[Dock-fans] Amber charges and GB/SA docking

david.evans at ulsop.ac.uk david.evans at ulsop.ac.uk
Tue Feb 22 04:38:20 PST 2005

I am starting to use Dock 5.1.1 to investigate DNA-ligand binding.
I have the program up and running, but have a few questions about the
scoring function parameters.
1) It states in the Dock4 manual (page 22) that you should use
KOLLMAN all-atom
charges in Sybyl when constructing the grid. In my Sybyl6.9, these
charges are not the
same as the best-known Cornell et al Amber94 charges, which you can
get from Sybyl by
using the Amber95dna dictionary. I'm thinking that I should go with
Amber94, but what are the
bulk of published works with DOCK based on, or the opinions of
current DOCK experts?
2) I've referred a lot to a paper by Kang et al (Biopolymers 73, pg
192), which used the GB/SA
scoring function to rank minor-groove binding ligands, with
reasonable success.
My initial, very preliminary, runs (on a Zinc database) seems to rank
ligands 'better' with the
plain dock energy function. By 'better' I mean molecules which lie
substantially inside the groove
are ranked higher than those only partially attached to the DNA.
I know many things might be happening, but specifically I'd like to
try using the refit GB/SA
scoring functions of the above paper (Table II). Can someone tell me
how to modify the Dock code
to do this? Line  621  in score_d4.cpp seems apposite, but I don't
quite understand the relation
between these variables and the equations as written in the paper.
Thanks for your attention,

-------------- next part --------------
A non-text attachment was scrubbed...
Name: david_evans.vcf
Type: text/directory
Size: 175 bytes
Desc: not available
Url : http://blur.compbio.ucsf.edu/pipermail/dock-fans/attachments/20050222/8ce51b02/david_evans.bin

More information about the Dock-fans mailing list