CHARMM Development Project Forums User Discussion & Questions Energy terms, Constraints, Restraints, and Solvation Implicit solvation models

Dear all,
I have a high interest in using implicit solvation models implemented in CHARMM. However, being an amateur in this field I I'd like to know more about performances in CHARMM in general, and the Generalized Born Molecular Volume approach in particular. For the latter, I've tried to do single-point energy calculations with the analytical approach, but have encountered some syntactic problems, and therefore I ask for a sample input file concerning a protein-DNA system, and using the 27 all atom rtf/param set.

Thanks!

Johan

Did you ever check $charmm_home/test/c27test/genborn22.inp?

Yes,
I've looked through all test-cases (always the first thing to do), but I want to learn more about the GB Molecular Volume method, since it is claimed to work on any force-field. In particular, I'm interested in the analytical approach and I've already tried the first and second methods of GBMV.

Johan

Are you experienced with implicit solvation techniques? I have some problems with charged systems in this aspect, so if you know how to work-around such problems, I'd really be happy to know.

I am sorry, I am not experienced with them. Actually, I just used GB recently, and my personal IMMATURE understanding is:
GB model is mainly to attain quantitative agreement with PB results. And even for such a simple goal, it is still parameterized to attain it. The GB formula, to me, is doubtful.

Therefore, I gave up GB. And since PB is not expensive to my problems, I am using PB.

I think we need get comments from those more experienced persons.

Yes,
but the GBMV method (in c30a1) gives more "reasonable" results, but I'll probably have to resort to PB as well . Are you experienced in treating metals within the framework of PB?

Maybe GBMV is better, I am using c29b2, :-(
Sorry, I don't have experience of dealing with metals. There some some useful information on professor Benoit Roux 's website.