unanswered Martini & Targeted Molecular Dynamics (NAMD)

  • irenicus777
  • irenicus777's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
9 years 1 month ago #4461 by irenicus777
Hello,

I'm creating this thread to ask whether it is possible to perform Targeted Molecular Dynamics using a coarse-grained force field such as Martini.

I've been trying to study the effects of specific structural movements upon a transmembrane protein system using the Targeted MD method implemented in NAMD ( www.ks.uiuc.edu/Research/namd/2.10b1/ug/node47.html ). In this process, a subset of the simulation is guided towards a final 'target' structure by means of steering forces, and the force for each atoms is obtained through the gradient of a potential calculated as a function of change in RMSD.Therefore, if you had a protein in two distinct conformations, you could simulate the transition from one to the other through TMD, and explore the impact of the structural movements to the rest of the system (for example, to the protein's partner in a complex).

However, the size of the system I'm studying (multiple chains forming a complex, large lipid bilayer etc) makes simulation difficult. Therefore, since NAMD can perform simulations with MARTINI, I was considering coarse - graining the system and running TMD for the martini model.

First of all, in your opinion, does that make sense? Would TMD work for a CG force field such as MARTINI in the manner it would do for a standard atomistic one?

If I do combine the two, what should I keeping mind when setting a value for the elastic constant applied in the forces' calculations? NAMD's user guide suggests 200 kcal/(mol.angs^2), and in all-atom TMD this seems to work well, however I'm not sure whether that same value would be appropriate for a coarse grained simulation.
Also, from what I've found out myself by running unbiased CG simulations, they tend to give larger RMSD values than the ones obtained through all-atom ones. Since, in TMD, RMSD is used to calculate the steering forces, how do you believe this difference between CG and atomistic RMSDs factors into the process?

I would be interested in reading your thoughts on the matter.

Please Log in or Create an account to join the conversation.

Time to create page: 0.082 seconds