On Oct 15, 2011, at 12:15 PM, Sajeewa Pemasinghe <sajeewasp.gmail.com> wrote:
> Hello everybody,
>
> I have calculated the RMSD of a protein which consists of 991 residues for
> 20ns in two ways.
>
> 1) I have considered the whole protein at once
>
> For example
>
> rms reference out totalprotein.dat :1-991.N,CA,C time 1
>
>
> 2) I have calculated the RMSD separately for each residue over the 20ns
> period
>
> For example
>
> rms reference out residue1.dat :1 time 1
> rms reference out residue2.dat :2 time 1
> rms reference out residue3.dat :3 time 1
> rms reference out residue4.dat :4 time 1
> rms reference out residue5.dat :5 time 1
Each time you run the rms command here, it realigns to the given mask. For a large system, this doesn't say very much. It can align a single residue by contorting the rest of the structure to any orientation. Furthermore, you do the total rms on only some backbone atoms, yet do the per-residue rms commands on all atoms in that residue, which is also not a fair comparison. Try adding the keyword nofit to the per-residue rms commands and make sure that the atoms used to calculate the rms is the same in each case.
HTH,
Jason
> .............................................................
> ...............................................................
> rms reference out residue991.dat :991 time 1
>
>
> Then I calculated the average of the values given by 1) which was a single
> value around 3.2 angstrom
> Then I calculated the average of each individual residue using the output
> of 2) which give 991 average values most of which are around or below 1
> angstrom (non of them above 3.2)
>
> * I was wondering If the whole protein had an average RMSD of 3.2 angstrom
> why wouldn't the average RMSD of individual residues be close to the 3.2
> mark?*
> *
> *
> This might sound like a silly question but your response is highly
> important to me.
>
> Thank you
>
> Sajeewa Dewage
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Oct 15 2011 - 10:30:02 PDT
