I'm trying to do Normal Mode Analysis for my MD trajectory. I'm following Amber tutorial (http://ambermd.org/tutorials/advanced/tutorial3/py_script/section5.htm) which uses an example of estrogen receptor protein and raloxifene ligand. Since I'm only getting errors from this, I'm having a couple of technical questions.
- As for solvent, my system is explicitly solvated. Skjaerven et al. in J Mol Str: THEOCHEM 898 (2009) 42–48 claims as follows:
"normal modes of proteins are most often calculated in the absence of solvent or with simple solvent models, as opposed to explicit solvent molecules in molecular dynamics simulations. (…) a common approximation is the absence of (explicit) solvent. Studies have shown that explicitly present water yields an increase in the density of local minima compared to vacuum simulations. In addition, water is found to increase the friction and amping on the atomic fluctuations."
Could NMA be used for proteins? I've heard an opinion that rigid rotor model used in NMA is not appropriate for proteins which are flexible.
The molecule in Amber tutorial example is a protein and it's solvated in explicit solvent (I think).
So, to conclude, how does it all make sense?