[Chimera-users] Chimera problems
meng at cgl.ucsf.edu
Sat Oct 20 12:00:10 PDT 2018
> On Oct 19, 2018, at 8:41 PM, Shriwas, Pratik <ps774614 at ohio.edu> wrote:
> I am trying to use USCF chimera for studying docking between my compound and the receptor called GLUT1 (PDB - 4pyp or 1SUK). My concerns are:
> • I find different scores everytime I use the same compound and receptor
Sounds like you are using the Autodock Vina interface of Chimera. Chimera does not calculate the docking, it just sends/receives data from an outside web service that is running the Autodock Vina program. It may be that the docking process uses some random numbers, and because the web service does not allow a high amount of sampling, it cannot always find the same top score(s). However, I do not know the details of Autodock Vina and you would have to look at its website <http://vina.scripps.edu/> and/or paper to learn about that. I’m guessing you are using Chimera’s Autodock Vina GUI to enter inputs, not the “vina” command. However, if you are using the command make sure you get a recent version of Chimera, because some problems with the command were fixed.
> • I am not able to show my hydorgen bonds between compound snd receptor in GUI. Although I can see them in the list box.
Maybe the atoms are not shown (for example, if the sidechain is not shown, the H-bond to the sidechain won’t be shown). To make sure the H-bonding atoms are shown, in the H-bond parameters dialog turn on the option “If endpoint atom hidden, show endpoint residue.” However, if that is not the problem, then maybe the graphics on your computer has trouble showing lines. You could try updating your graphics driver, or try showing the H-bonds as sticks instead of lines with command: setattr p drawMode 1. Or maybe it can draw lines but they are too thin for you to see. In that case, in the H-bond parameters dialog increase the “Line width” value.
> • I am not able to find exact binding energy.
This is the central problem of docking! Firstly docking scores are not binding energies, only approximations, and secondly, one has to do a lot of conformational and positional sampling to find the most accurate position. The Autodock Vina website does not allow a high amount of sampling, as it is a shared resource provided for free. If you want to do more intensive docking you can download Autodock Vina (or some other docking progam like DOCK, Gold, ...) and run it directly.
> • My ligand is not on pubchem/NIH servers. What is best way to find its smiles string. Different websites have different smiles string and I get different interactions.
Usually there are multiple ways of writing the SMILES string for the same compound. As long as the compound looks right to you (has the correct covalent structure, the right elements connected in the right way) then the results should be equally valid even though the coordinates might be slightly different (e.g. rotated or translated or different conformation). Any slight change in inputs will change docking outputs, since sampling is not exhaustive, but the change does not mean one answer is “wrong” and the other is “right".
> Pratik Shriwas
> Phd Candidate,
> Molecular and Cellular Biology program,
> Edison Biotechnology Institute,
> Ohio University.
For future reference, Chimera questions should be sent to chimera-users at cgl.ucsf.edu (CC’d here). For Autodock Vina questions, see <http://vina.scripps.edu/questions.html>
I hope this helps,
Elaine C. Meng, Ph.D.
UCSF Chimera(X) team
Department of Pharmaceutical Chemistry
University of California, San Francisco
More information about the Chimera-users