you need a box that can hold the unfolded state, not one that just
fits the folded state.
you should also ensure that the results do not depend on the initial
structure used.
hoestly, this is a very challenging problem, and difficult to do properly.
On Mon, Jun 11, 2012 at 2:52 AM, Chinh Su Tran To
<chinh.sutranto.gmail.com> wrote:
> Dear Dr. Simmerling,
>
> I used the box size of 12 angstroms (doubted if it was large enough,
> honestly). As visualized, the protein moved out to the box-edge though even
> iwrap=1 was used.
>
> I wanted to detect the intermediate unfolded states, and the drug-targeting
> ability at the active site of these intermediate unfolded states.
> So what i need is the unfolded structures at different time points of the
> unfolding pathway. And what I've got are unfolded representative structures
> with significant "overall" RMSD differences. However, they dont satisfy
> what I expected because the "little change" in the active site are actually
> NOT able to represent the whole unfolding process, or in other words, I
> havent got the "totally" unfolded conformations.
>
> Regarding to this scenario, could you please give me one advice that
> whether I should jump in to change some parameter (i.e. increase the
> temperature, or as you mentioned add the denaturants) to get the better
> unfolded structures OR I need to start it over? I actually tried out
> increasing temp to 600K at some time points (i.e. after the first 150ns),
> but the total energy scenario changes.
>
> Thank you.
>
> Regards,
> Chinh
>
> On Fri, Jun 8, 2012 at 5:57 PM, Carlos Simmerling <
> carlos.simmerling.gmail.com> wrote:
>
>> it's not really an answer to your question, but have you considered
>> the size box you need to hold the unfolded state? how large is your
>> box? it may be that the protein is interacting strongly with its
>> periodic images, stabilizing specific structures.
>>
>> regarding the "how long does it take"- proteins vary widely in this.
>> is there experimental data for unfolding rate? that might give some
>> clue, though of course it can't be done at 500K. there are other ways,
>> such as adding denaturant, but keep in mind that unfolded states and
>> pathways depend on the process sued to unfold it, and so what you do
>> depends on what you want to learn. In my opinion, an unfolded protein
>> at 500K doesn't really tell me much that has biological relevance.
>>
>> On Thu, Jun 7, 2012 at 11:17 PM, Chinh Su Tran To
>> <chinh.sutranto.gmail.com> wrote:
>> > Dear Amber users,
>> >
>> > I am doing unfolding for a 270-residue protein using Amber 9. By heating
>> it
>> > at 500K (dt=1.5fs), I observe its unfolded states (mostly by
>> visualization
>> > using VMD).
>> > It's been running for 276 ns, but its active site is still somehow
>> > "maintained" (not much changed --> I calculated distances among C-alpha
>> > atoms of these active residues, the values are not very much different
>> > among the unfolded states, e.g. the 6ns-state and the 240ns-state).
>> > The overall RMSD significantly differ though.
>> >
>> > This may be a "silly" question 'cause it is not a technical one, but I am
>> > interested in it.
>> > - I didnt put any restraint for the whole system during the unfolding,
>> but
>> > using iwrap=1 to keep it in the solvent box. Does this iwrap parameter
>> > somewhat give any restraint per se?
>> > - Im a newbie in unfolding such a big protein, how long normally does it
>> > take to totally get unfolded protein? One of my professors folded a
>> protein
>> > in ms or second scale, how about unfolding?
>> > - aside from heating, any other ways to unfold the protein? Is 500K an
>> > appropriate value?
>> >
>> > Thank you for any help. Any recommended papers are also very much
>> > appreciated.
>> >
>> > Regards,
>> > Chinh
>> > _______________________________________________
>> > 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
>>
> _______________________________________________
> 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
Received on Thu Jun 14 2012 - 09:00:03 PDT
