Why Cluster Pdbs By Rmsd?

Question

[Harpalss]

I've been looking into RMSD clustering of PDBs, alot of my google searches just returns RMSD clustering servers, so that lead me to posting this question.

Say for example I have 100 PDBs, and to an all against all superposition and calculate the average RMSD, I then cluster the averaged values, What information do you gain by clustering PDBs by their RMSD?

Answer 1

[miquelduranfrigola]

I'm not an expert, but roughly average RMSD measures structural similarity. By clustering structures using av.RMSD, basically you are grouping together similar structures. This can be useful, for instance, if you run a molecular dynamics. In this case, say you have a PDB file for each time unit, then upon clustering you can explore changes of conformation, that is, different clusters will represent different conformations.