Re: [AMBER] can not be a parameter fitting

From: Jason Swails <jason.swails.gmail.com>
Date: Mon, 2 Feb 2015 08:28:22 -0500

> On Feb 2, 2015, at 12:54 AM, Takuya Uto <t.uto.cc.miyazaki-u.ac.jp> wrote:
>
> Dear Sir
>
>
>
> As shown in the tutorial A23,
> I tried the fitting of force field parameters for small molecule that is
> different from the tutorial.
> Currently , about 1- phenyl ethyl alcohol as a model molecule ,
> I have tried the fitting parameters of CA -CA -CT -OH, CA- CT- OH- HO.
>
> I can not determine reasonable parameters are also doing many times.
> (Compared with the QM calculation , it does not reproduce the PES.)

Your description of what you are doing has very little detail, to the point where I really don’t have any idea what it is that you’re doing. Please try to be more descriptive.

> Although I tried to various verification ,
> Especially at the stage of the "Derive K",
> I noticed that the R ^ 2 value is extremely bad . (R ^ 2 = 0.13)


> For methoxybenzene is more simple molecules , I was tried to parameters
> fitting tee group of CA-CA-OS-CT, was going on is still the same thing .
>
> I had continued parameter fitting , but eventually R ^ 2 value was 0.95 or
> more , as compared to the PES of the QM calculations gave very unreasonable
> results .

I don’t understand what this means.

> I am very confusing. Please help me.

Please be precise with your descriptions of what you are doing. How are you defining the results to be “unreasonable”? Are you fitting these parameters by hand? Are you using paramfit by following the tutorial?

Good luck,
Jason

--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 02 2015 - 05:30:05 PST
Custom Search