[gmx-users] a (possible?) typo error in the manual and a lot of g_sas quastions

[nicolis at guideo.fr]

Hello,
I try to use g_sas to compute solvent accessible surfaces.
I read in the man g_sas or with "g_sas -h" that the program computes 
hydrophobic and solvent accessible surfaces.

I see in the output options that I can use -o to get an area.xvg file, but am I 
correct supposing that the first column is hydrophobic and the second 
total area? If yes how is it possible for some frames to have a zero area 
(both columns)?

Another point: with the option "-q connelly.pdb" I generate a pdb file with 
the surfaces, but it contains all the surfaces, including water solvent 
molecules even if I use "-n index" to select only my molecule (a small 18 
atom drug). How to select only the drug and not the solvent?

Yet another point (the man error): in the man page, in the "g_sas -h" and 
on the html manual I see that both options -qmax and -minarea define the 
maximum charge of a hydrophobic atom. Obviously there is a typo error 
and -minarea defines something else but what?

Another question: is there a gromacs program allowing to compute mean 
area over the mdrun or do I just take the area.xvg file and directly compute 
the area from the values in it?

And another point concerning only the beta altivec version: while 
computing, g_sas displays a message like the following for every frame:

Reading frame       1 time    0.500   *** malloc[4123]: Deallocation of a 
pointer not malloced: 0x87b000; This could be a double free(), or free() 
called with the middle of an allocated block; Try setting environment 
variable MallocHelp to see tools to help debug

but it generates the files normally.

And a final question: is it possible to compute the area of a single .gro/.itp 
molecule without to perform an mdrun to obtain just one area value?

Thanks if you managed to read till here and many more thanks for any 
answers.

Yiannis