I'm working with few protein structures and want to superimpose them, say more than 2 proteins at a time. So for the purpose I'm looking for software that can fulfill my need for the purpose. Please suggest a few.
Dear Swapnil! there are two main approaches for superimposition of protein structures. One is "superposition" which is suitable for closely related structures and another one "structural alignment" which allows superimposition of structures with low identity. Both of them are realised in YASARA View which is free or higher (commercial) YASARA versions. But superposition requires the same molecule naming and residue numbering in YASARA
Try the Least Squares Comparison (Lsq command) in the program "O" (http://xray.bmc.uu.se/alwyn/A-Z_of_O/everything_e-l.html#anchor1340987) which according to my experience gives the best fit of the superposed structures.
CCP4 supported program : gesamt. This program can superpose the search model with an entire PDB archive (i.e. a directory with all pdb files to align).
As Bartosz suggested, Pymol can be used, with the command "Super". Also you can try Coot, which has both SSM and Least Squares Superimposition. Or Theseus, that uses maximum likelihood.
I agree wit most of the answers here, I just add a comment: try a multiple alignement before superpose your structures, it will help you on identifying a core of common/similar residues you can use to feed superposition. In fact with big structures and many loops could be difficult to correct superpose them unless you don't specify a core. For simple superpositions I found good this software (Superpose 1° link), available also as webserver, it allows you multiple superpositions at one time.
If you need to do a complicated jobs with involve also multiple alignments look at the 2° link.
You can use PyMol for superimposing the structures. However it works better with FATCAT server. First get the structure aligned in this server then use PyMol to view it. Hope this works.
I will recommend you a software called MATRAS that was developed by Dr Kawabata. It is available at http://strcomp.protein.osaka-u.ac.jp/matras/index.html , and very fast.
As Niyaz Safarov suggested, for quick alignments I teach people to use the graphical interface of PyMOL. One useful addition is the ability to align selections of molecules to other molecules or to other selections. I'll type up some example tutorials here:
Basic:
Open PyMOL
Type fetch 3ql3
Type fetch 4m6k
Click the Zoom button at the top right to see everything in your scene
In the menu on the right, click the A button next to [4m6k]
Drag down/left to select Action:align>AlignTo:molecule>Object:3ql3
The molecule 4m6k will snap onto the other one- its easier to see the relationship with ribbon cartoons, so in the top of the menu click the A button next to all and drag down/left to select Action:preset>publication
Color the molecules: Click the rainbow colored C button on the far right of [3ql3] and pick Color:Reds>Red
Repeat for 4m6k to pick Cyan
Using selections:
Restart PyMOL to ensure we are on the same page
Type fetch 3vkh (Dynein dimer (motor protein that walks on microtubules composed of tubulin) with more complete stalk on one monomer)
Type fetch 3j1t (short piece of the microtubule-binding stalk of dynein bound to a tubulin dimer
Click the Zoom button at the top right to see everything in your scene
Show both structures as cartoon ribbons colored by chain: In the menu on the right click the A button next to all and select Action:preset>publication, then click the rainbow colored C button on the right of the same row to select Color:byChain>byChain. You should end up with cyan, green, and magenta ribbon structures.
Try a default alignment: In the menu on the right, click the A button next to [3j1t]
Drag down/left to select Action:align>AlignTo:molecule>Object:3vkh
This puts the 3j1t somewhere in the middle of the large motor domain of chain A of 3vkh- this is very bad! The problem is that PyMOL is trying to align the entire structure, including the tubulin of 3j1t to 3vkh- we only want to align the homologous part of the total structure
First make a selection, then align that selection as follows:
Top menu click Display>Sequence On
In the bottom row, click the first K in chain A, which is residue 3264 of chain A, then scroll to the end of chain A in the sequence viewer to find the last residue of chain A which is also a K (last letter in green), hold shift and click to select all of chain A...
You will see a new object in your menu called (sele). Rename this by clicking the A button and type "mtBinder"
Next to [mtBinder] click A Align:AlignTo:molecule>Object:3vkh
Now the stalk and the associated tubulin dimer of 3j1t (i.e. all of 3j1t) will properly align to the end of the stalk of chain A in 3vkh
Click the A button next to [mtBinder] and select zoom, now you can rotate around and color the mtBinder red to get a good comparison of the fit of the microtubule binding regions from these two different structures.
Hi, I work on several programs. From my experience, Coot makes better superpositions than PyMol. For making pictures I save superimposed pdb files in Coot and work on them in PyMol. Cheers and good luck.
I have used several softs and my suggestion is: 1. Use coot to extract molecular info. Coot makes good fit and you can notice much from these superimpositions. 2. For making pictures use PyMol. However load structures which had been superimposed in Coot and their coordinates were saved in Coot.
How do u superpose two protein complex structures? I have a structure with a bound ligand and i want to compare with another structure from pdb with slightly different ligand bound at the same place. That is i have two different proteins and two different ligands
Ligand protein complexes can be superposed in FATCAT. Use the 'download complete structure of protein A on protein B' option and edit pdb file manually. Alternatively, Coot has SSM and LSQ superpose that are quite good. For multi-proteins, you can use POSA or Coot. I have used coot for upto 5 structures. PyMOL is best for figure preparation