[gmx-users] RMSD graph over time and g_cluster RMSD distribution seem at odds with each other

[Tsjerk Wassenaar]

H Massimo,

The RMSD over time is the RMSD against a single reference structure. For
clustering you use the RMSD matrix, which has all pairwise RMSDs. The
distribution you show is the distribution of RMSD values in that matrix.

Cheers,

Tsjerk



On Wed, Aug 13, 2014 at 3:44 PM, ms <devicerandom at gmail.com> wrote:

> Hi Tsjerk,
>
>
> On 8/13/14 3:19 PM, Tsjerk Wassenaar wrote:
>
>> You have two clusters, both with low within-cluster RMSD. This shows up as
>> the first peak in the RMSD distribution. The second peak is the
>> between-clusters RMSD.
>>
> I am sorry, but I don't understand anything here - what does
> "within-cluster RMSD" means? what is the "between-clusters RMSD"?
>
> I would expect the RMSD distribution to mirror the RMSD vs time plot ;
> that is, if I take the points from the rmsd-time.xvg and I make an
> histogram, I would expect to recover the RMSD distribution. Am I wrong?
>
> thanks a lot for your help,
>
> m.
>
>  Cheers,
>>
>> Tsjerk
>>
>>
>>
>> On Wed, Aug 13, 2014 at 3:14 PM, ms <devicerandom at gmail.com> wrote:
>>
>>  Dear Gromacs users,
>>>
>>> I am analyzing the RMSD evolution of a trajectory with g_rms and
>>> clustering it with g_cluster (GROMOS, cutoff 0.1). I am noticing
>>> something
>>> quite odd in the RMSD distribution. I link here the RMSD evolution over
>>> time , the cluster ID over time, and the RMSD distribution (see
>>> http://imgur.com/a/0ftA2 ).
>>>
>>> RMSD evolution in time and cluster ID in time seem to make sense -the
>>> larger cluster is cluster 1, and it indeed populates most of the
>>> trajectory.
>>>
>>> The RMSD distribution is also correctly bimodal and peaked around the
>>> right values, at least visually. However, I would expect that the peak
>>> around 0.3-0.4 should be larger than the peak around 0.1, given the time
>>> evolution/distribution of RMSD, yet I see the opposite, apparently. What
>>> does it mean?
>>>
>>> Thanks,
>>> Massimo
>>> --
>>> 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.
>



-- 
Tsjerk A. Wassenaar, Ph.D.