From: Alexei Rossokhin (alrossokhin_at_yahoo.com)
Date: Fri Nov 26 2021 - 01:36:07 CST
Dear Peter, thanks for your answer.
When I first ran this script, I was surprised that when choosing the selection as "protein" (without specifying any atom), the script runs over all atoms in the residues, and these calculations are wrong, as I showed in the previous letter. Thus, it seems that some atom should still be specified in selection, for example, "protein and alpha", and in that case the script works correctly. I have just started to learn the tcl language and do not understand all the details. Could you please point me to which operators in the script operate with the name of the atom.
Thanks,Alexey
On Thursday, November 25, 2021, 07:49:24 PM GMT+3, Peter Freddolino <petefred_at_umich.edu> wrote:
Dear Alexey,I think the problem is that this script was written under the assumption that you're operating at a residue level, and thus, that there aren't repeated residue entries in the second argument to rmsd_residue_over_time (given the function name, one would not expect the user to give multiple atoms from the same residue in the function argument). If such entries do exist, it is going to behave oddly because the loops in that function act over the list of residues passed will then be repeatedly adding to the residue level rmsd, and then dividing by it. Now, that script *should* have been written defensively and passed the residue id list through [lsort -unique ...] prior to acting on it, but as it is, there is a clear deviation in your input from what the function is expecting.Best,Peter
On Thu, Nov 25, 2021 at 4:02 AM Alexei Rossokhin <alrossokhin_at_remove_yahoo.com> wrote:
Dear NAMD users, While browsing the NAMD tutorial I found the residue_rmsd.tcl script to calculate RMSD for individual residues. When I tried to run it, the results puzzled me a little.
When I used the selection "protein and chain A and resid 174 and name CA and noh" for my MD calculations I got the following result
RMSD of residue 174 is 1.3978458595465855
Average rmsd per residue: 1.3978458595465855
When I used the selection "protein and chain A and resid 174 and name CA CB CG CD and noh" I got the following result
RMSD of residue 174 is 5.591383438186342
RMSD of residue 174 is 0.022276428040583034
RMSD of residue 174 is 8.875070932503201e-5
RMSD of residue 174 is 3.535884833666614e-7
Average rmsd per residue: 1.4034372426311832
Using "protein and chain A and resid 174 and name CB and noh" I got
RMSD of residue 174 is 1.3978458595465855
Average rmsd per residue: 1.3978458595465855
If to run the script separately for CA or CB result is the same. If to run the script with the selection "name CA CB CG CD", the result for individual atoms is strange, although the average is close in all cases.
What does it mean, any bug in the script?Thank you for help.
Alexey
PS I attached the residue_rmsd.tcl script.
This archive was generated by hypermail 2.1.6 : Fri Dec 31 2021 - 23:17:12 CST