[gmx-users] g_densmap options and use

[Mark Abraham]

On 13/11/2010 4:58 AM, sa wrote:
>
>     On 13/11/2010 4:07 AM, sa wrote:
>     > Dear All,
>     >
>     > I would like to compute the average 2D density distribution of the
>     > water around 6 peptides aggregated in the cluster within the
>     > simulation box with gromacs, for that I think that g_densmap is the
>     > the good tool (correct ?).
>     >
>     > However it is not very clear for how to use g_densmap. Below the
>     > command I used with g_densmap (ver GMX 4.5.3)
>     >
>     > g_densmap_mpi -f ./TRAJ/XTC/6_Pep_Urea_Pref_All.xtc -s
>     > ./TRAJ/TPR/em.tpr  -bin 0.02 -amax 30 -rmax 30 -b 0 -e 80000 -o
>     > 6_Peptide_53A6_densmap.xpm -od 6_Peptide_53A6_densmap.dat
>     >
>     > When I use the above command, g_densmap asks me to choose two groups
>     > to define the axis and an analysis group:
>     >
>     > Reading file ./TRAJ/TPR/em.tpr, VERSION 4.5.1 (single precision)
>     > Reading file ./TRAJ/TPR/em.tpr, VERSION 4.5.1 (single precision)
>     > Select two groups to define the axis and an analysis group
>     > Group     0 (         System) has 86359 elements
>     > Group     1 (        Protein) has   546 elements
>     > Group     2 (      Protein-H) has   396 elements
>     > Group     3 (        C-alpha) has    48 elements
>     > Group     4 (       Backbone) has   144 elements
>     > Group     5 (      MainChain) has   192 elements
>     > Group     6 (   MainChain+Cb) has   234 elements
>     > Group     7 (    MainChain+H) has   246 elements
>     > Group     8 (      SideChain) has   300 elements
>     > Group     9 (    SideChain-H) has   204 elements
>     > Group    10 (    Prot-Masses) has   546 elements
>     > Group    11 (    non-Protein) has 85813 elements
>     > Group    12 (          Other) has 17320 elements
>     > Group    13 (            URE) has 17320 elements
>     > Group    14 (             CL) has     6 elements
>     > Group    15 (          Water) has 68487 elements
>     > Group    16 (            SOL) has 68487 elements
>     > Group    17 (      non-Water) has 17872 elements
>     > Group    18 (            Ion) has     6 elements
>     > Group    19 (            URE) has 17320 elements
>     > Group    20 (             CL) has     6 elements
>     > Group    21 ( Water_and_ions) has 68493 elements
>     > Select a group: 1
>     > Selected 1: 'Protein'
>     > Select a group: 16
>     > Selected 16: 'SOL'
>     > Select a group: 16
>     > Selected 16: 'SOL'
>     >
>     > I chose protein and SOL, the program ask me to choose a third group
>     > (?) What to choose ?
>
>     Doesn't g_densmap -h explain the three groups?
>
>
>  Yes I have read the help of the tool but it is not clear to me why i 
> have to choose three groups since i want to compute the water density 
> map around my peptides (-> two groups)

You need to define an axis by the line between the centers of mass of 
two groups. Almost surely the COM of your protein cluster and the COM of 
your water is a random line. I don't know what there is about your 
system that makes you want a 2D density distribution, but if there is, 
you need to come up with a sensible axis around which to compute. Only 
then does the distribution of third group get computed.

>
>     > I choose SOL again, the program computes something but i can not
>     > inspect the results are what i want since no xpm is generated by
>     > g_densmap (is this a bug ?)
>
>     Probably badly-formed input is silently breaking something somewhere.
>
>
> I don't understand your response since with the above command and -o 
> argument show that an xpm with the name "6_Peptide_53A6_densmap.xpm" 
> should appear

Agreed, but lets get your input formed sensibly before deducing that 
there's a bug or anything.

I had a quick look at the code, and Peter Kasson designed it so that if 
you use -od then you won't get the -o output. You could get both from 
two invocations of g_densmap, which doesn't sound like a good design, 
and certainly contradicts the documentation. For the moment, I'll fix 
the documentation.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101113/7e219dd7/attachment.html>