Hello,
I don't think this approach will work. You can't make a linear combination
of distance restraints AFTER the simulation has been run with 2 independent
restraints. This is because the force exerted on the system will not be
equal to the linear combination of those 2 restraints. For example,
consider the combination d1-d2. If d1=d2, then there is no restraint if the
appropriate LC is taken. If they're treated separately, the system DOES
have an applied force from the restraints.
Defining a LCOD will give you a 1-D PMF with no need to convert a 2-D to a
1-D. The method you propose doing here won't work, because there's no way
to predict the optimum pathway through a 2-D surface without seeing the
surface. The LCOD approach will map out the 1-D hypersurface (rather than
the 2-D hypersurface with independent restraints) from the reactants to the
products.
Hope this helps,
Jason
On Fri, Aug 6, 2010 at 3:57 AM, Catein Catherine <askamber23.hotmail.com>wrote:
>
> Dear Sir/Madam,
>
>
>
> I am interested to study a reaction coordinate as follows:
>
>
>
> X + C-Y ==> X-C-Y ==> X-C + Y
>
>
>
> It seems to me that I should define the reaction coordinate as the
> distances of XtoC and CtoY.
>
>
>
> I am think of using PMF with umbrella sampling.
>
>
>
> I know how to make two restrainst for umberlla sampling. I also know how to
> do the 2D-PMF from wham. Is it the only way or best way to study this
> reaction coordinate?
>
>
>
> I am thinking if it is a sensible way to convert the results files to
> single column before doing WHAM, i.e.
>
>
>
> After the 2D-PMF calculation, I should get the result file as follows:
>
>
>
> time column 1 column 2
>
> 0 2 5
>
> 1 2 4.9
>
> ......
>
> where the column 1 is the distance between X and C atoms, column 2 is the
> distance between C to Y.
>
>
>
> Borrowing the idea of LCOD to this analysis. I could convert the column 1
> and column 2 to a single combined variable, i.e. column 1 - column 2.
>
>
>
> By doing this convertion, I may be able to calculate the 1D-PMF against the
> combined variable using 1D-PMF.
>
>
>
> Do you think it is a sensible way to get the work done using umbrella
> sampling?
>
>
>
> Best regards,
>
>
>
> Cat
>
>
>
>
>
> Best regards,
>
>
>
> Cat
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 06 2010 - 06:30:05 PDT
