The Molecular Graphics Laboratory Forum
http://mgl.scripps.edu/forum/

Docking with flexible side chains
http://mgl.scripps.edu/forum/viewtopic.php?f=12&t=235
Page 1 of 1

Author:  oleg_trott [ Thu Jul 02, 2009 6:03 pm ]
Post subject:  Docking with flexible side chains

If you know how to do rigid receptor - flexible ligand docking, but do not understand how to make some side chains flexible in the receptor, read this explanation by Wim Nerinckx:

http://mgldev.scripps.edu/pipermail/autodock/2009-April/005741.html

Author:  wim nerinckx [ Wed Jul 15, 2009 11:46 am ]
Post subject:  Reflection on flexible residues

From the ADL mailing list as well as the new Vina forum, I gather that new dockers often are setting quite many receptor residues unnecessarily as flexible. Flexibility should be kept for only those few residues that really need it: too many flexibles hampers the prediction (in my experience). May I humbly advise new dockers to reflect more on your receptor (say, collect all known 3D structures and make a combined automatic overlap - e.g. the Swiss-PDB-viewer can do this - to see where the differences are, check the b-factors of suspect residues, consider the catalytic mechanism if your receptor is an enzyme, ...) before deciding which few residues must remain flexible. And even then you may consider to do test-dockings to possibly further eliminate some flexibles that only minimally affect ligand binding. Plus, the less flexibles, not only the better the prediction, but also quite faster the docking, which is a very important criterion if you are planning virtual screens with several 100.000 compounds.

Page 1 of 1 All times are UTC
Powered by phpBB® Forum Software © phpBB Group
https://www.phpbb.com/