CCL: MMPBSA in case of highly water accessible molecules ?

 Sent to CCL by: "Marek Maly" [marek.maly[A]ujep.cz]
  Dear all,
  I am just curious if the commonly used and for example in Amber MM/PBSA
implementation highly recommended value 1 for dielectric constant of the solut (in Amber it is parameter INDI) is ideal also in case of MM/PBSA calculations of free energy of binding 
  of two molecules which interior is pretty accessible to water molecules.
  I am asking because I have experienced some small positive values of dG
  (4.91, 2.48) in case of nicely complexated singlestrand
  DNA and small dendrimer  which interior is
  almost fully accessible for water molecules.
  Although dH was negative in agreement with evident complexation,
  positive entropic penalty -TdS more than compensated
  enthalpic gain.
  I tried to increase solute diel. constant (INDI = 2,3 - 7) and I got
nice improvement (more negative values dH) with saturation of dH around INDI = 6. 
  But of course I would like to prevent me from the experiments which might
be in contradiction with theory. In this case would be for me better to stay with INDI=1 and explain small positive values of dG as the result of inaccuracy of the methodology (especially in case of normal mode 
  analysis)
  I have to say that I tried both, single and three trajectory approach
  with  the similar behaviour.
  I think that my statistic is pretty good (analyses (including entropy) of
  400 frames from the last 4ns of 14ns simulation ).
  Thanks in advance for any helpful information/advice !
       Best wishes,
             Marek
 --
Tato zpráva byla vytvořena převratným poštovním klientem Opery: http://www.opera.com/mail/