I would like to know the stable and variable region of a protein. I have the structure/PDB of the protein, I think long simulation can tell me the stable and variable region of the protein. Please advis. .
What do you mean by variable? If you are referring to flexibility then you can look at the RMSF (root mean square fluctuations) for instance of the backbone atoms. I don't know which program you are using, but in principle all major MD packages should be able to do this analysis. It might be a good idea not to use just one long simulation but multiple to get a feeling of the error associated with the RMSF. Also make sure to look at the "equilibrated" protein only.
The way I'm doing, not sure if is the right way:
1)Made a database of protein in MySQL
2)Filter the protein with good resolution
3)Named each atom to have a common name on each aminoacid as C1, C2..
4)Use MySQL to collect all combination of distance between same protein and the neighbourhoods.
5)Match similar situation inside protein and neighbourhoods to find minimal and maximal distances.
You can use the biases toward order promoting (hydrophobic) or disorder promoting residues to predict stable/disordered regions from the sequence.
An example is Foldindex: Prilusky et al., “FoldIndex: a simple tool to predict whether a
given protein sequence is intrinsically unfolded”, Bioimformatics, vol. 21(16), p. 3435-3438, 2005
I believe you want to tell rigid from flexible regions in a protein structure, stable and variable as you call them. If so, PDB b-factors tell you about that and you can visualize them by coloring the structure by b-factors, in pymol, or by temperature, in rasmol.
- Badges
- Science topic