[gmx-users] Re: how to write the correct top file for membrane protein

[Mark Abraham]

liang wrote:
> Dear Mark,
> 
> Thanks for your advice, but from my understanding, Berger's force field 
> is based on OPLS/Amber.
> Lipid itp files from Tieleman's website just use the atom type 
> definitions from GROMOS87.
> And GROMOS87 is out of fashion now.
> So thats why i think it would be better to use OPLS for protein and 
> "ffgmx.itp" for lipid.

Fine, but you can't use them both at the same time (unless you want to 
produce indefensible gibberish), and ffgmx is deprecated anyway.

> I am not sure whether I misunderstand the usage of Berger's force field, 
> if i am wrong,
> please give me further instruction.

See my advice last time.

>> Don't mix forcefields unless you already have hard evidence that this
>> combination works - and I'm not aware of any. The link Justin provided
>> warns you about that. You need to find a force field that has parameters
>> optimized that suit your system. Mixing anything else is about 99% likely
>> to make an expensive (and poor) random number generator.

Look in the literature for systems similar to yours and see if their 
ideas on force fields are suitable for you to use. Making it up based on 
stuff you found on the web is for high school students doing assignments :-)

Mark