[gmx-users] rot_diff

[Berk Hess]

Hi,

Note that for a large molecule in a relatively small box the rotational diffusion
is suppressed significantly due to hydrodynamic forces which decay as 1/r.

Berk.


Date: Mon, 7 Jul 2008 18:35:57 -0400
From: rams.crux at gmail.com
To: gmx-users at gromacs.org; X.Periole at rug.nl
Subject: Re: [gmx-users] rot_diff
CC: 

Dear XAvier,

Thanks for your reply and for the explanation. I am not an NMR guy so I would like to know little bit of more about the way we can calculate the rotational diffusion. The way I understood is the following and let me know if I am wrong.


After obtaining the rotaional correlation function using Gromacs tools (g_rotacf), I need to calculate the correlation time I suppose. 
The obtained correlation time is related with the local diffusion constant (d) by the relation

d = 1/l(l+1) t
t is the correlation time obtained above and l = 1 or 2 depends upon the order of the legendre polynomial we will use in the g_rotacf and the experimental results with which we are comparing.   

then by solving the following relation

d=n'Qn (n is a unit vector lies along the vector connecting the two spins), we can obtain "Q" which inturn is in relation with D (the diffusion tensor). 

Thats the overall idea I have but I am sure I need to worry alot of finer other details while I start putting my hands into it. If the overall idea is alright I could put the things in a more detailed way.


Ram.

On Sat, Jul 5, 2008 at 12:51 PM, Xavier Periole <X.Periole at rug.nl> wrote:

On Sat, 5 Jul 2008 10:40:21 -0400

 "rams rams" <rams.crux at gmail.com> wrote:


Dear users,



Is it possible to evaluate the rotational diffusion of proteins using

gromacs tools ??


No directly. However you can use g_rotacf to generate the autocorrelation

function of vectors (option -d). By defining vectors representing your

molecule/protein you can access the rotational correlation time of your

representative vector. You can imagine different way to get a statistically

significant value. One would be to define many vectors between backbone

atoms and average your results. Another would be to again define many

vectors but this time between the center of mass of the protein and each

Ca atoms and average ...



You can also hack the g_rms code to extract the rotational matrix during

the overlay of your protein to a reference structure and apply it to

a unit vector from whose trajectory you can again use g_rotacf to get

the autocorrelation function of that vector ...



An important point in the comparison of your result to experimental

values is the way the rotational correlation time is extracted

experimentally. They select different mode of relaxation (1 or 2) and thus

you have to use the corresponding Legendre polynomial when calculating the

autocorrelation function. From NMR relaxation l=2.



XAvier.



-----------------------------------------------------

XAvier Periole - PhD



Molecular Dynamics Group

- NMR and Computation -

University of Groningen

The Netherlands

-----------------------------------------------------

_______________________________________________

gmx-users mailing list    gmx-users at gromacs.org

http://www.gromacs.org/mailman/listinfo/gmx-users

Please search the archive at http://www.gromacs.org/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.

Can't post? Read http://www.gromacs.org/mailing_lists/users.php




_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20080708/254e57f2/attachment.html>