VMD-L Mailing List

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu Oct 04 2007 - 14:28:18 CDT

Just as an addendum, this is more likely to be the problem if your
solvent box is relatively small, such that what I've described can occur
for a nontrivial number of atoms.
Peter

Peter Freddolino wrote:
> Hi Narender,
> this seems very odd... the only thing I can think of that would cause
> this increase is the following: vmd's atomselections do not take
> periodic boundary conditions into account, but namd does. Thus, if you
> have atoms that are between 15 and 20 angstroms from segname PTO1 when
> measured across a periodic boundary, but not within the unit cell, these
> atoms will appear in the interactions calculated by namdenergy only with
> the longer selection. This is particularly likely to be a problem if
> your solvent is not wrapped in your input trajectory (in which case you
> should wrap it prior to using namdenergy).
>
> When you say you're only getting 50 outputs, how many output files are
> there? 100 separate log files should be generated...
>
> Also, for your case you may find it easier to simply use the gui with
> the selections "segname PTO1" and "not segname PTO1" rather than forcing
> the selection to update at each time step. In addition, if you only care
> about vdw interactions there's no need to turn PME on, as that just
> results in extra work.
>
> Best,
> Peter
>
>
> Narender Singh Maan wrote:
>
>> Hello,
>> recently with the help of Peter, I used this script to do NAMDENERGY
>> analysis of my simulated trajectory and it worked fine (almost).
>> My system details: waterbox of 15 Ang from the surface of protein
>> (segname PTO1) containing other protein and ions also. And my aim is
>> to calculate VDW interaction energies between segname PTO1 with its
>> surroundings (to do binding free energy analysis)
>> .............................................................................................................................................................................
>> package require namdenergy
>> set sel1 [atomselect top "segname PTO1"]
>> set sel2 [atomselect top "(within 15 of segname PTO1) and not segname
>> PTO1"]
>> for {set i 0} {$i < [molinfo top get numframes]} {incr i} {
>> $sel1 frame 0
>> $sel2 frame 0
>> namdenergy -sel $sel1 $sel2 -vdw -skip [molinfo top get numframes]
>> -par par_all27_prot_na.prm -extsys x.xsc -ofile elect1-$i.dat -switch
>> 10 -cutoff 12
>> animate delete end 0
>> }
>> ...........................................................................................................................................................................
>>
>> Now i have some doubts and i was wondering if someone can again help
>> me out please:
>> 1) when I changed the set sel2 (line 3) in this script from
>> "(within 15 of segname PTO1) and not segname PTO1"
>> to
>> "(within 20 of segname PTO1) and not segname PTO1"
>> the output of VDW energies changed drastically, which is the way it
>> should behave, but what i don't understand is that later in the script
>> i am specifying the switch and cutoff distance (10 and 12 Ang.) which
>> are both smaller thn the selected values in line 2 so why should there
>> be any difference in the output VDW energies ( i.e. it shouldn't
>> matter what lies beyond 12 Ang. cutoff).
>>
>> 2) Also from this script, in the output i get only half the number of
>> VDW energies, compared to the number of frames i have in my dcd file
>> (i.e. "molinfo top get numframes" is 100 but i only get 50 VDW values
>> in my output)!!
>>
>> I would greatly appreciate any input form the members.
>> thank you
>> narender
>>
>>
>>
>>