[gmx-users] (no subject)

[PeiQuan Chen]

Dear gmx-users：
  Thank you for your(Erik) kindly help!
  Yes. I built the topology directly.I will try to creating a building
block in the rtp file as you told me.But I still have something 
uncertianly to build the topology as follow:
  In the ffoplsaa.rtp,it lists:

[ bondedtypes ]
; bonds  angles  dihedrals  impropers all_dihedrals nrexcl 
     1       1          3          1	    1         3 
and in the ffoplsaa.itp,it lists:
[ defaults ]
; nbfunc	comb-rule	gen-pairs	fudgeLJ	fudgeQQ
   1		   2		 yes	     0.5      0.5

And the user manual tell me that the OPLSAA force field calculate the 
1-4 interaction by scalling,and the [ pairtypes ] section of 
ffoplsnb.itp file is empty.Does it means the [ pair ] section in 
topology of my molecule should be empty?

In the early,someone have told me that for OPLS I don't need to speicfy 
pairs, and you need nrexcl to be two.
But why in the ffoplsaa.rtp file,the nrexcl is to be 3? 
 
If I build block in the rtp file,It must will creat a topology file that 
contain a [ pair ] section that is not empty this time.
Needn't I try to modify the nrexcl to be 2 for making my topology file, 
or just let nrexcl to be 3 ? when the pdb2gmx generate the topology file 
for me,needn't I try to modify the [ pair ] section to be empty?

You also refer that the oplsaa is rather restrictive with improper 
dihedrals.But when I want to keep the pyrrole ring planar during the MD 
simulation, should they be use for our type of conformations?


Thank you in advance!!!!




>Hi,

>Unfortunately I don't have the time to check topologies manually for 
>you, but it looks as if you built the topology directly. I would 
>definitely recommend creating a building block in the rtp file instead, 
>since pdb2gmx can then assign all dihedrals, bonds, angles, etc. 
>automatically.

>In general, OPLS is rather restrictive with improper dihedrals, but 
>check the available OPLS literature to see if they should be use for 
>your type of conformations.

>Other things to do is to make sure you have the right atom types, check 
>the result from the energy minimization (do you get a good energy?), 
>and check the energy from a single step to see if you have huge 
>energies e.g. in bonds/angles/dihedrals when you start from an 
>experimental structure. If you do, locate the faulty interaction by 
>testing what happens when you turn the off (basically a binary search).

>Cheers,

>Erik


Sincerely yours
---- 
PeiQuan Chen
________________________________________________________________________
PeiQuan Chen 	Department of Chemistry, Nankai University
WeiJin Road 94, TianJin, 300071, P.R.China
phone:	(86)2223506728		Mobile Phone: (+86)13920675030
gromacs at 163.com	
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++