CHARMM Development Project Forums User Discussion & Questions General Chemistry Discussions how to consider the hydrophobic force in charmm?

Dear Charmm Users,
As we known, in traditional generalized Born (GB) model, the hydrophobic effect is considered by taking the solvent accessible surface area (SASA) as a proxy of the its extend in energy. The nonpolar solvation free energy isn't the function of coordinate, so the hydrophobic force isnot obtained by the derivative of energy with respect to its coordinate. The model may be controversial because we only account for its impact in energy and the hydrophobic force is neglected in MD simulation using AMBER software. In the charmm, is the nonpolar solvation free energy also calculated by the aSASA+b? Is the hydrophobic force considered in MD, if yes, how to add the hydrophobic force in every MD step? Thank you very much !

Two ways: Use explicit solvent. Most implicit solvent implementations in CHARMM also contain a "hydrophobic" component, see eg the SA and GBVDW keywords for the GBMV methods (gbmv.doc). Read the 2009 J Comp Chem CHARMM paper, section on implicit models.

I have read the papers: J Comput Chem 24: 1348–1356, 2003 and the 2009 J Comp Chem CHARMM paper. I donot find the discussion for hydrophobic force in MD using implicit model. Thank you !

"hydrophobic" and "hydrophobicity" are mentioned several times in Section III.D of the 2009 paper, and also in the 2003 paper. Most of the models in CHARMM deal with solvation free energies, which include both polar and non-polar components
So I think the answer to your question is, yes, forces due to the "hydrophobic" effect are included at each MD step.
For technical details you can start by looking at the papers referenced in the two JCC papers.

Also, I might have misunderstood the original post, but the SASA is most definitely a function of the coordinates.

If the SASA is a function of the coordinates,and there is a derivative of this function to get hydrophobic forces. For the linear protein, the added hydrophobic force will be very strong and it will cause the linear protein become the global protein. Is it unreasonalbe? Thanks!

Globular proteins are common, but there are also proteins with quite elongated shape.

The question again is very difficult to understand, but the "hydrophobic force" (which really is an umbrella term for the interplay of interactions and effects that makes it favorable from a free energy point of view to minimize the solvent-exposed hydrophobic surface) is an important driving force in protein folding. Most practical implicit solvent models are able to reproduce this effect at least to some extent, else they wouldn't be of much value.