If my protein is large and has many domains of which one domain has the active site that I want to study ligand interaction for with MD simulation, can I cut out the domain of interest in the protein along with the bound ligand for MD simulation. The entire protein is very big and the time it would take for a 100ns simulation on a GPU computer (nvidia RTX 2070 is 3 months) so I would like to know if its okay to just extract the domain that has the active site to which the ligand is bound for MD simulation?
Dear Olatomide,
in most cases it is fine to proceed like you suggested. Protein domains are more or less defined by having a stable tertiary structure without the adjacent domains. Actually, most structural data for large, multi-domain, proteins has been initially obtained by solving the structures of their subdomains, often with bound ligands and for catalytic domains in the presence of different substrates. Also, being relevant for your case, numerous MD studies on isolated protein domains have been already published.
In your case, if the focus is on ligand binding to the active site, you should be on the safe side using only the catalytic/receptor domain.
Nevertheless, one has to be careful in not missing e.g. interactions with other domains that influence the active/ligand binding site. If no such interactions are present or appear to be possible, there should be no danger. And, of course, you might miss allosteric interactions between domains.
I hope this help!
Best regards,
Johann
Dear Olatomide
In most cases there will be no problem at least as far as computational studies are concerned. You will certainly have a slight deviation from the realistic value of protein binding due to internal forces developing between groups but generally I think it would be okay to do that. I wish you good luck with your project.
- Badges
- Science topic