The first step would be to look the pubished work about your enzyme. Specially if it is crystallized (http://www.rcsb.org/), this kind of papers have info about the active site. If the active site is unknown, try to mutate residues not conserved. Ussually they are not important. But you will never know...

don't know about about the best sites, but I would do the following: find related proteins, form the MSA, and look for conserved positions.... those are the positions that I WOULDN'T touch..... now you have a great varieaty of options - where to look for related proteins, how to form the MSA......

As mentioned above, the catalytic pocket or the Protein-protein interaction interface which is usually conserved should not be changed. Maybe you can look the published paper for help concering the protein u are interested in.

Hi everyone!

Thank you for your quick responses! The particular amino acid codon I'm targeting is not important for either the DNA binding domain or the ligand binding domain of my receptor of interest, and from what I've seen, these are the domains that have been targeted in the crystallography experiments. I'm just concerned about its importance in determining the overall structure of the protein.

Make a multiple sequence alignment of your protein. If it is already in the PDB, than you can get the ready-made alignment from the HSSP databank. If there is no published structure, you can make an alignment with the XSSP webserver (it is very easy, but takes a while to run). At the bottom of these alignments there is a big table with the conservation per position and which amino acids are found at that position. You can see whether it is safe to mutate from this table.