Le Tuesday 27 September 2011 à 10:09:10AM, Kasprzak, Wojciech (NIH/NCI) [C] a écrit :
> Dr. Walker commented that "GB will be terrible for highly polar systems such
> as DNA / RNA," and my experience seems to second that. On the other hand,
> GB protocol tutorial I read uses a small DNA fragment. Again, any enlightening
> comments would be appreciated (say, good for small nucleic acid structures,
> based on crystallography data, bad in other cases.)
Hi Wojciech,
In a recently published paper (
http://dx.doi.org/10.1021/ct200384r),
we have evaluated DNA simulations with different force fields (AMBER
ff99bsc0 and CHARMM all27) and different GB methods (AMBER GBhct, GBobc1,
GBobc2, GBn and CHARMM GBMV) from a structural point of view, with respect
to explicit solvation and experimental data.
Unfortunately, it appears that the AMBER ff99bsc0 force field and
pairwise GB methods implemented in AMBER have similar defects that add
up and lead to important structural deformations. This is particularly
visible when looking at the major groove width. CHARMM all27 + GBMV is
also not to recommend.
On the other hand, reasonable DNA modeling can be obtained by using the
CHARMM all27 force field in combination with AMBER GB methods (in
particular GBhct), thanks to defect compensation.
I would thus not be so pessimistic for DNA and recommend the all27+GBhct
(igb = 1) protocol (using CHARMM force fields in AMBER is possible
thanks to the chamber program), which gives average structural
parameters comparable to explicit solvation. However I don't know how
this is transferable to RNA.
Good luck,
--
Thomas Gaillard
Maître de conférences
Laboratoire de Biochimie
Ecole Polytechnique
91128 Palaiseau cedex
tel: +33 1 69 33 48 62
fax: +33 1 69 33 49 09
thomas.gaillard.polytechnique.edu
http://bioc.polytechnique.fr/~gaillard
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Sep 27 2011 - 11:00:03 PDT
