[gmx-users] how to indicate solvent flexibility?

[David van der Spoel]

On 2013-01-10 10:45, Albert wrote:
> Hello Justin and Leandro:
>
>    thanks a lot for kind advices. I am trying to us the g_msd to
> calculate the density:

try g_msd -h
wrong tool.

>
> first I made a index file called density.ndx with g_select, defined the
> solvent within 6A of a residue
>
> after that I try to run g_msd with command:
>
> g_msd_mpi -f md.xtc -s analysis.tpr -n density.ndx -mol diff_mol.xvg -o
> msd.xvg
>
> a dialoug popped up with above command:
>
>
> .....
> Group 13963 (close_27926.000) has     7 elements
> Group 13964 (close_27928.000) has     5 elements
> Group 13965 (close_27930.000) has     7 elements
> Group 13966 (close_27932.000) has     7 elements
> Group 13967 (close_27934.000) has     7 elements
> Group 13968 (close_27936.000) has     8 elements
> Group 13969 (close_27938.000) has     9 elements
> Group 13970 (close_27940.000) has     6 elements
> Group 13971 (close_27942.000) has     9 elements
> Group 13972 (close_27944.000) has     9 elements
> Group 13973 (close_27946.000) has     9 elements
> Group 13974 (close_27948.000) has    10 elements
> Group 13975 (close_27950.000) has     9 elements
> Group 13976 (close_27952.000) has    12 elements
> Group 13977 (close_27954.000) has    10 elements
> Group 13978 (close_27956.000) has     9 elements
> Group 13979 (close_27958.000) has    10 elements
> Group 13980 (close_27960.000) has     8 elements
> Group 13981 (close_27962.000) has    10 elements
> Group 13982 (close_27964.000) has    10 elements
> Group 13983 (close_27966.000) has     7 elements
> Group 13984 (close_27968.000) has     9 elements
> Group 13985 (close_27970.000) has     8 elements
> Group 13986 (close_27972.000) has     8 elements
> Group 13987 (close_27974.000) has     7 elements
> Group 13988 (close_27976.000) has     9 elements
> Group 13989 (close_27978.000) has     6 elements
> Group 13990 (close_27980.000) has     9 elements
> Group 13991 (close_27982.000) has     8 elements
> Group 13992 (close_27984.000) has     8 elements
> Group 13993 (close_27986.000) has    11 elements
> Group 13994 (close_27988.000) has    10 elements
> Group 13995 (close_27990.000) has    11 elements
> Group 13996 (close_27992.000) has    10 elements
> Group 13997 (close_27994.000) has    11 elements
> Group 13998 (close_27996.000) has    11 elements
> Group 13999 (close_27998.000) has     9 elements
> Group 14000 (close_28000.000) has    12 elements
>
>
> I select 14000 which is the last one, but it failed with messages:
>
> rogram g_msd_mpi, VERSION 4.5.5-dev-20121121-3e633d4
> Source code file: /home/albert/software/gromacs/src/tools/gmx_msd.c,
> line: 739
>
> Fatal error:
> The index group does not consist of whole molecules
> For more information and tips for troubleshooting, please check the GROMACS
> website at http://www.gromacs.org/Documentation/Errors
> -------------------------------------------------------
>
> "Can't You Make This Thing Go Faster ?" (Black Crowes)
>
>
> thank you very much
> Albert
>
>
>
>
>
> On 01/08/2013 06:15 PM, Justin Lemkul wrote:
>>
>>
>> On 1/8/13 11:42 AM, Albert wrote:
>>> hello:
>>>
>>>    I've finished a 60ns MD simulation with Gromacs and I found that the
>>> flixbility of solvent molecules inside the protein is different when
>>> it binds
>>> with different ligands: ie. in one case the solvent can move very
>>> fast with bulk
>>> environment, and in other case the solvent forms type Hbonds with
>>> resdiues
>>> inside protein.  I am just wondering how which module of Gromacs can
>>> I use to
>>> indicate the solvent difference in flexibility? Is it possible to
>>> calculate the
>>> entropy in certain region (let's say: 20<Z<30 ) of the solvent?
>>>
>>
>> It sounds like g_rmsf and g_msd may be useful here.  The only way to
>> specify geometric criteria for index groups is to use g_select, but
>> then the analysis has to be done on each individual frame, not the
>> trajectory.  Dynamic selections will be more conveniently implemented
>> in a future Gromacs version.
>>
>> -Justin
>>
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se