Re: simulation of membrane-protein systems

From: Andrew D. Fant (fant_at_pobox.com)
Date: Mon Dec 18 2006 - 13:42:31 CST

L. Michel Espinoza-Fonseca wrote:
> Thank you very much for your suggestions and for the script. After a
> slight modification, it worked perfectly and got a system with no
> water around the membrane. I also ran a couple of simulations and got
> a much better system, with no membrane "crushing" at the edges and of
> course no water around the hydrophobic region of the membrane. I will
> definitely acknowledge you in the resulting paper! You helped me
> to save a lot of time :)

I'd like to second the thank you. As a relatively new reader and relative
novice, this thread has been really helpful in sorting out the messier
practicalities of these simulations.

I'd like to throw in one set of related questions for my own edification. I'm
looking at some predicted receptor structures and want to do some sort of
conformational search to get a sense for the range of motion. My first instinct
was simulated annealing, but that struck me as a really bad idea with a
relatively delicate bilayer involved, So some sort of Monte Carlo process would
be my next shot. Is there a good way to generate any sort of set of random
conformers in NAMD or VMD, or do I need a separate tool for that? Also, what
sort of time scales is reasonable for this kind of exploration in conformational
space. My friends in CFD would tell me "until the residuals decline
sufficiently", but I'm far enough afield from my previous experience that any
ballpark figures or rules of thumb would be highly appreciated.

Thanks, and Happy Holidays to everyone,

Andy

-- 
Andrew Fant    | And when the night is cloudy    | This space to let
Molecular Geek | There is still a light          |----------------------
fant_at_pobox.com | That shines on me               | Disclaimer:  I don't
Boston, MA     | Shine until tomorrow, Let it be | even speak for myself

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:58 CST