[Chimera-users] How to to do that in chimera?

[Elaine Meng]

Hi Denis,
Whew, that's a lot of questions!  Bear in mind that we (Chimera developers) are not really users of the docking programs ourselves and have just used example files generated by others for most testing.

(1) Chimera reads the PDBQT format of Autodock and Autodock Vina output files.  We know it is possible for more than one docking position to be in a single file because the Vina output is like that.  For AutoDock, 
(i) see if there is an option of AutoDock to output all the positions in one file, and if not, or to handle the ones you already generated, 
(ii) concatenate the individual files into a single file and then open that single file in Chimera.  I just now tested that: it works fine, but you have to have a MODEL N (e.g. MODEL 1) line at the start of each ligand file and an ENDMDL line at the end.  Look in your Vina output file with a text editor for an example. The numbering isn't used, so it's OK if they all have MODEL 1... you don't have to worry about numbering them correctly.  You should look at your single-position files in a text editor to see if thay have these lines already.  If not, you you will need to add them.  If a file already has more than one position, however, it already has this MODEL and ENDMDL stuff, and you don't have to add it yourself.

(2) There is no tool for "hydrophobic interactions" per se, but there are two possibly related features:
(i) show receptor surface colored by hydrophobicity (for example, using the Presets menu) to see where the pocket is more or less hydrophobic.  You still have to know what parts of the ligand are more or less hydrophobic to be able to judge.
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/menu.html#menupresets>

(ii) use Find Contacts/Clashes to identify all contacts, but that will include both polar and nonpolar and favorable and unfavorable ... basically everything nearby.  You can specify by selection which atoms to include, however, and in that way only look at contacts between carbon atoms.
<http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/findclash/findclash.html>

Also, presumably the docking scores take hydrophobic interactions into account.

(3) see (1)(ii) above

(4) if different format (pdbqt vs mol2), no you can't concatenate those into a single file.  However, you can just start ViewDock more than once and list the ones from one program in one ViewDock dialog and the ones from the other program in another ViewDock dialog, and display any subset of the positions from each at the same time in the receptor site.

(5) You could just display all docked positions at the same time and look at the resulting "blob," or even calculate a (qualitative) density map from them using the command "molmap."
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/molmap.html>

 Other than that, you may want to use some other program outside of Chimera for clustering the docking results or doing some kind of probability calculation; I don't know enough to recommend anything specific, however.  Chimera does have an Ensemble Cluster tool that you could try, but only for different positions of exactly the same ligand (i.e. each ensemble member must have exactly the same atoms with the same names).  I've never tried it with docking output.
<http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/ensemblecluster/ensemblecluster.html>

I hope this helps,
Elaine
-----
Elaine C. Meng, Ph.D.                       
UCSF Chimera(X) team
Department of Pharmaceutical Chemistry
University of California, San Francisco

> On Apr 9, 2020, at 1:06 PM, Krivic, Denis <denis.krivic at medunigraz.at> wrote:
> 
> To whom it may concern,
> 
> I have just started working on a project which includes docking and analysis of the results several weeks ago. So figuring out software for docking and the docking procedure itself was one set of problems (solved), and now I could really use your help in order to analyze my data - docking results (output files).
> At the moment, I am using Windows 10 (x64), Chimera 1.14, AutoDock Vina 1.1.2 and AutoDock 4.2.6.
> 
> So, I have a protein (ion-channel) and several different ligands, for which I want to see how they and if they (at all) may interact with the protein. Therefore I dock the ligand in the two mentioned programs and got different output files. Vina generates a single file with 9 different positions, and 4.2. generates 9 different files for each position.
> 
> Now, the first problem I encountered was that in Chimera I can open and analyze only Vina output (single file). How do I open 4.2. output?
> 
> The second problem is that I am able (know how) to analyze polar interactions (HBonds). But can I somehow check for hydrophobic or other kind of interactions?
> 
> The third thing is that, for example, I have Vina output files for ligand docked in 2 different (but adjacent) places in protein (so 18 positions in total). Is there any chance to merge those files so I can see the energy differences among all of them?
> 
> Fourthly, is there any way to merge vina and an other program's output (4.2 or DOCK for example) and see if there are common structures (same one generated by both) and the ones that are not common for the two?
> 
> Finally, is there any graphical way of representing the docking poses and see the likeliness of the binding of a certain poses? Or some sort of clustering of poses into more probable binding candidates or something similar?
> 
> Thank you in advance,
> 
> Denis