I'm new to structure visualization and I'm trying to process a large collection of structures and visualize regions of evolutionary relatedness. I have marked pairs of residues on the structures and I'm trying to find a way of orienting the camera to provide 'the best view' of the selected residues before generating the images.
While I don't know what the general way to define the 'best view' of a region of interest in a protein structure I'm guessing it mostly has to do with choosing a camera position which has the least number of uninteresting residues blocking the marked ones.
Does anyone know of a, preferably pymol script, or tool that can determine an optimal viewing position. I have ~1500 structures that I need to visualize so its impracticable to do this by hand. I can probably write something in python but it would be easier if there was already something out there.
Thanks,
-Will
I think this one is question is discussed/answered in another question http://www.biostars.org/post/show/9437/looking-for-a-toolset-for-semi-automatically-coloring-3d-pictures-from-1d-sequence-segments/ ? Have you tried MolScript ?
Will Molscript position and the camera for optimal viewing?
You can select the residue(s) of interest and generate optimal view(s) based on 3D co-ordinates.
Can you be a little more specific on how how get rasmol to choose optimal viewing parameters, i haven't been able to find any documentation for how to do this. For example how can I get an 'optimal' image for the PDB structure 1A1N chain A ind 5 and chain A 38.
What do you hope to gain by visualizing so many structures? Are the structures very similar? I assume that they are not. This means you will have a large number of motifs of interest embedded in a variety of very different structures, and this means that it may be difficult to script something that gives the "optimal" view every time. You can probably set a certain number of rules in a script that give approximately what you want, but your mileage may vary depending on the different structures and how good your script-fu is.
Will visualizing this many structures be helpful to you? What about doing a multiple sequence alignment and then coloring by motif? If I were looking at similarities and differences between structures and absolutely HAD to visualize them, I would separate them by similarity or category. Instead of having an image for each structure, I would have a composite/overlay figure for each of my categories or for the top categories with the most structures (something like this: http://www.topsan.org/@api/deki/files/7443/=PX4068B-activesite.png ). I would also include the motif under the 3d visualization for each category or possibly a multiple sequence alignment.