[Chimera-users] Help in docking with Chimera

[Elaine Meng]

Hi Zahir,
Sorry for the delay, there were at least two different problems and it took a lot of time to investigate them. One problem is a display issue that can be mostly fixed with some extra steps in Chimera, but the other is a problem with how the AutoDock ligand prep script treats proline, for which I don't have a solution.

Problem (1): Autodock Vina changes the order of the atoms so that you get part of one residue and then part of another residue, etc. The residues are all scrambled up. That is why they look crazy.  You can fix the display without too much trouble, but only for showing atoms and bonds as stick or ball-and-stick.  The ribbons will still be confused and you cannot show the ligands as ribbons except with a LOT more work of editing the output to put the residues back together.  How to fix the atom display in Chimera is to hide ribbons, show all atoms, delete all bonds and then add reasonable bonds back.  So for example, if the docked ligands are model #5, you could use commands:
~ribb #5
display #5
~bond #5
sel #5
...then menu: Tools... Structure Editing... Build Structure, go to Adjust Bonds section, "Add" all reasonable bonds among the selected atoms  (sorry there is no command for that step).  Then you can clear selection.

I ran a test with a Gly-Pro-Gly peptide that I built, except after creating the PDB file I manually edited to be all one residue named UNK instead of GLY1,PRO2,GLY3.  This prevented the problem of the wrong bonds, but again you cannot show ribbon for this UNK molecule, so it is not worth the trouble to do that for your own peptide ligands.

Problem (2).  I have no solution for this, but the AutoDock ligand prep script that identifies rotatable bonds decides to allow the proline ring to break open.  That was another reason I did the test above, making a single residue UNK containing Gly-Pro-Gly atoms, but it didn’t help.  The ring still got broken.

A third issue is that your ligands have a very large number of rotatable bonds and the relatively small amount of sampling allowed by the web service is probably not adequate.  You may need to spend more time and get Autodock Vina running on your own system and learn about how to use it directly (without Chimera). Then you could explicitly control which bonds are rotatable (i.e. prevent the proline from breaking open) before running the docking step with much more sampling.  Of course that will take more time and effort than just running the web service with the Chimera interface.

I will also put this response in the bug ticket that you submitted.

Sorry for the problems… I hope this at least clarifies the situation.  Best,
Elaine

> On May 25, 2017, at 1:49 PM, Elaine Meng <meng at cgl.ucsf.edu> wrote:
> 
> 
> Hi Zahir,
> It might just be display (like one bond not shown) without any affect on the calculations because of the format of the file output by Autodock Vina. Autodock Vina is not our program, so we cannot control the output format.  However, impossible for us to tell if that is the issue unless you send us an example file or files. You could use  menu: Help… Report a Bug and include brief explanation and your email address and attach the example file or session that shows the problem.  Keep in mind it might not be a Chimera bug.
> Best,
> Elaine
> -----
> Elaine C. Meng, Ph.D.                       
> UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
> Department of Pharmaceutical Chemistry
> University of California, San Francisco
> 
>> On May 25, 2017, at 1:13 PM, Md Zahir Uddin <m_uddin at u.pacific.edu> wrote:
>> 
>> Hi,
>> We are seeing a new problem in the docking of our peptides. After docking of our peptides using Autodock Vina function in Chimera, we are seeing in the output that the peptides are broken (splitting into two parts) at the protein's binding site. Our peptides are 8 amino acid long (one is 5 amino acid cyclic peptide by disulfide bond) and we prepared the peptides using the Build Structure function in Chimera and then dockprep & minimize structure.
>> Can you please help me in this? We would really appreciate your help.
>> Thanks,
>> Zahir
>> 
> 
> 
> _______________________________________________
> Chimera-users mailing list: Chimera-users at cgl.ucsf.edu
> Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>