Steven -
There are a couple of ways to calculate the free energy difference, which
will tell you about the population of the conformers as well as the
equilibrium ratios. We published a paper where we concentrated on the chi
torsions of RNA nucleosides. Even though you want to look at the sugar
puckerings of nucleosides, the methodology we followed might be useful for
your work. Look at the MD part of the paper.
"Re-parameterization of RNA CHI Torsion Parameters for the AMBER Force
Field and Comparison to NMR Spectra for Cytidine and Uridine," I.
Yildirim, H.A. Stern, S.D. Kennedy, J.D. Tubbs, and D.H. Turner, J. Chem.
Theory and Comput. 6, 1520-1531 (2010).
Here are my suggestions:
1. You can simple solvate the nucleotides and run a long MD simulation in
constant pressure (NPT). With the current computer technology, I will
think that running a microsecond MD on a nucleoside in explicitly solvated
system will not take too long. If the simulation time is long enough, then
you can do cluster analysis on the trajectory. The significant parts in a
nucleoside will be chi and sugar puckering, so a 2D clustering profile
will give you a nice picture of where the conformations are
occupied/concentrated in the MD. I will expect that you will end up seeing
4 regions. You can then calculate the % of structures in each region,
which you can then calculate the relative free energy differences. See the
above paper.
By the way, I will not expect you to see a big difference when you run the
MD in constant volume (NVT) dynamics.
2. You can do umbrella sampling MD simulations in 2D where chi and sugar
puckering are the reaction coordinates for each nucleoside. You might even
stick with 1D calculations with only sugar puckering being the reaction
coordinate. You will need to sample the conformational space enough,
however, if you are going to do 1D calculations. As I wrote above, in a
single nucleoside, chi and sugar puckering are the two main torsions that
will have a big impact on the conformation of the nucleoside.
This approach is, however, much more complicated than the regular MD
simulations. I have done this calculations for DNA nucleosides (not
published though). At the end, you will end up with 2D free energy
surfaces. These type of free energy surfaces will give you an idea on how
the conformational transitions might happen (the pathways) as well as the
free energy differences of different conformational minimas. I think
Carlos Simmerling, Benoit Roux, Alex MacKerell have used this approach to
analyze different nucleic acid systems. Carlos might say something on
this approach. They published a paper where they used umbrella sampling MD
to analyze stacking/unstacking of RNA bases but I cannot find it in their
website.
Hope this helps. Good luck.
Ilyas Yildirim, Ph.D.
-----------------------------------------------------------
= Department of Chemistry - 2145 Sheridan Road =
= Northwestern University - Evanston, IL 60208 =
= Ryan Hall #4035 (Nano Building) - Ph.: (847)467-4986 =
=
http://www.pas.rochester.edu/~yildirim/ =
-----------------------------------------------------------
On Fri, 29 Jun 2012, Steven M. Graham wrote:
> It pains me to say this, but I am not 100% certain of your terminology. Free energy difference is, of course, delta G, and would thus include the entropy of the north/south change. 'Absolute energy difference' is...just delta H? What I want, at the end of the day, is the equilibrium ratio of the solvated north/south conformers to compare to in vacuo values.
>
> Steven M. Graham. Ph.D.
> Associate Professor
> Department of Chemistry
> St. John's University
> 8000 Utopia Parkway
> Queens, NY 11439
> 718-990-5217
> grahams.stjohns.edu
>
> -----Original Message-----
> From: Ilyas Yildirim [mailto:i-yildirim.northwestern.edu]
> Sent: Thursday, June 28, 2012 6:26 PM
> To: AMBER Mailing List
> Subject: Re: [AMBER] Re [amber] north/sout energy difference
>
> Do you want to calculate the absolute energy difference of these two
> conformations or free energy difference?
>
> Ilyas Yildirim, Ph.D.
> -----------------------------------------------------------
> = Department of Chemistry - 2145 Sheridan Road =
> = Northwestern University - Evanston, IL 60208 =
> = Ryan Hall #4035 (Nano Building) - Ph.: (847)467-4986 =
> = http://www.pas.rochester.edu/~yildirim/ =
> -----------------------------------------------------------
>
>
> On Thu, 28 Jun 2012, Steven M. Graham wrote:
>
>> I'm re-posting this as I got only one response (thanks Jiri; see below). Not a criticism - I know if I don't respond right away the post tends to get lost in my inbox.
>>
>> I'm going on the assumption that there are no 'silly questions'...
>>
>> How would one calculate the energy difference between a north conformation and a south conformation of a nucleoside in solution? Clearly, one could build a periodic box of H2O for each structure and optimize, but wouldn't the water energy dominate? One could delete the water molecules, but now that's a vacuum calculation. Let's assume using a distance-dependent dielectric is not going to give realistic values. Perhaps a solvent model, like GB or PB?
>>
>> Jiri: Read your chi(OL) paper - I was already familiar with it. The problem in I want the N/S energy difference at the minima, which you've set to zero to compare the chi torsion profiles and chi dihedral terms. But thanks anyway.
>>
>>
>> Steven M. Graham. Ph.D.
>> Associate Professor
>> Department of Chemistry
>> St. John's University
>> 8000 Utopia Parkway
>> Queens, NY 11439
>> 718-990-5217
>> grahams.stjohns.edu
>>
>> _______________________________________________
>> 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 Fri Jun 29 2012 - 11:00:03 PDT
