From: P.-L. Chau (pc104_at_pasteur.fr)
Date: Fri May 04 2012 - 03:33:57 CDT
Thanks for your message.
> Getting the coordinates from the PDB is straightforward. You must check
> your script. The script you used works for the files in the tutorial,
> but now if you use your own PDB, first check that such PDB is similar to
> the one in the tutorial.
My PDB has only ATOM lines, and is generated by VMD (I manually deleted
the non-ATOM lines), so it should be quite standard.
> Now, I can bet $1000 that the coordinates in the PDB match the ones you
> get from the tcl console. [...]
Thank you for the script. I should explain that the purpose of my work is
not just to obtain the cartesian coordinates of five atoms. I would like
to obtain the coordinates of these five atoms at each timstep of an NAMD
molecular dynamics run, then calculate the best-fitting circle for these
five atoms, and then apply a radially outward force from the centre of the
circle to each atom.
I have been slowly building up my script. I have got the atom
identification part working. Now I am trying obtain the coordinates of
these five atoms, and I have so far failed.
So thank you for the tcl script, but I do not plan to do this on a tcl
console, but within a NAMD run. I have gone to brass tacks and asked NAMD
to print out all the atomic coordinates:
[...]
wrapmode input
proc calcforces {step unique Rrate Rtarget K} {
global aalist
puts $aalist
while {[nextatom]} {
set rtest [getcoord]
puts $rtest
set atomid [getid]
if { [lsearch $aalist $atomid] >= 0 } {
set rvec [getcoord]
puts $rvec
foreach { x y z } $rvec {break}
puts "$x $y $z"
}
}
}
[...]
and the list of $rtest I see from my NAMD output is completely different
from the coordinates I put in. More worryingly, NAMD seems to be adding
digits to the coordinates read in. For example, the first atom of the PDB
file is (-60.120,-65.596,21.040), but NAMD thinks its coordinates are
(-63.3342148731, -56.9164711539, -74.0969765657). Where on earth does NAMD
get the trailing digits from?
Something seems to be failing in a very fundamental way. I should be truly
grateful for any suggestions.
Thank you very much indeed!
P-L Chau
email: pc104_at_pasteur.fr
Bioinformatique Structurale
CNRS URA 2185
Institut Pasteur
75724 Paris
France
tel: +33 1 45688546
fax: +33 1 45688719
This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:21:57 CST