Re: vmd-l: problems with CG protein-lipidbilayer-water system

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Sat Nov 21 2009 - 16:54:04 CST

On Sat, 2009-11-21 at 19:00 +0100, Francesco Pietra wrote:
> Hi:

francesco,

[...]

> If cgsolvate.tcl is added to
>
> /usr/local/lib/vmd/plugins/noarch/tcl/solvate1.3
>
> it is not seen by the command
>
> package require ...

no surprise. this is not how these things work in tcl.

>
> Then I put cgsolvate.tcl in my working directory. Command
>
> source cgsolvate.tcl
>
> reported
>
> conflicting versions provided for package "solvate": 1.3, then 1.2.

again, this is perfect tcl behavior. the script you want to
use is a hack, and an ugly one to boot, as it contains hardcoded
paths to directories on the schulten group NFS server, so it
would have required some editing. the clean solution would
have been to rename all references of solvate to cgsolvate.
the solvate package has since evolved and augmented with additional
features. from the cvs changelog, it looks as if it predates
the optimizations for systems containing millions of atoms
that were done in january and febuary of 2009 and also the
support for arbitrary solvents that was added in january 2007.
so the hack is quite old, actually.

the fact that the current solvate package allows you
to use an arbitrary solvents, should make this hacked
version not really needed anymore.

> My first question is: should I downgrade to VMD 1.8.6? Anton's files
> are not that old, 31 AUG 2009.

remember that those are copies and thus most likely
you see the time they were copied over. it doesn't say
anything about its real age which could well be four
years or more (see above).

> My second question: are the above files for lipids enough for my task?

why don't you try and see?

> My third question: at present is it still advisable to coarse grain
> the components rather that the all-atom final system using the auto cg
> builder?

yeah, it just makes sense to me. it is much easier to merge components
later, and you make it easier for the CG tool to get it right. it is
also easier to identify and debug problems.

axel.

>
> Thanks for answering
>
> francesco pietra

-- 
Dr. Axel Kohlmeyer  akohlmey_at_gmail.com 
Institute for Computational Molecular Science
College of Science and Technology
Temple University, Philadelphia PA, USA.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:22:33 CST