Re: changing solventDielectric in GBIS

From: Andres Colubri (
Date: Mon Jun 13 2011 - 08:17:36 CDT

Ok, thanks for the tip. Using separate runs seems like a good idea in this

About your second comment, I agree with your point and it is quite likely
that the force field is not precise enough to be affected by these changes
in the dielectric.

In any case, I was interested in comparing the results from the GBIS model
with explicit solvent simulations. So I was trying to reproduce the
simulation protocol I had in the configuration files for my periodic water
box sims, which include a heating cycle.

Actually, this leads me to another question. The heating cycle in the
explicit solvent simulations is between 10 and 300 degrees, but I didn't
find much reliable information about water dielectric under the freezing
point, so this is why I started at 275 in the implicit solvent sims. In the
case I want to use GBIS to perform runs under 273 degrees, which
should I use?


On Mon, Jun 13, 2011 at 9:50 PM, Ajasja Ljubetič

> I would solve this by doing separate runs. You can
> access environment variables in TCL using $env(NAME_OF_VARIABLE). So you can
> have one conf file and a separate bash scripts that sets
> the appropriate environment variables and then runs NAMD.
> Also, is the force field precise enough that such small changes in the
> dielectric constant are even relevant?
> Best regards,
> Ajasja
> On Mon, Jun 13, 2011 at 14:36, Andres Colubri <>wrote:
>> Hello,
>> I'm simulating a heating cycle using the new GBIS model in NAMD,
>> starting at T=275 K and ending at T=300K, with increments of 5 degrees.
>> Since the dielectric of water goes from 88 at 0C to 78 at 28C, the
>> simulation script should update the solventDielectric parameter accordingly
>> at each temperature step. So I have the following in my config file :
>> # Heating
>> for {set i $temperature0} {$i <= $temperature1} {incr i 5} {
>> set temperaturei $i
>> langevinTemp $temperaturei
>> langevinPistonTemp $temperaturei
>> solventDielectric $dielectric($temperaturei)
>> }
>> with:
>> set temperature0 275
>> set temperature1 300
>> and
>> set dielectric(275) 87.31
>> set dielectric(280) 85.46
>> set dielectric(285) 83.61
>> set dielectric(290) 81.76
>> set dielectric(295) 79.91
>> set dielectric(300) 78.06
>> However, I get this error in the first iteration of the loop:
>> TCL: Setting parameter solventDielectric to 87.31
>> FATAL ERROR: Setting parameter solventDielectric from script failed!
>> Any help with this will be greatly appreciated.
>> Thanks!

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:24:04 CST