A nice writeup on developing torsion parameters is here:
http://hincklab.uthscsa.edu/html/soft_packs/amber6/AMBER-sh-19.3.html
and you'll want to focus in and around "Before fitting the torsional
parameters, we must generate the energy profile for the molecular
mechanical nonbonded potential as was done for the quantum potential,
subtract this curve from the quantum curve, and fit the torsional
potential to the difference potential."
That is from 2000 or so however. You need to setup, analyze, and apply
any quatum mechanical rotational barrier calculations in the context of
the same calculation strategies used to parameterize the rest of the
forcefield you are using.
In a very gross un-empiracal approximation, think of the torsion
parameters as correction. You are adding correcting terms, typically
not super-large in the grand scheme, to bring the Molecular Mechanics
rotational barriers profile into close agreement with the quantum
mechanically derived barriers and curve shapes.
I've not done it in a few years - but I found that, once the QM
rotational barriers were calculated, and pre-torstion-correction MM
barriers were available, it was a very quick process to get coefficients
an dphase-shifts for the needed Torsion parameters using fitting
functions in Mathematica.
And of course, if your atom types are highly analagous to existing atom
types, I suppose you can cheat and guestimate from analagous
parameters. The "gaff" forcefield may also offer some inspiration.
But, everyone should go through the calculations at least once in life....
On 05/30/2017 08:26 AM, Elvis Martis wrote:
> HI,
>
> with this limited information, I am afraid anyone will be able to help you.
>
>
> Best Regards
>
> [photo]
>
>
>
> Elvis Martis
> Ph.D. Student (Computational Chemistry)
> at Bombay College of Pharmacy
>
>
> A Kalina, Santacruz [E], Mumbai 400098, INDIA
> W www.elvismartis.in<https://webapp.wisestamp.com/www.elvismartis.in>
> Skype. adrian_elvis12<https://webapp.wisestamp.com/#>
>
>
> [https://ci3.googleusercontent.com/proxy/P0F8-p0kwxKdscp6zsT-ZSRttk9OJEsBGiaXej_H2ERz8n2ma5SLHFAWJdKL-wqOlXSGjbmEyga9C8lmU1bs-_yPIq3CnazA5eJVDYjce1r-34uwxqjjRnmAtE473lEq28nSHQ=s0-d-e1-ft#https://s3.amazonaws.com/images.wisestamp.com/icons_for_colors_32/linkedin.png]<http://www.linkedin.com/in/elvisadrianmartis/>
>
>
>
>
>
> ________________________________
> From: sylvester kisembo <akiikius.yahoo.com>
> Sent: 30 May 2017 18:45:39
> To: amber.ambermd.org
> Subject: [AMBER] No Torsion terms
>
> Hello, I have tried to run leap and i get a message:
> No torsion terms for:oH-cA-cA-hEoS-cC-cA-hAoH-cA-cA-hE
> Looking at past advice, it is advised to add the terms by hand. My question is where can i derive the values from? I found the torsion terms for other dihedrals in the Lipid14.dat file and added them to my frcmod file but i can't find the specific values for the above torsions.
> Thanks!
>
> _______________________________________________
> 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
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jun 01 2017 - 09:00:03 PDT
