Re: calculating forces from dcd trajectory

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu May 21 2009 - 06:54:08 CDT

Hi Bin,

namdenergy isn't necessarily capable of doing everything that
pairInteractions is; it just provides a convenient interface for the
more common tasks. You should be able to make your own config file using
pairinteractions that will use appropriate groups to give you the force
information that you're after. You should be able to use namdenergy with
the debug flag to get an example config file. Note that the namdenergy
flag for which interaction you're after (eg, -nonb) only affects the
parsing of the logfile, not the namd run itself, so you should be able
to use namdenergy with the groups you're interested in, without
modifying the config file, and then look at the log file to get the
forces you're after.
Best,
Peter

BIN ZHANG wrote:
> Dear Peter:
>
> Thanks for your response.
> It seems to me that only the option *-nonb* available while there are
> two selections in namdEnergy plugin. In this case, would the sum of all
> the forces be outputed in the pairInteraction force column, or only the
> ones coming from nonbonded interactions?
>
> Thanks,
> Bin
>
>
> On May 20, 2009, at 4:51 AM, Peter Freddolino wrote:
>
>> Hi Bin,
>>
>> BIN ZHANG wrote:
>>> One way I can imagine to obtain the force is using the option
>>> *pairInteraction*, and choose the group1 to be a single alpha carbon.
>>> However, this would only give me all the forces coming from the
>>> nonbonded interactions.
>> Why? If group1 is the CA atom and group2 is everything else in the
>> system, you should get what you want.
>>>
>>> The *loadtotalforces* command in the TclForceScript seems to be
>>> another promising way. But one thing I'm not clear is whether
>>> tclforce script requires to run real dynamics.
>>>
>> I believe this should work as well, but you might want to verify it
>> with a simple test case (and pairInteractions should give a simpler
>> way to get the same thing).
>>
>> Best,
>> Peter

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:50 CST