ADL: no inhibition constante (Ki)

abanuelos at cinvestav.mx abanuelos at cinvestav.mx
Tue Mar 4 10:33:18 PST 2008


Hello Tee & Miguel:

We had a similar problem running autodock 4 in cygwin, we obtained very
low binding energies (in the order of -3x10e7) and a Ki=0.0 , we thought
that was the size of the torsion matrix in a large grid, the problem was
solved downsizing the grid, maybe was a memory allocation problem , the
calculus worked fine in autodock 3 for cygwin and in autodock 4 for
Linux(we employ a fedora 8 distro), i suggest you trying lowering the grid
size and/or the number of active torsion in the ligand.

Best Regards

Angel Bañuelos

========================================================================
Hi ! Miguel
I see your problem. This happened because some components(atoms) of your
ligand
crashed with Macromolecule. You can extract file to display what happened
using 3D
program such as Weblab viewer. That why the binding energy was so far and
Ki didn't
appear. To solve this problem, try to redock by increasing number of run
(about 30
to 50 runs but It's sure that more time is consumed). Then, see the result
in your
out.dlg to see which one has resonable value of dock energy ( not too much
high or
being positive).. After that, extract the comformation corresponded to the
value
using egrep command or get-docked( from autodock package) and redocking
again. I
guess this time will give you reasonable result.

Best luck,
Tee Piroj


----- Original Message ----
From: Miguel Quiliano Meza <qilime at yahoo.es>
To: autodock at scripps.edu
Sent: Tuesday, March 4, 2008 1:00:19 AM
Subject: ADL: no inhibition constante (Ki)

Hi dockers.

I have been performing several dockings as a training. But in some cases
inhibition
constant term is not coming (in the out.dlg). Also I have read a same
question in
the list but no-one responds.

Also in some cases my binding energy  is  16700000 ..!!!!

so I would like to know  in which cases the Ki will come? and why not?

Thanks in advance.



More information about the autodock mailing list