ADL: Is it possible to correct for high torsional free

David M. Carter david.carter3 at mail.mcgill.ca
Thu May 29 15:16:08 PDT 2008


Hi,

It is my understanding that the torsional free energy term is a constant and
depends on how many torsions are set active.  I am trying the docking
problem you describe myself and have had limited success.  I have a 12-mer
peptide and originally set 32 torsions active with no luck, just positive
binding energies.  Then I reduced the torsions down to a value of around 14
or so and cranked up the number of energy evaluations to 25,000,000.  Only
then did I start seeing negative binding energies (~ -2.5 kcal/mol) after
about 10 days of computations.  Even so, I am now realizing that for me to
obtain convergence of docking solutions I will need to do around 250,000,000
energy evaluations which will take our fastest quad processor machine about
one month to complete.  As far as I know this will go much faster on a
cluster of processors, but have yet to take that step.  Sounds like you need
to increase the number of evaluations you are doing.

Hope this helps,

David

-----Original Message-----
From: autodock-bounces at scripps.edu [mailto:autodock-bounces at scripps.edu] On
Behalf Of autodock-request at scripps.edu
Sent: May-29-08 3:00 PM
To: autodock at scripps.edu
Subject: autodock Digest, Vol 50, Issue 24

Send autodock mailing list submissions to
	autodock at scripps.edu

To subscribe or unsubscribe via the World Wide Web, visit
	http://mgldev.scripps.edu/mailman/listinfo/autodock
or, via email, send a message with subject or body 'help' to
	autodock-request at scripps.edu

You can reach the person managing the list at
	autodock-owner at scripps.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of autodock digest..."


Today's Topics:

   1. Re:  ADT bug ? (Ruth Huey)
   2. Re:  autogrid4/autodock4 does not work!!! (anantha ramanan)
   3. Re:  small query (anantha ramanan)
   4.  Is it possible to correct for high torsional free energies
      when docking peptides? (giang at temple.edu)


----------------------------------------------------------------------

Message: 1
Date: Wed, 28 May 2008 17:37:52 -0700
From: Ruth Huey <rhuey at scripps.edu>
Subject: Re: ADL: ADT bug ?
To: autodock at scripps.edu
Message-ID: <483DFAE0.7070308 at scripps.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi,
(1) Only the side chains of the designated residues are allowed to be 
flexible. Thus for a flexible Thr residue, at most 4 atoms would be 
written in the flexible file.
(2) The receptor file for prepare_flexreceptor4.py needs to be in 
".pdbqt" format.  You can use ADT or prepare_receptor4.py to  convert 
your pdb file to pdbqt.

         Ruth


  
Amor A. San Juan wrote:
> Hi,
>
> I noticed that ADT in PMV parse residue atoms when implemented in the
> recepter preparation. For example, only few atoms of residue Thr were in
> the *_flex.pdbqt output. This is due to ADT bug I presume. In effect,
> input flex residues are not sensible and so the output from dlg as well.
>
> I try to use manually the ADT script prepare_flexreceptor4.py in order to
> prepare the protein flex residues. But error flags showed up:
>
> Hope anybody can give suggestion/help.
>
> Thanks,
> Amor
>
> [biiamor at hilbert flex_test]$ pythonsh
>
/HOME01/BII/biiamor/MGLTools-1.4.6/MGLToolsPckgs/AutoDockTools/Utilities24/p
repare_flexreceptor4.py
> -r 1IPB.pdb -s THR205 -x test.pdb
> setting PYTHONHOME environment
> Traceback (most recent call last):
>   File
>
"/HOME01/BII/biiamor/MGLTools-1.4.6/MGLToolsPckgs/AutoDockTools/Utilities24/
prepare_flexreceptor4.py",
> line 127, in ?
>     non_rotatable_bonds=all_bnds)
>   File
>
"/HOME01/BII/biiamor/MGLTools-1.4.6/MGLToolsPckgs/AutoDockTools/MoleculePrep
aration.py",
> line 2201, in __init__
>     self.write_flex(flex_residues, flexres_filename)
>   File
>
"/HOME01/BII/biiamor/MGLTools-1.4.6/MGLToolsPckgs/AutoDockTools/MoleculePrep
aration.py",
> line 2241, in write_flex
>     self.writeResidue(res, outfileptr)
>   File
>
"/HOME01/BII/biiamor/MGLTools-1.4.6/MGLToolsPckgs/AutoDockTools/MoleculePrep
aration.py",
> line 2327, in writeResidue
>     self.writeSubtree(outfileptr, at, at2)
>   File
>
"/HOME01/BII/biiamor/MGLTools-1.4.6/MGLToolsPckgs/AutoDockTools/MoleculePrep
aration.py",
> line 2344, in writeSubtree
>     charge = startAtom.charge
>   File
> "/HOME01/BII/biiamor/MGLTools-1.4.6/MGLToolsPckgs/MolKit/molecule.py",
> line 385, in __getattr__
>     return self._charges[self.chargeSet]
> KeyError
>
>
> ________________________________________________
> --- ADL: AutoDock List  --- http://autodock.scripps.edu/mailing_list ---
>
>   



------------------------------

Message: 2
Date: Wed, 28 May 2008 21:58:19 -0700
From: "anantha ramanan" <ananthu.bioinfo at gmail.com>
Subject: Re: ADL: autogrid4/autodock4 does not work!!!
To: autodock at scripps.edu
Message-ID:
	<3ec817200805282158j7af200abj61ab2e41562f3899 at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

Do not keep any molecule files inside the usr/local/bin/.  ADT will ask u to
save the files in default place whenever u are trying to save the files in
the format of  PDBQT, GPF etc... , Save all the files into this default
drive  and run ur program hopefully it ll work.


-- 
Best Regards
Anantha ramanan.R


------------------------------

Message: 3
Date: Wed, 28 May 2008 22:01:02 -0700
From: "anantha ramanan" <ananthu.bioinfo at gmail.com>
Subject: Re: ADL: small query
To: autodock at scripps.edu
Message-ID:
	<3ec817200805282201o7ded28cfmf1f89f6c22a9e91d at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

Do not keep any molecule files inside the usr/local/bin/.  ADT will ask u to
save the files in default place whenever u are trying to save the files in
the format of  PDBQT, GPF etc... , Save all the files into this default
drive  and run ur program hopefully it ll work.



-- 
Best Regards
Anantha ramanan.R


------------------------------

Message: 4
Date: Thu, 29 May 2008 12:03:01 -0400 (EDT)
From: <giang at temple.edu>
Subject: ADL: Is it possible to correct for high torsional free
	energies when docking peptides?
To: autodock at scripps.edu
Message-ID: <20080529120301.CIZ73676 at po-a.temple.edu>
Content-Type: text/plain; charset=us-ascii

We are trying to dock a peptide (37 amino acids) and have
limited the number of rotatable bonds to 32.  Unfortunately,
all the dockings have very high positive binding energies (>20
kcal/mol), and so no Ki can be calculated. 

It seems that the torsional free energy swamps out the
negative free energies gained by binding. 

Does anyone have advice about how to get around this problem?

Please help.  Thanks.

Kathy Giangiacomo


------------------------------

________________________________________________
--- ADL: AutoDock List  ---
http://www.scripps.edu/pub/olson-web/doc/autodock/ ---

End of autodock Digest, Vol 50, Issue 24
****************************************



More information about the autodock mailing list