AMBER Archive (2005)

Subject: Re: AMBER: if setting ntf=4, will it cause someproblem?

From: Li Su (suli_at_msu.edu)
Date: Tue May 31 2005 - 16:07:25 CDT


Carlos Simmerling writes:

> the timestep limit is related to the frequencies of motion
> in your system. these frequencies for harmonic oscillators
> are related to the square root of the force constant over
> the mass. So, increasing the force constant increases the
> frequencies and you need smaller time steps. Ignoring the
> normal angle term will not help if you still use a high
> force constant in the angle restraint.
>
> You also need to be careful of making things stiff when
> calculating the PMF- depending on what you are calculating,
> bond angles may change the pmf profile. For example, the
> rotational barriers are affected by the angle bending terms.
>
> Li Su wrote:
>> Hi,
>> I need to make my bond angles very stiff so when setting restraints I
>> have to use very high force constants, but the trick is it seems to me
>> the smallest timestep available is 1 fs. Therefore, I'm thinking about
>> fix all angles involving hydrogen atoms for I will not need to change
>> them to remove the highest frequencies in bong angle calculations. There
>> is a swich "NTF" in amber, when it is set to 4, it means calculation of
>> all bonds and bond angles involving H atoms will be omitted. I am
>> wondering whether it will cause some problem when I am simulating PMF.
>> Thanks a lot!
>>
>>
>> Best Regards,
>> Li
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>

Dear Carlos Simmerling,
Thank you for your kind response. I understand what you have said. But for
some need for comparison, I wanna to fix all the bond angles but vary one.
However, those valence angles involving H atoms are not a big deal in my
case. Actually, if I can constrain those angles, it will be more appropriate
in my case. But the function to constrain angles is not available in amber.
So I'm just wondering if setting ntf=54, what will be done to those angles
involving H bond angles. Does it mean the whole calculation of the force of
thoes angles will be ommitted? And will it interferes my further
calculation?
Thanks a lot.

Best,
Li

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu