[Dock-fans] message in amber score

[Scott Brozell]

Hi,

On Mon, 28 Apr 2008, liane saiz wrote:

> I was running the dock6 amber scoring function when the following sign
> appeared in my screen:
> 
> select_atoms: atom-expr too complicated
> 
> the Amber parameters I used were as follow
> 
> Amber Score Parameters
> --------------------------------------------------------------------------------------
> amber_score_primary                                          yes
> amber_score_secondary                                        no
> amber_score_receptor_file_prefix
> Model_1_min_chimera
> amber_score_movable_region                                   distance
> amber_score_movable_distance_cutoff                          10
> receptor_site_file                               Model_1_min_chimera.sph
> amber_score_before_md_minimization_cycles                    100
> amber_score_md_steps                                         3000
> amber_score_after_md_minimization_cycles                     100
> amber_score_gb_model                                         5
> amber_score_nonbonded_cutoff                                 18.0
> amber_score_temperature                                      300.0
> 
> the calculations are still in progress but I don't know whether they are
> reliable since I've ever seen this message before. Could somebody kindly
> tell me more about this??

I have never encountered this message before.
Apparently, the size of the receptor in combination with the 10A
amber_score_movable_distance_cutoff lead to a generated
NAB atom expression that exceeds AEXPR_SIZE in src/dock/nab/select_atoms.c

Since NAB ignores its own error here, the calculation is probably not
correctly selecting all residues within 10A of the active site.

A patch would be to increase AEXPR_SIZE on line 9 and reinstall dock:
#define  AEXPR_SIZE  1000

Note that the -v option to dock will emit the generated
NAB atom expression.

> Previous calculations with amber_score_movable_region=ligand were
> succesfully completed with no strange messages.

Only amber_score_movable_region of distance or nab_atom_expression
could permit this message.

Scott