[Dock-fans] Fwd: grid size (former "parameters" and "very large receptors")

Trent E. Balius tbalius at aol.com
Fri Jan 21 06:45:15 PST 2011


I would try test #3 from my earlier email.  When you have spheres in multiple pockets dock may try and orient the ligand to spheres located in different pockets causing the ligand to be placed in a vary unfavorable position and pruned leaving you with no orients.  


As a test:

Use the program showsphere to create a pdb of the spheres.  

http://dock.compbio.ucsf.edu/DOCK_6/dock6_manual.htm#Showsphere 

Visualize the big sphere set with your protein with a program like chimera. Choose the a reduced set in one pocket and try docking.  Do, say, 5 more iterations where you increase sphere to increase coverage of the protein.  

I hope this helps,

 

Trent E. Balius
Graduate Student, Rizzo Group,
Department of Applied Mathematics and Statistics,
Stony Brook University.
Office: Math Tower 3-129, Phone: (631) 632-8519
URL: http://www.ams.sunysb.edu/~tbalius/

 

 

-----Original Message-----
From: Francesco Pietra <chiendarret at gmail.com>
To: Trent E. Balius <tbalius at aol.com>
Sent: Thu, Jan 20, 2011 4:54 pm
Subject: Re: [Dock-fans] Fwd: grid size (former "parameters" and "very large receptors")


I have now tried anchor_and_grow for the same ligand (30 heavy atoms)





with an even coarser grid (0.6) made of only 8,726,355 points. Out





file:











Initializing Library File Routines...





Initializing Orienting Routines...





Read in 3526 spheres for orienting.





Initializing Conformer Generator Routines...





Initializing Grid Score Routines...





 Reading the energy grid from





/home/francesco/work_hasic1a_mod21-model/mod21_reduce/grid.whole-corrected.0.6/grid.nrg





 Done reading the energy grid.











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





VERBOSE MOLECULE STATS











  Number of heavy atoms = 30





  Number of rotatable bonds = 1





  Formal Charge = -0.00





  Molecular Weight = 397.56





  Heavy Atoms = 30





Orienting 28 anchor heavy atom centers





********************











Then I tried anchor_and_grow for methane, same receptor, same grid





above, same computer with same settings. Out file:











VERBOSE MOLECULE STATS











  Number of heavy atoms = 1





  Number of rotatable bonds = 0





  Formal Charge = -0.00





  Molecular Weight = 16.04





  Heavy Atoms = 1





Orienting 1 anchor heavy atom centers





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





VERBOSE GROWTH STATS : ANCHOR #1











60/100 anchor orients retained (max 100) t=0.1s

















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





Molecule: methane.pdb











 Elapsed time for docking:  4 seconds











 Anchors:       1





 Orientations:      395





 Conformations:     60











    Grid Score:           -4.239931





      Grid_vdw:           -4.234622





       Grid_es:           -0.005309





    Int_energy:            0.000000

















1 Molecules Processed





Total elapsed time: 5 seconds





*********************











The orienting code is the culprit.











For curiosity I examined autodock4: to create a box for the whole my





receptor, the grid space needs to be as large as 1A. I did not try if





docking worked for my 30-heavy-atoms molecule.











Well, any comment would be appreciated











francesco pietra





























On Tue, Jan 18, 2011 at 2:52 PM, Trent E. Balius <tbalius at aol.com> wrote:





> Hi Francisco,





>





> As I am interested to work with the whole energy of the receptor





>





> Even if the grid box does not contain the whole receptor the energy stored





> at each grid point is calculated from the entire receptor.





>





> That being said I understand that you want to dock all over receptor





> surface.





>





> I guess you believe that this is a problem with the orienting code?





>





> It looks like you are docking a rigid molecule.





>





> Possible tests:





>





> (1) Perform a single point calculation on the grid to demonstrate that it is





> not a grid energy problem.





>





> (2) Preform only orienting (no minimization and no growth, make sure that





> your pruning cut off is vary high).  visualize the orients with receptor and





> spheres.





>





> (3) Dock with reduced sphere sets, all spheres located in the same pocket.





> Gradually increase the sphere set to include more spheres within other





> pockets about the protein.





>





> Let us of know of your progress with is problem.  Blind docking is a useful





> application of DOCK.





>





> See the following for previous blind docking discussion:





>





> http://mailman.docking.org/pipermail/dock-fans/2010-July/002398.html





>





> I hope that these test suggestions help.





>





> Trent E. Balius





> Graduate Student, Rizzo Group,





> Department of Applied Mathematics and Statistics,





> Stony Brook University.





> Office: Math Tower 3-129, Phone: (631) 632-8519





> URL: http://www.ams.sunysb.edu/~tbalius/





>





>





> -----Original Message-----





> From: Francesco Pietra <chiendarret at gmail.com>





> To: dock-fans <dock-fans at docking.org>





> Sent: Tue, Jan 18, 2011 5:32 am





> Subject: [Dock-fans] Fwd: grid size (former "parameters" and "very large





> receptors")





>





> Forgot to add that I have also put in my .bashrc the statement





>





>





>





>





>





> ulimit -s unlimited





>





>





>





>





>





> (Debian GNU/LINUX amd64)





>





>





>





>





>





>





>





>





> ---------- Forwarded message ----------





>





>





> From: Francesco Pietra <chiendarret at gmail.com>





>





>





> Date: Tue, Jan 18, 2011 at 8:53 AM





>





>





> Subject: grid size (former "parameters" and "very large receptors")





>





>





> To: dock-fans <dock-fans at docking.org>





>





>





>





>





>





>





>





>





> Hello:





>





>





> I am combining here my two former threads "vdw parameters" and "very





>





>





> large receptors".





>





>





>





>





>





> That about the vdw parameters was a wrong message by dock6.4.





>





>





> Actually. it was the format. To fit the MW code, the correct format is





>





>





> (i.e., C.3, H, etc). The one I got from dockprep in latest version of





>





>





> chimera was no good (c3, hc ...)





>





>





>





>





>





> @<TRIPOS>ATOM





>





>





>      1 C           2.5960   -1.2050    1.7710 C.3       1 A31   -0.0961





>





>





>      2 H           2.6980   -1.2650    0.6640 H         1 A31    0.0407





>





>





>





>





>





>





>





>





> That corrected, the problem is the grid size. This is a portion of





>





>





> anchor_and_grow.out:





>





>





>





>





>





> Initializing Library File Routines...





>





>





> Initializing Orienting Routines...





>





>





> Read in 3433 spheres for orienting.





>





>





> Initializing Conformer Generator Routines...





>





>





> Initializing Grid Score Routines...





>





>





>  Reading the energy grid from





>





>





> /home/francesco/work_hasic1a_mod21-model/mod21_reduce/grid.whole.0.5/grid.nrg





>





>





>  Done reading the energy grid.





>





>





>





>





>





> -----------------------------------





>





>





> VERBOSE MOLECULE STATS





>





>





>





>





>





>  Number of heavy atoms = 30





>





>





>  Number of rotatable bonds = 1





>





>





>  Formal Charge = -0.00





>





>





> Molecular Weight = 383.56





>





>





>  Heavy Atoms = 30





>





>





> Orienting 28 anchor heavy atom centers





>





>





>





>





>





>





>





>





> At this point it crashes, the screen reporting "segmentation fault",





>





>





> which per se tells nothing useful. However, contrary to previous





>





>





> erroneous messages, it seems to me that it is not a problem of out of





>





>





> memory. I tried also on a shared-memory machine with a lot of RAM,





>





>





> getting exactly the same error message.





>





>





>





>





>





> It seems to me that dock6.4 is unable to deal with 3433 spheres (which





>





>





> was to cover the whole receptor at grid space 0.5). I tried now





>





>





> anchor_and_grow with a smaller system investigated in the past with





>





>





> previous versions of dock and it was finely completed by my present





>





>





> installation of dock6.4, even on a machine with only 8GB RAM.





>





>





>





>





>





> As I am interested to work with the whole energy of the receptor, and





>





>





> I can't use dock3.5 for the reasons that John pointed out, I would be





>





>





> happy to have the response of dock6.4 developers, if they agree or not





>





>





> with my above conclusions. As always, I am prone to carry out tests on





>





>





> their suggestions. This is not to limit interventions by





>





>





> non-developers may have the right ideas about.





>





>





>





>





>





> Thanks





>





>





> francesco pietra





>





>





> _______________________________________________





>





>





> Dock-fans mailing list





>





>





> Dock-fans at mailman.docking.org





>





>





> http://mailman.docking.org/mailman/listinfo/dock-fans





>





>





>






 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.docking.org/pipermail/dock-fans/attachments/20110121/2b605c43/attachment.html 


More information about the Dock-fans mailing list