[gmx-users] different sets of fudgeQQ and fudgeLJ

[Mark Abraham]

On 18/09/2011 8:58 AM, Yun Shi wrote:
> Hi all,
>
> I want to apply different values of LJ and QQ scaling factors for two 
> interacting molecules A and B. Since I already have the .itp files for 
> each molecule, should I just add something like:
>
> [ defaults ]
> ; nbfunc        comb-rule       gen-pairs       fudgeLJ  fudgeQQ
>    1                    2                     yes             0.5      
>       0.8333
>
> at the very beginning of A.itp and B.itp respectively (the actual 
> values are different from this eg.)?
>
> I wonder if .itp file format was designed for this kind of purpose? 
> :)  I guess we could even define different sets of atom types for A 
> and B, right?

As you will see in the examples in chapter 5, only one [defaults] 
section can exist in an entire topology. Further, the fudge values only 
apply to a subset of non-bonded interactions. For the kind of 
calculation you have in mind, you will need to define new atom types for 
the modified interactions.

Mark