While running a simulation on any peptide or protein, is it possible for the secondary structures to change? For example, seeing a steady rise in helical content over the whole simulation?
Dear Jordan,
yes, it is possible...of course dependent on the simulation time. Straightforward examples that secondary structures change or are formed are "folding simulations", where stretched peptides acquire secondary and tertiary structure in a MD simulation. A 'classical' example is the folding of the Villin headpiece protein from a stretched conformation into the structure also found in crystals in a 6 µs simulation. This and other examples can be found here: http://www.ks.uiuc.edu/Research/folding/
Best regards
Johann
Dear Jordan,
yes, it is possible...of course dependent on the simulation time. Straightforward examples that secondary structures change or are formed are "folding simulations", where stretched peptides acquire secondary and tertiary structure in a MD simulation. A 'classical' example is the folding of the Villin headpiece protein from a stretched conformation into the structure also found in crystals in a 6 µs simulation. This and other examples can be found here: http://www.ks.uiuc.edu/Research/folding/
Best regards
Johann
Yes, it is possible. I observed it in my recent MD simulation of my project.
Dear Johann P Klare ,
Thanks a lot for the answer! I learned quite a lot from it
Dear Jordan,
Of-course it is possible to see such changes during your simulation. However, this change will only depend on your system and the number of contacts it is associated with during the simulation time. Folding of a protein is very general with large number of external and internal factors associated with the protein. If your peptide or protein is disordered, it can get helical nature in presence of other binding protein or other system.
It is very possible to observe such changes, particularly for turn/coil regions, or regions with less stable secondary structure elements (3-10 helices, pi-helices etc). In fact, in many cases it is expected. However, whether such changes are structurally and functionally relevant depends upon your system and your simulation parameters.
One very important parameter to consider is the choice of force field. Some commonly used force fields tend to be too "helical", i.e. they are biased towards spontaneously forming helical structures - the so-called "coil to helix transition" issue (and, conversely, some - fewer - force fields have been found to be biased towards the formation of beta strands). This has been documented in dozens of publications testing and comparing several different force fields (for example, the CHARMM force fields tend to be "helical", while the GROMOS force fields are more biased towards beta structures. The 99sb-ildn variants of AMBER seem to be among the more balanced options, at least according to the literature).
I would suggest that you consult the relevant literature on the matter. See for example this paper: Article Are Current Molecular Dynamics Force Fields too Helical?
It is a bit old, but still relevant, considering that many of the force field models it covers are, to this day, used by many labs. From there, move on to other similar publications, and use them as a guide to evaluate your own results, and consider whether these changes you're seeing are partly caused by the model of representation.
Fotis Baltoumas Thanks for the guidance! I'll surely read up on this topic to learn more.
- Badges
- Science topic