ADL: How to add distance constraints to Autodock?

mswingle at jaguar1.usouthal.edu mswingle at jaguar1.usouthal.edu
Wed Oct 31 13:55:36 PDT 2007


Hi Tony,

AFAIK, there is no simple way to do something like this. Distance constraints WITHIN a ligand are 
straightforward (you just add a CONSTRAIN record to the ligand's coordinate file) but constraints between 
ligand and receptor are another matter. This is because autodock uses pre-calculated interaction maps (i.e. 
only autogrid actually cares about what's in your receptor coordinate file, autodock just uses autogrid's 
maps). In autodock3 there was a way to specify a covalent attachment point and use a special "covalent" 
atom type in the ligand. This is a very poorly documented feature. I've played around with it a bit without 
acceptable results (probably because I wasn't setting the parameters properly). Autodock4 has probably 
retained this feature so you might give it a try.

Regards,

Mark

----- Original Message -----
From: tony at mrc-lmb.cam.ac.uk
Date: Wednesday, October 31, 2007 1:11 pm
Subject: ADL: How to add distance constraints to Autodock?
To: autodock at scripps.edu


> Dear All,
> 
> 
> I'm trying to dock a chemically modified peptide inhibitor (165 atoms) 
> to
> the active site of an enzyme. The exact location of the binding site is
> known. In addition, due to a common mode of binding observed in other
> structurally unrelated inhibitors, I can anticipate not only the relative
> position of the enzyme and the inhibitor but also the possible
> interactions between both molecules. Thus, a set of atoms of my inhibitor
> should to positioned within a certain distance to a set of atoms of the
> enzyme molecule so that they can form hydrogen bonds.
> 
> 
> I've performed several runs using Autodock 4.0 with a grid centered in 
> the
> binding site but none of the resulting docked conformations satisfied 
> the
> above criteria.
> 
> 
> Is there any way to apply distance constraints in Autodock 4.0?
> Simething like: Consider only conformations where atom A from the 
> ligand is
> within distance B from the atom C in the receptor molecule.
> 
> 
> Alternatively, could you suggest me any other approach to perform flexible
> docking of two molecules using user defined distance constraints.
> 
> Any help will be highly appreciated.
> 
> Thank you in advance.
> 
> Tony
> 
> PS.A rigid body docking is not helpful since at least the ligand has to
> undergo a conformational change.
> 
> ________________________________________________
> --- ADL: AutoDock List  ---  ---


More information about the autodock mailing list