I have simulated my viral protein at a temperature of 300K (room temperature). However, it does not seem to stabilize even after 400 ns of simulation. I have observed many papers where simulation has been carried out at 310K (body temperature). Could this be a possible reason?
I don't think so. Because in Molecular dynamics temperature is not a "real temperature" but an abstract parameter. Water at 100C doesn't boil and doesn't freeze at 0C. The increasing temperature could speed up the fold (because the atoms will move faster) but could not change the experiment completely. The "gold rule" is that MD simulation should be run at least 1 mks for predicted or unfolded proteins. Or you can use Replica Exchange Molecular Dynamics. It could help exit the energy barrier and find the structure with the global energy minimum, not a local one. And structure I hope my answer will be helpful.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6484850/
The difference between 300 and 310 K is unlikely to be your problem. A standard strategy for starting an MD simulation attempts to gently release the protein into its simulation box with a conformation that is close to the original crystal structure.
1) freeze protein atoms and minimize the solvating water.
2) minimize protein and water (and any ions)
3) run NVT simulations for a while with the C\alpha carbons frozen
4) run more NVT simulations with C\alpha carbons dynamically restrained
5) transition to NPT simulations for production (cannot constrain NPT).
If your molecule ventures far from its crystal structure position, it is unlikely to ever return. The relevant time scale for protein folding/unfolding is probably measured in milliseconds. You will never find it in a microsecond simulation.
If, however, you are starting with a predicted structure, then your results may also be interpreted to mean that the prediction is wrong. Despite media fanfare that the structure prediction problem has been solved by AI, this assertion is just hyperbole.
- Badges
- Science topic