[Dock-fans] Double bonds

James Foster (PG Research) j.r.foster at dundee.ac.uk
Sat Dec 17 14:09:05 PST 2016


Thank you Brian, this worked an absolute treat. Hope you have a great Christmas!


James

________________________________
From: Brian Fochtman <brianfochtman at gmail.com>
Sent: 17 December 2016 21:39:46
To: James Foster (PG Research); dock-fans at docking.org
Subject: Re: [Dock-fans] Double bonds

Hey,
I was able to recreate the behavior you are seeing when docking your ligand into an arbitrary structure I had available.  Sorry I sent you on a goose chase - the script I sent had a bug.  I've included an updated version of the script and confirmed that after docking the cis-double bonds are preserved.
   - Brian

On Sat, Dec 17, 2016 at 3:51 PM, James Foster (PG Research) <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:

Ah yes, that's a good point. As you can probably tell I'm new to docking..


I'm trying to see if its possible to dock 2-arachidonoylglycerol against the recent CB2 receptor crystal structure. I know a c-shaped binding mode was made by the authors of the paper, but I'd like to see if dock can make a similar prediction.


James


________________________________
From: Brian Fochtman <brianfochtman at gmail.com<mailto:brianfochtman at gmail.com>>
Sent: 17 December 2016 20:02:23

To: James Foster (PG Research)
Subject: Re: [Dock-fans] Double bonds

James,
    No problem.  A couple things.
1. Unrelated to the double bond issue is that you don't have partial charges on your small molecule and therefore are not considering electrostatic interactions - is this by design?
2. I'm really confused by this double bond issue.  DOCK has no way of removing atoms or bonds so I'm unclear of what is happening here.  It makes sense that if you post process the anchor_and_grow.mol2 file you sent you would get the final_output_scored.mol2 file you sent.  If you want to send over your protein structure I can rerun your calculation and see if I get the same issue.
-Brian

On Sat, Dec 17, 2016 at 12:34 PM, James Foster (PG Research) <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:

Hi Brian,

Thank you for sticking with me on this issue. I think I understand the rationale behind what we're trying to do. On the temp_for_docking.mol2 file I can see the fake rings have been added and the double bonds are all cis as they should be. But when I run the anchor and grow script the output file (anchor_and_grow_scored.mol2) has trans bonds and the fake rings have gone. Then when I run the python script using the "anchor_and_grow_scored.mol2" file the fake rings have been added again?


Please find attached pictures of the same double bond at each step of the process, and my 'anchor_and grow_scored.mol2 file.


Thanks,

James

________________________________
From: Brian Fochtman <brianfochtman at gmail.com<mailto:brianfochtman at gmail.com>>
Sent: 17 December 2016 14:34:54

To: James Foster (PG Research)
Subject: Re: [Dock-fans] Double bonds

Hey James,
What the script does is draw fake rings about double bonds, as dock doesn't sample torsion angles in rings.  This results in docked molecules with with same torsion angles around double bonds that you started with.

temp.for.docking.mol2 had these rings which is exactly what we want.

anchor_and_grow_scored.mol2 should also have these rings.

When running
python 2014.05...cyclize_db.py anchor_and_grow_scored.mol2  final_output_scored.mol2 we should have a final...mol2 file without rings but the final...mol2 file you sent is cyclized. (The same script removes these artificial rings when it finds them.)

My guess is that your input for the second python step isn't using the newly generated anchor_and_grow_scored.mol2.

To test this we need to take a look at anchor_and_grow_scored.mol2.

If this ends up being right I'd like to run the docking myself.

Hope this helps.
-Brian

On Dec 17, 2016 5:02 AM, "James Foster (PG Research)" <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:
Sorry Brian, that was from a different dock. Please find attached the right file..
________________________________
From: Brian Fochtman <brianfochtman at gmail.com<mailto:brianfochtman at gmail.com>>
Sent: 17 December 2016 00:45:33
To: James Foster (PG Research)
Subject: Re: [Dock-fans] Double bonds

I think the issue is that you didn't set  "ligand_atom_file" as "temp_for_docking.mol2" in your dock.in<http://dock.in> file.

>From dock.out:

Molecule Library Input Parameters
------------------------------------------------------------------------------------------
ligand_atom_file                                             /Users/jdogg/Desktop/gloriamg2a/linoacid.mol2
limit_max_ligands                                            no



On Fri, Dec 16, 2016 at 7:28 PM, James Foster (PG Research) <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:

Oops sorry, is this the right one?

________________________________
From: Brian Fochtman <brianfochtman at gmail.com<mailto:brianfochtman at gmail.com>>
Sent: 17 December 2016 00:04:43

To: James Foster (PG Research)
Subject: Re: [Dock-fans] Double bonds

Sorry should have been more clear. The non mol2 output from dock.  Example if you run dock like
dock -i dock.in<http://dock.in> -o dock.out

I would like to see the dock.out file.

On Fri, Dec 16, 2016 at 7:01 PM, James Foster (PG Research) <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:
Please see attached :)
________________________________
From: Brian Fochtman <brianfochtman at gmail.com<mailto:brianfochtman at gmail.com>>
Sent: 16 December 2016 23:48:03
To: James Foster (PG Research)
Subject: Re: [Dock-fans] Double bonds

Sorry, also your dock.out file.

On Fri, Dec 16, 2016 at 6:26 PM, James Foster (PG Research) <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:

Sure, please find attached..

________________________________
From: Brian Fochtman <brianfochtman at gmail.com<mailto:brianfochtman at gmail.com>>
Sent: 16 December 2016 22:19:36

To: James Foster (PG Research)
Cc: dock-fans at docking.org<mailto:dock-fans at docking.org>
Subject: Re: [Dock-fans] Double bonds

Can you send your mol2 files?

On Fri, Dec 16, 2016 at 3:56 PM, James Foster (PG Research) <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:

Hi Brian,

 Thanks very much for the reply. I tried preparing the molecule with your script, running a flexible docking algorithm and then post processing the output file your script but the double bonds still come out as trans instead of remaining cis? Maybe it's not right to use a flexible docking algorithm on a molecule containing a long alkyl chain with double bonds?


Thanks again,

James

________________________________
From: Brian Fochtman <brianfochtman at gmail.com<mailto:brianfochtman at gmail.com>>
Sent: 16 December 2016 03:18:41
To: James Foster (PG Research)
Cc: dock-fans at docking.org<mailto:dock-fans at docking.org>
Subject: Re: [Dock-fans] Double bonds

Hi Jim,
     There are a couple ways to to change this behavior within DOCK.  The method I prefer is as follows:

     I've attached a script that should be used to pre and post process your molecules.

      1.   python     2014.05.16_cyclize_db.py<http://2014.05.16_cyclize_db.py>      input.mol2      temp_for_docking.mol2
      2.   run dock as you normally would except specify "temp_for_docking.mol2" from above as the ligand_atom_file
      3.   python      2014.05.16_cyclize_db.py<http://2014.05.16_cyclize_db.py>      output_scored.mol2       final_output_scored.mol2


     What this process does is lock all double bonds in the molecule in the original conformations.  Running the script the second time cleans up the output file.  One disadvantage of this process is that RMSDs have to be calculated after the face on final_output_scored.mol2.    Let me know if you have any questions or need help getting the script running.

-Brian


On Thu, Dec 15, 2016 at 6:52 PM, James Foster (PG Research) <j.r.foster at dundee.ac.uk<mailto:j.r.foster at dundee.ac.uk>> wrote:

Hello,

I've tried using rigid and flexible docking to dock a molecule containing a cis c-c double bond. After running the dock the conformation of the bond turns into trans? Please do you know of any solution to correct this?
Thanks,
Jim

The University of Dundee is a registered Scottish Charity, No: SC015096

_______________________________________________
Dock-fans mailing list
Dock-fans at docking.org<mailto:Dock-fans at docking.org>
http://mailman.docking.org/mailman/listinfo/dock-fans



The University of Dundee is a registered Scottish Charity, No: SC015096


The University of Dundee is a registered Scottish Charity, No: SC015096


The University of Dundee is a registered Scottish Charity, No: SC015096


The University of Dundee is a registered Scottish Charity, No: SC015096


The University of Dundee is a registered Scottish Charity, No: SC015096

The University of Dundee is a registered Scottish Charity, No: SC015096


The University of Dundee is a registered Scottish Charity, No: SC015096


The University of Dundee is a registered Scottish Charity, No: SC015096
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.docking.org/pipermail/dock-fans/attachments/20161217/a36f2dd6/attachment-0001.html 


More information about the Dock-fans mailing list