[Dock-fans] Unknown Close Contact Warning

Scott Brozell sbrozell at rci.rutgers.edu
Thu Sep 23 00:27:33 PDT 2010


On Tue, Sep 14, 2010 at 11:36:53AM -0400, Fuhrman, Kit wrote:
> I'm re-scoring my compounds with the Amber score.  Some of my ligands display an error in the amberize_ligand log
> "Warning: Close contact of 0.618328 angstroms between .R<<0> 1>.A<H24 37> and .R<<0> 1>.A<H28 44>"
> The problem is that when I measure the distance between the atoms, H24 and H28, the distance is not even close to .618328.  In this case it's 0.786114712.  This happens for quite a number of ligands.  When I view the mol2 in chimera or pymol, I can't find anything terribly wrong with the molecules.  If I ignore the errors, I get a reasonable amber score.
> I've attached the ligand.  Has anyone seen this before?  Any suggestions on how to zero in on the cause?

The close contact reported is actually the square of the distance:
sqrt(0.618328) = 0.786
At the very least the warning should be clearer, but this may be a
legacy issue within Amber.  I'll investigate.


More information about the Dock-fans mailing list