ADL: Docking robustness

Stefano Forli forli at scripps.edu
Tue Dec 9 17:23:00 PST 2014


Hi Dariush,

you may want to take a look at this page [1] for a general discussion about variations 
among docking results.
As a rule-of-thumb, inconsistencies across docking run imply lack of convergence, 
therefore very likely the search is not accurate enough to sample the energy landscape of 
your problem.

To overcome this issue you can either reduce the complexity of the search (e.g., reduce 
docking box dimensions or split it into multiple boxes; rigidify parts of the ligand) or 
increase the search length (e.g., increase exhaustiveness).

About random seeds, they are set automatically by the docking programs, and the only 
reason to manually set one would be to exactly reproduce a particular docking job.

Hope this helps,

S



[1] 
http://autodock.scripps.edu/faqs-help/faq/why-do-the-results-differ-when-multiple-dockings-are-done-with-the-same-input?searchterm=calculation

On 12/09/2014 02:50 PM, Dariush Mohammadyani wrote:
> Thank you Amr!
> The issue is lack of robustness. If you use same ligand and receptor and
> run several dockings using different seed, your results would be different
> in each run.
> Then, how you can find out the main binding pose or study binding energies?
> Specifically, in a case of comparing different ligands vs. the same
> receptor.
>
>
> Regards,
> Dariush
>
> On Tue, Dec 9, 2014 at 5:01 PM, Amr AL-HOSSARY <amr_alhossary at hotmail.com>
> wrote:
>>
>> Vina is supposed to do a stochastic search using both Monte Carlo and BFGS
>> methods.
>> The seed is used to control the randomness by giving a fixed start point if
>> you need.
>>
>> I don't see any "issue" there to mitigate.
>>
>> Amr
>>
>>
>> -----Original Message-----
>> From: autodock-bounces at scripps.edu [mailto:autodock-bounces at scripps.edu]
>> On
>> Behalf Of Dariush Mohammadyani
>> Sent: Tuesday, December 09, 2014 5:28 AM
>> To: autodock at scripps.edu
>> Subject: ADL: Docking robustness
>>
>> Hello All,
>>
>> I am wondering to know why changing the seed (using -seed option) results
>> in
>> significantly different results.
>> For instance, in one docking most of 9 models may represent a binding pose,
>> which can be different from the results of another docking using different
>> random seed number.
>>
>> How can I mitigate this issue?
>>
>> Thank you for answering!
>>
>> Regards,
>> Dariush
>> ________________________________________________
>> --- ADL: AutoDock List  --- http://autodock.scripps.edu/mailing_list ---
>> ________________________________________________
>> --- ADL: AutoDock List  --- http://autodock.scripps.edu/mailing_list ---
>>
> ________________________________________________
> --- ADL: AutoDock List  --- http://autodock.scripps.edu/mailing_list ---
>

-- 
  Stefano Forli, PhD

  Staff Scientist
  Molecular Graphics Laboratory
  Dept. of Integrative Structural
   and Computational Biology, MB-112F
  The Scripps Research Institute
  10550  North Torrey Pines Road
  La Jolla,  CA 92037-1000,  USA.

     tel: +1 (858)784-2055
     fax: +1 (858)784-2860
     email: forli at scripps.edu
     http://www.scripps.edu/~forli/


More information about the autodock mailing list