From: Richard Wood (rwoodphd_at_yahoo.com)
Date: Mon Feb 19 2007 - 18:19:46 CST
You wrote, "If we don't do more simulations of mixed bilayers, it is certainly not because of
the capabilities of such or such VMD plugin, it is because we mostly don't know how to simulate them properly." If this is the case, how can we simulate ion channels or protein-lipid systems "properly"?
I mentioned the VMD plug-in because people seem to rely on these programs with their plug-ins to build various things; in the case of NAMD, it's lipid bilayers. It seems to me that one could build an unsturated lipid as a pdb file and then create the proper topologies and
go from there, much like we do with proteins (proteins aren't built using a "neat" plug-in).
Do I get the sense that people think that "different" problems should be handled "differently"? It's kind of like being told a "simulation isn't a simulation, it depends on the conditions...". Maybe this is why our field is dysfunctional as far as the gap between academia and industry?
Richard
Richard L. Wood, Ph. D.
Computational Chemist
Cockeysville, MD 21030
rwoodphd_at_yahoo.com
----- Original Message ----
From: Jerome Henin <jhenin_at_cmm.chem.upenn.edu>
To: Richard Wood <rwoodphd_at_yahoo.com>
Cc: namd-l_at_ks.uiuc.edu
Sent: Monday, February 19, 2007 7:09:49 PM
Subject: Re: namd-l: a philosophical question...
On Monday 19 February 2007 18:16, Richard Wood wrote:
> As far as someone commenting to me, "do you have any philosophical
> questions about MD?", I would say that I do. In fact, I have many. Here
> is one. I have noticed that several (many?) people are working in the
> field of lipid simulations, in the hope of learning something about cell
> membrane behavior. I am wondering if any of these people take into account
> the fact that a cell membrane is not made up of a single type of lipid, but
> generally more than one, and that some of these lipids are unsaturated (to
> assure that the membrane won't freeze). I would like to know how this is
> handled. The membrane building tool of VMD won't allow the building of a
> mixed lipid bilayer. How does one get around this?
By carefully navigating around tricky questions and mined areas. Lipid
bilayers are just very, very difficult systems, be it for experimental or
computational studies (and the two are completely linked). What we are in now
is a kind of middle-age of lipid simulations, if I may paraphrase KS. If we
don't do more simulations of mixed bilayers, it is certainly not because of
the capabilities of such or such VMD plugin, it is because we mostly don't
know how to simulate them properly. Actually, even our models of most
single-component bilayers are pretty unreliable.
That's why most people prefer to ignore lipid-specific problems and just treat
them as "whatever greasy thing my protein is embedded in". I suppose it is
okay most of the time.
Jerome
____________________________________________________________________________________
8:00? 8:25? 8:40? Find a flick in no time
with the Yahoo! Search movie showtime shortcut.
http://tools.search.yahoo.com/shortcuts/#news
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:24 CST