This title appears in the Scientific Report :
2014
Please use the identifier:
http://dx.doi.org/10.1021/ct500084y in citations.
Extension of the FACTS Implicit Solvation Model to Membranes
Extension of the FACTS Implicit Solvation Model to Membranes
The generalized Born (GB) formalism can be used to model water as a dielectric continuum. Among the different implicit solvent models using the GB formalism, FACTS is one of the fastest. Here, we extend FACTS so that it can represent a membrane environment. This extension is accomplished by consider...
Saved in:
Personal Name(s): | Carballo-Pacheco, Martín |
---|---|
Vancea, Ioan / Strodel, Birgit (Corresponding Author) | |
Contributing Institute: |
Strukturbiochemie; ICS-6 |
Published in: | Journal of chemical theory and computation, 10 (2014) 8, S. 3163 - 3176 |
Imprint: |
Washington, DC
American Chemical Society (ACS)
2014
|
PubMed ID: |
26588287 |
DOI: |
10.1021/ct500084y |
Document Type: |
Journal Article |
Research Program: |
Structural Biology |
Publikationsportal JuSER |
The generalized Born (GB) formalism can be used to model water as a dielectric continuum. Among the different implicit solvent models using the GB formalism, FACTS is one of the fastest. Here, we extend FACTS so that it can represent a membrane environment. This extension is accomplished by considering a position dependent dielectric constant and empirical surface tension parameter. For the calculation of the effective Born radii in different dielectric environments we present a parameter-free approximation to Kirkwood’s equation, which uses the Born radii obtained with FACTS for the water environment as input. This approximation is tested for the calculation of self-free energies, pairwise interaction energies in solution and solvation free energies of complete protein conformations. The results compare well to those from the finite difference Poisson method. The new implicit membrane model is applied to estimate free energy insertion profiles of amino acid analogues and in molecular dynamics simulations of melittin, WALP23 and KALP23, glycophorin A, bacteriorhodopsin, and a Clc channel dimer. In all cases, the results agree qualitatively with experiments and explicit solvent simulations. Moreover, the implicit membrane model is only six times slower than a vacuum simulation. |