[gmx-users] OPLS force field issue....

[Justin Lemkul]

On 2/3/14, 10:14 PM, Sidath Wijesinghe wrote:
> Justin,,
>
> i rechecked the opls/ff that i used and found some errors, after correction them
> i got the .top file.
> note that i was able to define the charge groups as well. when i use this .top
> file with grompp,
> now am getting 12 errors...as follwos..
>
>
> ERROR 1 [file topol.top, line 1399]:
>    No default Ryckaert-Bell. types
>
>
> ERROR 2 [file topol.top, line 1400]:
>    No default Ryckaert-Bell. types
>
>
> ERROR 3 [file topol.top, line 1415]:
>    No default Ryckaert-Bell. types
>
>
> ERROR 4 [file topol.top, line 1416]:
>    No default Ryckaert-Bell. types
> and so on...
>
> i did the mapping as you suggested,as an example line 1399 in .top
> as follwos...
>   2     1    82    81     3  ( here 2 1 82 & 81 are carbon atoms)
>
> and the mapping to all 12 errors.....then  modified the ffbonded.itp file's
> diheadral block as follows...
>
> C2     C1    C82    C81       3    0.00  0.00 0.00 0.00
>
> so on....
>
> then bond tyype block as follows....
>
> C2     C1      1 0.00 0.00... so on....
>
> added those atom types in atomtypes.atp like below...
> C1
> C2
> C3
> C4
> C11
> C12
> C13................so on
>
> after that when i do grompp  gives me a fatal error like
>
> Program grompp, VERSION 4.5.5
> Source code file: toppush.c, line: 631
>
> Fatal error:
> Unknown bond_atomtype C1
>
> could you please tell me what i have done wrong?
>

You used atom names, not atom types.  In [atoms] in your .top, each atom type is 
assigned as opls_XXX.  Looking in ffnonbonded.itp, the second column of the 
corresponding opls_XXX type is its bonded type equivalent.  That is the 
identifier you need to be using.  The force field doesn't care what atoms are 
named, but it does care what type they are.

> and at the same time....i tied delete those 12 lines in .top and do grompp...it
> worked....but when i
> did mdrun.....i am getting the signs of blowing up system....what i can do over
> here....how can i get rid of these errors
> and carry out my simulation?
>

Of course if you delete an essential bonded interaction, the simulation will go 
haywire.  Deletion may be convenient for getting grompp to shut up, but it's not 
sound practice.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================