[gmx-users] trjconv and g_clustsize

[Mark Abraham]

On 12/26/2011 7:53 PM, mohammad agha wrote:
> Dear Prof.
>
> I have several questions about trjconv, please help me.
>
> In my simulation is created 2 micelles.
> 1- When I do 3 steps of micelle clustering as follows:
>
>   a- |trjconv -f md.xtc -o a_cluster.gro-e 600000 -pbc cluster|
> | b- ||grompp -f md.mdp -c a_cluster.gro -o a_cluster.tpr|
> | c- ||trjconv -f md.xtc -o a_cluster.xtc -s a_cluster.tpr -pbc nojump|
>
> ||
> I view 2 micelles according to result of my simulation in VMD but in 
> ngmx isn't view 2 micelles but just is view 1 micelle and reminder of 
> monomers are collected in several groups in different places.

So you don't have a single cluster of atoms... so as trjconv -h notes, 
using -pbc cluster isn't very useful when you don't have a cluster of 
atoms. All my advice of two days ago on this point still applies.

> Is it right if my criterion be view in VMD and is my micelle 
> clustering correct?

It's "correct" if it looks the way you want it to.

> ||
>
> 2- If I select surfactants for clustering group and system for output 
> group in step (a) and I select system for output group in step (c), 
> command g_clustsize doesn't work correct, consequently, I did under 
> step and made a .xtc file only for surfactants:
>
>     i- trjconv -f md.trr -o md-surfactant.xtc -n index.ndx
>
> Then I did steps a, b and c for md-surfactant.xtc and in all of steps 
> I selected surfactants for clustering and output group. Next, I did 
> g_clustsize and it answered me correct.
> May I know that my way is correct, please?
>
> 3- By upper way I have 2 micelles that I got index.ndx file for 
> cluster with maximum size by g_clustsize (maxcluster.ndx).
> May I know how can I use from this for next calculation if I want 
> consider only maximum cluster with all of system (water and ion) in 
> calculations (for example rdf, gyration and etc)?

Again I think you are not using the word cluster in the sense the 
GROMACS tools are. Each micelle could be a cluster of atoms, or the pair 
of micelles could be a cluster of micelles. How to use the tools will 
differ according to what you are trying to calculate on what kind of 
cluster. Anyway, the atom numbers do not change with the configuration 
of the atoms, so the same index file group identifies the surfactant 
atoms in each frame. To identify the *time* of the configuration with 
the largest diameter, something like g_mindist -max might serve. However 
you may need to resolve your PBC issue first.

Mark

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20111226/42118fd8/attachment.html>