Article ID: 883 - Last Modified: December 4, 2010
I recently regenerated the Glide grid for a target, using a different ligand length. With the new grid, my old docking pose was not found. Why is this?
In Suite 2009 Update 2, the force field used to generate the grid was changed from OPLS_2001 to OPLS_2005. If you generated the grid with the new force field, this could influence the results. Changes in grid parameters such as grid center or size can also influence the results. Usually, these variations are significant only for poor binders, where there isn't a strong basin of attraction. Glide docking involves complex numerical algorithms, so there will always be some dependence on initial conditions. However, if the target hasn't changed, altering the grid size shouldn't change the grid potentials much, if at all, because the grid is generated from the entire target, not just the part inside the grid box.
To check whether the problem is the ranking of the poses, you can try generating multiple poses per ligand, and comparing the poses found. Or you could try running a score-in-place or refine job on the older pose to see how its GlideScore compares with the scores of the newer poses.
Keywords: Grid size, docking pose
Type the words or phrases on which you would like to search, or click here to view a list of all
Knowledge Base articles