambermd.org>
>> Sent: Tuesday, February 26, 2013 10:02 PM
>> Subject: Re: [AMBER] adding rings using sequence command
>>
>> Minimization will only find the lowest local energy minimum using only
>> very local knowledge of forces. So, it has no good way of resolving
>> an issue like that. In general, it will take less energy to expand
>> all bonds in a ring slightly, and leave a branch intruding, than to
>> greatly stretch one bond in the ring to allow the branch to escape.
>>
>> We see this frequently, and are developing tools for automating
>> resolution. But, that might take a while. In the meantime, you have
>> to fix it yourself.
>>
>> When you have those situations, it is best to resolve them before
>> minimization. Of course, you might use a brief minimization to
>> determine if you have a problem, then go back and fix the
>> pre-minimized structure. Use the impose command in tleap to alter
>> angles. Altering angles far from the intrusion site will require the
>> least change in any given angle. The AmberTools12 manual contains a
>> good description of the command, and there are additional examples in
>> the section on building oligosaccharides.
>>
>>
>> On Tue, Feb 26, 2013 at 2:49 PM, Jio M <jiomm.yahoo.com> wrote:
>> > Dear Jason,
>> >
>> >
>> > Thanks for the suggestions. But please clarify this also (and correct
>> me):
>> > By mistake (which I realized after MD run) I had used the same structure
>> in which many rings were crossed by polymer chain fragments and run
>> minimization and later MD for 10 ns. I think such criss cross from rings
>> should have been removed by doing this, but it did not help.
>> >
>> > On the contrary if we run some simulation we never see branches crossing
>> the rings; I mean I never saw the "bonds" of fragments (which are just
>> bonds by distance) crossing the rings. So I think its vdw radii which
>> prevents this. I was wondering why after running minimization and later MD
>> did not resolve this issue and the polymer fragments remained stuck inside
>> the rings?
>> >
>> > regards,
>> >
>> > Jiomm
>> >
>> >
>> > ________________________________
>> > From: Jason Swails <jason.swails.gmail.com>
>> > To: Jio M <jiomm.yahoo.com>; AMBER Mailing List <amber.ambermd.org>
>> > Sent: Tuesday, February 26, 2013 5:53 PM
>> > Subject: Re: [AMBER] adding rings using sequence command
>> >
>> > On Tue, Feb 26, 2013 at 9:55 AM, Jio M <jiomm.yahoo.com> wrote:
>> >
>> >> Hi All,
>> >>
>> >> I am working on some polymeric system and want to introduce naphthalene
>> >> like ring system in polymer at intervals. I am using sequence command to
>> >> add ring residues to the polymer. As polymer is a complex system so
>> when I
>> >> add a ring residues at intervals of polymer using sequence command few
>> >> rings are added in such a manner that some polymer fragments criss cross
>> >> the ring. Its not a good idea to somehow relax the rings using xleap
>> one by
>> >> one as there are many ring systems.
>> >>
>> >
>> > Maybe try changing your residue template for your polymer building
>> block...
>> > The tleap sequence functionality is very simplistic (and therefore
>> limited)
>> > -- you can look into some of the commands to adjust structure (like
>> > 'impose', I believe) -- the GLYCAM folks have done some very clever
>> things
>> > getting tleap to build carbohydrates, but it is not easy (at least it's
>> not
>> > easy for me).
>> >
>> > You may be better off trying to build your structure elsewhere (or put a
>> > good deal of effort into learning how to 'really' use tleap).
>> >
>> > HTH,
>> > Jason
>> >
>> > --
>> > Jason M. Swails
>> > Quantum Theory Project,
>> > University of Florida
>> > Ph.D. Candidate
>> > 352-392-4032
>> > _______________________________________________
>> > AMBER mailing list
>> > AMBER.ambermd.org
>> > http://lists.ambermd.org/mailman/listinfo/amber
>> > _______________________________________________
>> > AMBER mailing list
>> > AMBER.ambermd.org
>> > http://lists.ambermd.org/mailman/listinfo/amber
>>
>>
>>
>> --
>> :-) Lachele
>> Lachele Foley
>> CCRC/UGA
>> Athens, GA USA
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>
>
>
> --
> Jason M. Swails
> Quantum Theory Project,
> University of Florida
> Ph.D. Candidate
> 352-392-4032
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
:-) Lachele
Lachele Foley
CCRC/UGA
Athens, GA USA
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Feb 26 2013 - 13:00:03 PST
