Entering edit mode
2.2 years ago
garasubin9
•
0
Hi, I am a new beginner in the field of bioinfomatics. I would like to obtain protein-protein/peptide complex sequences. How can I achieve this goal from using PDB such as RCSB?
To identify complexes, you can use an advanced search:
https://www.rcsb.org/search/advanced
There are various options in dropdown menus, and you need to find a combination that satisfies your criteria. As to getting only the residues that are part of the interaction, I don't know of an automated way to get that. Maybe you can get close to it from the PDBbind database:
http://www.pdbbind.org.cn/
Thank you for your reply! I have a question. Can I use information given by PDB file to fetch such sequences? For example, can I filter and obtain a specific residue(amino acid) whose distance far from another ligand is below x angstrom just according to the PDB file?
It is possible from PDB files, but I think you need complete files for it. There are programs that will graphically show the interactions between amino-acids and ligands.
https://www.ebi.ac.uk/thornton-srv/software/LigPlus/
It is very kind of you to give me such treasurable suggestions! I think it is trivial to use PDB file to achieve this. Maybe I should use some tool to get this....
All the major structural exploration / visualization tools such as PyMOL, Chimera, or Jmol / Jmol/ Jsmol have a
select withinor equivalent command that lets you specify a distance. Usually you need to exclude the residue or ligand you are using by adding in something along the lines ofand not <this residue>. You can then usually have the appropriate software report/print what the selected residues are.Maybe you meant this for visualizing? Usually you download the entire structure and then limit the selection to the portion close to what you want using the
select withinapproach and then turn off the rendering for the protein outside of that to focus in on the interactions of concern. While you can then extract the selected items on their own easily to a separate object in PyMOL , & maybe others, a lot of time that additional step is unnecessary as you can simply hide other parts from view. Or use 'slab' option to move them out of the view. Part of the reason is that later you may want to see how far something outside what is shown occurs from the current selection and you won't have that part if you limited your structure to only a part of the experimentally determined structure.For trying Jmol/Jsmol right your browser, you can go to FirstGlance in Jmol enter your favorite PDB identifier and then bring up your structure. Then right-click on the console and bring up a console where you can enter the
select withincommands for Jmol. For something easier, Proteopedia's Scene Authoring Tools have a nice 'select within distance' sectiont that lets you easily build scenes and selections using 'select within'. You can more easily control how things are represented there using the 'representations' and 'colors' tabs. It is for authoring scenes but it really has built in a nice viewer tool that works right in your browser.If it is not for visualizing, PDBsum and PDBePISA will tell you a lot about interactions with other chains and with ligands. At the PDBsum, each structure deposited at the PDB has a page with a 'Prot-prot' tab that details the protein protein interactions, see here for an example. Down on the right bottom side you can get those details as text. You click on the 'Interface' selections on the left side to see each in turn. Each appropriate structure at the PDB also has a 'Ligand' tab that you can click to see interactions with the ligands. For the same structure, you can see that here. PEBePISA lets you put in your PDB identifier and then select the Interfaces button at the bottom right to bring up a table of each interface that you can further drill down for details from that table. If you need to scale up collecting that information from a lot of structures, see here and here. The latter are all demonstrated in sessions that come up by going here and pressing
launch binderto bring up a list of notebooks demonstrating retrieving such data and parsing it via Python.See Proteopedia's Interface analysis servers page for other options.
Thank you for your detailed reply! Actually, I am searching for protein complex interface including sequences. Is there any good solution to achieve this goal?