What's happening is that tleap is reading along the parm file for
parameters that address the arrangements c3-c3-os-c3 or c3-os-c3-c3. It
first finds c3-os-c3-c3 and logs a cosine term of amplitude 0.240 kcal/mol
and periodicity 3. The negative in front of the 3 indicates that this is
NOT the last term for the dihedral, so it keeps reading terms until it
finds one with a positive value of the peridiocity. That's the next line,
with an amplitude of 0.160 kcal/mol. I'm not sure how the latter lines got
include at all, or why only the one with periodicity 1 and amplitude 0.000
would be included. It may be something in the code that says "stop reading
negatives for this dihedral once a positive periodicity is found" but
doesn't make provisions for the existence of additional positive
periodicities.
This is a pretty weak point in the code, and has been discussed before in
the context of similar problems.
Dave
On Mon, Aug 5, 2019 at 3:51 PM Song, Lin <songlin3.chemistry.msu.edu> wrote:
> Hi Johannes and Dr. Case,
>
> May I ask if you have figured out which lines should be used as the
> c3-c3-os-c3 torsion parameter? I’m confused about this as well. It seems to
> me there’re two sets of parameters excluding the first line, one is for
> TorType=2, the other is for TorType=3. Could you also explain what does the
> TorType mean?
>
> Thank you!
>
> Best,
> Lin
>
> > On Mar 18, 2019, at 6:41 AM, Träg, Johannes <johannes.traeg.fau.de>
> wrote:
> >
> > Am 2019-03-15 20:20, schrieb David Case:
> >> On Fri, Mar 15, 2019, Träg, Johannes wrote:
> >>>
> >>> I have a question regarding the assignment of GAFF2 parameters
> >>> (gaff2.dat, Version 2.1, April 2016) to the dihedral angle
> >>> c3-c3-os-c3.
> >>>
> >>> For this dihedral angle, the gaff2.dat file lists the following lines:
> >>>
> >>> X -c3-os-X 3 1.150 0.000 3.000
> >>> JCC,7,(1986),230
> >>>
> >>> c3-os-c3-c3 1 0.240 0.000 -3 p29 GA
> >>> c3-os-c3-c3 1 0.160 0.000 2 p29 GA
> >>>
> >>> c3-c3-os-c3 1 0.910 0.000 -3
> >>> c3-c3-os-c3 1 1.000 0.000 -2
> >>> c3-c3-os-c3 1 0.000 0.000 1
> >>>
> >>> Of course the fist line has to be omitted, since there are explicit
> >>> entries for c3-c3-os-c3/c3-os-c3-c3. However, the .frcmod file
> >>> prepared
> >>> by parmchk2 lists the following lines:
> >>
> >> Can you say exactly what arguments you gave to parmchk2 to get this
> >> result? Did you include the "-s 2" option? (cc-ing to Junmei for
> >> comments as well.)
> >>
> > I used tha following command: parmchk2 -i ff.mol2 -f mol2 -o ff.frcmod
> > -s 2 -a Y
> >
> >
> >>>
> >>> c3-c3-os-c3 1 0.240 0.000 -3.000
> >>> c3-c3-os-c3 1 0.160 0.000 2.000
> >>> c3-c3-os-c3 1 1.000 0.000 -2.000
> >>> c3-c3-os-c3 1 0.000 0.000 1.000
> >>>
> >>> Finally, the .prmtop file prepared by leap contains only the values of
> >>> theses lines:
> >>>
> >>> c3-c3-os-c3 1 0.240 0.000 -3.000
> >>> c3-c3-os-c3 1 0.160 0.000 2.000
> >>> c3-c3-os-c3 1 0.000 0.000 1.000
> >>>
> >>> Why is the line (1/0.910/0.000/2) missing in .frcmod? How (and why) do
> >>> antechamber/leap choose the final values for the .prmtop file and not
> >>> the lines (1/0.910/0.000/-3 1/1.000/0.000/-2) instead?
> >>>
> >>
> >> ...dac
> >>
> >>
> >> _______________________________________________
> >> AMBER mailing list
> >> AMBER.ambermd.org
> >>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.ambermd.org_mailman_listinfo_amber&d=DwIGaQ&c=nE__W8dFE-shTxStwXtp0A&r=3FEhv0ODXmqNZDIrYntxr0EDkmWDtT3vpmO7cMu9ps0&m=pkYs2Xl7aZfzDxn0UqRZZoPCjw5CyK6FzmnxlIHLnCM&s=N3ZxJmHgDK2wjMFRS-Ip9gAQg0N_m0Lb64002pW4Tvw&e=
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> >
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.ambermd.org_mailman_listinfo_amber&d=DwIGaQ&c=nE__W8dFE-shTxStwXtp0A&r=3FEhv0ODXmqNZDIrYntxr0EDkmWDtT3vpmO7cMu9ps0&m=pkYs2Xl7aZfzDxn0UqRZZoPCjw5CyK6FzmnxlIHLnCM&s=N3ZxJmHgDK2wjMFRS-Ip9gAQg0N_m0Lb64002pW4Tvw&e=
>
> _______________________________________________
> 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 Aug 05 2019 - 13:30:01 PDT
