[gmx-users] Re: Cross-correlation maps

[Mark Abraham]

On 24/08/2012 4:51 PM, James Starlight wrote:
> up :)
> It's appeared two additional questions.
>
> 1) In addition to the pca's cross-correlation maps I wounder to know
> about possibility of calculation of such cross-correlation's from the
> trajectories indirectly without calculation of the covariance
> matrices.
>
> 2) is there any way to calculate degree of fluctuations of side-chains
> ( degree of  torsion's dynamics) from the different trajectories and
> to compare it ?
> E.g I have one protein in two different (apo and holo) forms. I've
> calculated two trajectories for both structures and observed different
> degree of dynamics in case of each structure ( e.g fluctuations in
> case of apo form were  more frequent than in case of liganded form).
> IS there any way to direct mesure and comprison of such side-chain's
> dynamics for two trajectories?

Yes, but not with GROMACS tools. You can generate a .tpr for use with 
that old version of g_covar with an old version of grompp in a manner 
similar to the way you generate one now, but there are bunch of details 
different. You may not require a .tpr, of course. Look at the file types 
for g_covar_mod -s in the output for -h.

Mark

>
>
> James
>
> 2012/8/15, James Starlight <jmsstarlight at gmail.com>:
>> Dear Gromacs users!
>>
>>
>> I want to obtain Cross-correlation maps ( for indication of the
>> cross-correlated fluctuations of the residues).
>> The example of such maps can be found here
>> http://pubs.acs.org/doi/abs/10.1021/ja076046a
>>
>>   I found that modificied version of the G_covar from users
>> contributions can do such things. But because of the older version of
>> that program (3.3.3) I've obtained the below error using it with 4.5.5
>> gromacs
>>
>>
>> Reading file md_GO.tpr, VERSION 4.5.5 (single precision)
>>
>> -------------------------------------------------------
>> Program g_covar_mod, VERSION 3.3.3
>> Source code file: tpxio.c, line: 1192
>>
>> Fatal error:
>> reading tpx file (md_GO.tpr) version 73 with version 40 program
>> -------------------------------------------------------
>>
>>
>> Is there newest versions of the G_covar for such things or
>> alternativelly any others ways to calculate such correlations maps ?
>>
>> Thanks for help
>>
>> James
>>