[gmx-users] g_msd and Diffusion Constant

[V.V.Chaban]

If the error bar exceeds the absolute value of diffusion coefficient,
just round the number, as I exemplified below.

The error bar should correspond to the last digit of the result value,
e.g. 5 +/ -2; 0.0002 +/- 0.0001; 0.4E6 +/- 0.2E6, etc.

My immediate guess is that the problem in the code is somewhere with rounding.



On Wed, Jul 22, 2015 at 1:49 AM, Ganesh Shahane
<ganesh7shahane at gmail.com> wrote:
> Yes, for some calculations the error bar values did exceed the diffusion
> coefficient values.
>
> I am using the double-precision version of gromacs. But then again the
> question is whether to consider the value displayed on the terminal screen
> or the one displayed in the .xvg file.
>
> On Tue, Jul 21, 2015 at 10:25 PM, V.V.Chaban <vvchaban at gmail.com> wrote:
>>
>> I have a strong belief that an error bar for such numbers exceed the
>> diffusion coefficient values.
>>
>> Newer gromacs versions (or is it just because of a double-precision
>> compilation?) output more digits. I do want to hope these digits are
>> meaningful....
>>
>> If I got this situation, I would report 10^(-8)cm^2/s.
>>
>>
>>
>>
>>
>> On Tue, Jul 21, 2015 at 2:14 AM, Ganesh Shahane
>> <ganesh7shahane at gmail.com> wrote:
>> > Hi!
>> >
>> > I am using gromacs version 4.5.5. The values are different. The value
>> > displayed on the terminal screen is 0.0007417*1e-5 cm^2/s while the
>> > value
>> > displayed in the chol.xvg file is 0.000903331*1e-5 cm^2/s.
>> >
>> > As you can see, both values are quite close though the value in chol.xvg
>> > doesn't appear to be round up.
>> >
>> > On Tue, Jul 21, 2015 at 10:20 AM, Chandan Choudhury <iitdckc at gmail.com>
>> > wrote:
>> >>
>> >> Hi!!
>> >>
>> >> Are the values different? Can you show us both the values?
>> >> I presume both values would be same.
>> >> The value in the xvg file would be round up.
>> >>
>> >> Chandan
>> >>
>> >>
>> >> On Tue, Jul 21, 2015 at 1:33 AM, V.V.Chaban <vvchaban at gmail.com> wrote:
>> >>
>> >> > which version fo you use?
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> > On Mon, Jul 20, 2015 at 8:28 AM, Ganesh Shahane
>> >> > <ganesh7shahane at gmail.com> wrote:
>> >> > > Dear Gromacs Users,
>> >> > >
>> >> > > I have performed a simulation of a POPC+Cholesterol bilayer. I wish
>> >> > > to
>> >> > know
>> >> > > the diffusion constant of a particular cholesterol molecule.
>> >> > > Towards
>> >> > this,
>> >> > > I performed a g_msd using the command as follows-
>> >> > >
>> >> > > g_msd -f 30chol.xtc -s prod.tpr -lateral z -endfit 800000 -n
>> >> > > index.ndx
>> >> > -mol
>> >> > > chol.xvg
>> >> > >
>> >> > > where index.ndx contains a single index for the cholesterol atoms
>> >> > > in
>> >> > > question. In the end, I got 2 different values of diffusion
>> >> > > constant,
>> >> > with
>> >> > > one of them being displayed on the terminal screen and the other
>> >> > > value
>> >> > > in
>> >> > > the chol.xvg file.
>> >> > >
>> >> > > Does anyone have any idea as to which value should be taken as the
>> >> > > value
>> >> > of
>> >> > > diffusion constant? Any help would be appreciated. Thank you!
>> >> > >
>> >> > > --
>> >> > > Best Regards,
>> >> > > Ganesh Shahane
>> >> > > --
>> >> > > Gromacs Users mailing list
>> >> > >
>> >> > > * Please search the archive at
>> >> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> >> > posting!
>> >> > >
>> >> > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>> >> > >
>> >> > > * For (un)subscribe requests visit
>> >> > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>> >> > > or
>> >> > send a mail to gmx-users-request at gromacs.org.
>> >> > --
>> >> > Gromacs Users mailing list
>> >> >
>> >> > * Please search the archive at
>> >> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> >> > posting!
>> >> >
>> >> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>> >> >
>> >> > * For (un)subscribe requests visit
>> >> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> >> > send a mail to gmx-users-request at gromacs.org.
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> Chandan Kumar Choudhury
>> >> National Chemical Laboratory, Pune
>> >> India
>> >>
>> >> *"All work and no play makes Jack a dull boy...”*
>> >> --
>> >> Gromacs Users mailing list
>> >>
>> >> * Please search the archive at
>> >> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> >> posting!
>> >>
>> >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>> >>
>> >> * For (un)subscribe requests visit
>> >> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> >> send
>> >> a mail to gmx-users-request at gromacs.org.
>> >
>> >
>> >
>> >
>> > --
>> > Best Regards,
>> > Ganesh Shahane
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send
>> a mail to gmx-users-request at gromacs.org.
>
>
>
>
> --
> Best Regards,
> Ganesh Shahane