[gmx-users] Different g_sas values between AMBER, CHARMM and GROMOS ff for DPC

[David van der Spoel]

On 2011-04-09 19.06, sa wrote:
> Dear All,
>
> I have simulated three DPC micelles with the same size (54 lipids) with
> different force fields (CHARMM, AMBER et GROMOS53A6) and computed the
> average accessible surface areas for each lipids with g_sas (gmx4.5.3) I
> obtain the three average values for total DPC, the headgroup
> (phosphocholine) and the alkyl tail.
>
>            Total (A2)        Head          Tail
>
> GROMOS53A6    158.4 ± 4.8     54.8 ± 3.4         103.2 ± 3.8
> AMBER              248.2 ± 8.7     76.2 ± 5.1         172.5 ± 5.2
> CHARMM           242.7 ± 7.7     70.4 ± 4.1          172.4 ± 4.8
>
> As you can see the ASA values are close between the micelles simulated
> with the CHARMM and AMBER ff, but larger than the values obtained from
> the micelle simulated with GROMOS ff especially for the headgroup. I am
> aware that  CHARMM and AMBER are explicits ff whereas GROMOS is an
> united ff  It is the reason I obtain a smaller values for GROMOS
> compared to two others ones ?
>
>
> Does g_sas tool take into account this. I have also noted when I use
> g_sas i obtain the following message
>
> WARNING: masses and atomic (Van der Waals) radii will be determined
>           based on residue and atom names. These numbers can deviate
>           from the correct mass and radius of the atom type.
>
> WARNING: could not find a Van der Waals radius for 54 atoms
>
> It is important ?

Given your observations above, don't you think it might be?
In order for g_sas to work with gromos force field one would have to 
increase the radius of the carbon atoms. For the 54 atoms the default 
value will be used.
>
> Thank you in advance for your advices.
>
> SA
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se