[gmx-developers] Re: g_chi 3.2

[Graham Smith]

Hi - 

I've added another output option to g_chi, -cs chiprodentro.xvg; 
this uses the probabilities of rotomer occupancies p
to calculate a configurational entropy for a sidechain 
S = -R sum p ln p, and an effective number of rotamers visited 
n_eff = exp(S/R). The xvg file contains 
i, S, n_eff, n_eff/n_rot 
got the idea from
                                              
Shenkin PS, Farid H, Fetrow JS.
Pediction and evaluation of side-chain conformations for
protein backbone structures.
Proteins. 1996 Nov;26(3):323-52.

I know you're going to put out 3.2.1 soon, and I hope you'll be able
to include this bit of code. Therefore I haven't uploaded it to
contributions, but you can get it from
www.bmm.icnet.uk/~smithgr/newchi_3.2.tar.gz
(contains 
gromacs-3.2/src/tools/anadih.c
gromacs-3.2/src/tools/gmx_chi.c
gromacs-3.2/src/tools/pp2shift.h
gromacs-3.2/include/gstat.h
)

Cheers 

Graham