Hello AMBER list,
I figured out my problem but I wanted to document the explanation so no one else runs into the same pitfall. The mistake was using wildcards when the PN is odd: In this molecule, the s-trans conformation is higher in energy than the s-cis, but applying the same PN=1.0 dihedral parameter to all four dihedrals resulted in two dihedrals canceling out the other two. What I infer, then, is if your PN is even you're free to use wildcards, but when your PN is odd you have to worry about terms canceling each other out. I fixed the problem by defining all my dihedrals explicitly and adjusting the phase by 180° on half of the PN=1 dihedrals.
Hope this helps someone with a similar problem down the road.
-Scott
On Jan 6, 2014, at 12:32 PM, Jason Swails wrote:
>
> On Mon, 2014-01-06 at 00:52 +0000, Allen, Scott Edward wrote:
>> Hello,
>>
>> I'm trying to parametrize a dihedral in my small molecule. I find that when I incorporate my results into my .frcmod file, AMBER (tleap?) only reads the second term of the X -ca-cc-X dihedral (the phase=180 term). Has anyone had a similar problem? The full text of the .frcmod file is given below.
>
> What makes you say this? Did you use ParmEd or rdparm to verify that
> the first term is missing from the relevant torsion terms?
>
> Also note that "X -ca-cc-X " is only used when a specific torsion is not
> present in any of the parameter databases. It's possible that one of
> the terms you hoped to modify with your torsion wild-card has a specific
> definition in the gaff.dat file. In this case, the only way to override
> that torsion term is to use specific torsions (i.e., with full atom
> types as the first and last atoms rather than wildcards).
>
> HTH,
> Jason
>
>>
>> Thank you,
>> Scott Allen
>>
>> remark goes here
>> MASS
>>
>> BOND
>>
>> ANGLE
>>
>> DIHE
>> X -ca-cc-X 4 1.800 0.000 -1.000 from GA
>> X -ca-cc-X 4 3.500 180.000 2.000 from GA
>> cc-ss-cc-cd 1 1.100 180.000 2.000 same as X -c2-ss-X
>> cc-ss-cc-h4 1 1.100 180.000 2.000 same as X -c2-ss-X
>> cc-ss-cc-nd 1 1.100 180.000 2.000 same as X -c2-ss-X
>> cc-ss-cc-ca 1 1.100 180.000 2.000 same as X -c2-ss-X
>>
>> IMPROPER
>> ca-nd-cc-ss 1.1 180.0 2.0 Using default value
>> cc-h4-cd-nd 1.1 180.0 2.0 Using default value
>> cd-h4-cc-ss 1.1 180.0 2.0 Using default value
>> ca-cc-ca-nb 1.1 180.0 2.0 Using default value
>> ca-ca-ca-ha 1.1 180.0 2.0 General improper torsional angle (2 general atom types)
>> ca-h4-ca-nb 1.1 180.0 2.0 Using default value
>>
>> NONBON
>>
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jan 06 2014 - 12:30:02 PST
