[gmx-users] gmx msd unusual unit

[David van der Spoel]

Den 2020-03-16 kl. 15:10, skrev Justin Lemkul:
> 
> 
> On 3/16/20 10:07 AM, Joe Paul-Taylor wrote:
>> Hello
>>
>> With running gmx msd I am told in the output the y axis unit is nm/ 
>> S2/ N, which is unexpected. Is this simply an print error or is there 
>> an error within the calculation? I have provided a copy of the output 
>> below, noting the units of the self-diffusion coefficient are 
>> sensible. Any insight would be much appreciated.
> 
> The unit is correct. GROMACS does all distance calculations in terms of 
> nm. You get an interpreted output of the MSD in traditional units for 
> the diffusion constant.

The confusion maybe about the xmgrace formatting :)
nm^2 if you use the right software...

> 
> -Justin
> 
>> # Created by:
>> #                       :-) GROMACS - gmx msd, 2019.2 (-:
>> # Command line:
>> #   gmx msd -s topol.tpr -f npt.xtc -b 3000 -e 13000 -o 
>> msd_IRMOF-NHPr_3OCT_8_run_1.xvg
>> # gmx msd is part of G R O M A C S:
>> #
>> # Gromacs Runs One Microsecond At Cannonball Speeds
>> #
>> @    title "Mean Square Displacement"
>> @    xaxis  label "Time (ps)"
>> @    yaxis  label "MSD (nm\S2\N)"
>> @TYPE xy
>> # MSD gathered over 10000 ps with 1001 restarts
>> # Diffusion constants fitted from time 1000 to 9000 ps
>> # D[       UNK] = 0.0031 (+/- 0.0004) (1e-5 cm^2/s)
>>
>> Thank you
>>
>> Joe
>>
>>
> 


-- 
David van der Spoel, Ph.D., Professor of Biology
Head of Department, Cell & Molecular Biology, Uppsala University.
Box 596, SE-75124 Uppsala, Sweden. Phone: +46184714205.
http://www.icm.uu.se