The difference between energy minimization and molecular dynamic simulation can be summarized like this. All molecules can exist with relative atom placements that are energetically unfavorable (i.e. it requires input of energy to put a molecule into a particular conformation). All molecules also have one (or more) optimal atom placements, where any change in any atom position will require input of energy. This is a minimum energy (ME) conformation. Energy minimization is the process of computationally finding the closest ME conformation. Because of the way minimization algorithms work, the located minimum may not actually be the global ME but instead a local minima. In other words, any slight atom movement away from the local ME conformation will require imput of energy, so it is indeed a minimum. However, there could be other quite different atomic arrangements that actually have a lower ME but energy minimization alone may not find these.

Molecular dynamics is the computational process of simulating molecular motion. Similar (or identical) "forcefields" are used as is the case with energy minimization algorithms. In this case, the goal is not to reduce energy to the closest ME conformation but to more or less conserve energy while allowing atoms to move about (while constrained by the forcefield). If allowed to proceed long enough, such dynamic simulations can often sample quite different molecular conformations, especially if the amount of starting energy is high (e.g. "high temperature"). Coupled with energy minimization, such dynamic simulation protocols (e.g. monte carlo) have a better chance of finding correct global ME conformations than simply starting with a (random?) conformation and minimizing that.... (You also can learn things about how the molecule moves and what conformations are likely populated most of the time)

Hope that helps.

-Ryan

p.s. Sorry, but I don't know anything about Abalone, so I can't help you there.

Dear Seyed,

Energy depends on the coordinates of the system (Molecule, Protein, Solvent) and the used force field. If two Programs use the same force field, the energy for the same system and coordinates are (almost) the same. Different force fields have different accuracy, limitations and parameterization.

Check reviews by e.g. Van Gunsteren in Angewandte Chemie or a recent paper from D.E. Shaw in PLOSone for differences and general trends.

MD prdoduces a series of coordiantes for one system at a given Temperature, thus you can get ensemble average energies at e.G. room temperature and can compare with experimental data.

Fully minimized systems can give you a different view and you may compare distinct, differnt conformations (local minima) and their relative energy differences.

Abalone should do both, geometry optimization and MD, so you can easily try it out.

spdbviewer has no solvent and provides no further modeling options, so be careful.

Happy Modeling,

Marco

Energy minimization raise down the energy of a structure.

Molecular dynamic simulation resolve atomic movement equations to simulate structure function.

The difference between energy minimization and molecular dynamic simulation can be summarized like this. All molecules can exist with relative atom placements that are energetically unfavorable (i.e. it requires input of energy to put a molecule into a particular conformation). All molecules also have one (or more) optimal atom placements, where any change in any atom position will require input of energy. This is a minimum energy (ME) conformation. Energy minimization is the process of computationally finding the closest ME conformation. Because of the way minimization algorithms work, the located minimum may not actually be the global ME but instead a local minima. In other words, any slight atom movement away from the local ME conformation will require imput of energy, so it is indeed a minimum. However, there could be other quite different atomic arrangements that actually have a lower ME but energy minimization alone may not find these.

Molecular dynamics is the computational process of simulating molecular motion. Similar (or identical) "forcefields" are used as is the case with energy minimization algorithms. In this case, the goal is not to reduce energy to the closest ME conformation but to more or less conserve energy while allowing atoms to move about (while constrained by the forcefield). If allowed to proceed long enough, such dynamic simulations can often sample quite different molecular conformations, especially if the amount of starting energy is high (e.g. "high temperature"). Coupled with energy minimization, such dynamic simulation protocols (e.g. monte carlo) have a better chance of finding correct global ME conformations than simply starting with a (random?) conformation and minimizing that.... (You also can learn things about how the molecule moves and what conformations are likely populated most of the time)

Hope that helps.

-Ryan

p.s. Sorry, but I don't know anything about Abalone, so I can't help you there.

Energy minimization is related to molecular mechanics. It defines the most stable structure. THIS IS the STATIC POINT OF VIEW

Molecular dynamics is related to the evolution versus time of the structure of the molecule. THIS IS the DYNAMICS POI9NT OF VIEW;

Thank you everyone for your helpful comments

Just to extend Ryan's description, alrready a good one

-In Molecular Mechanics (MM) you "approximate" the geometry of an atomic system using a mechanical system analogue and endowing it with special properties, all taken from the classical mechanics (ignoring the contribution of quantum mechanics). The approximation is fitted to known molecular systems through the adjustment of empirical coefficients. Now you can use MM to several aims.

-In energy minimization you change the position of "atoms" in space and search for a conformation with low energy. The strategy to move the atoms around may vary, but commonly you pay attention to the forces imposed to each atom by its neighbors and use the gradient generated by these forces to guess how each atom should be moved to, and by how much. In this strategy atoms are assumed to be at rest, i.e. they have zero kinetic energy (which in practice does not happen for a true atomic system, for it would be against the second law of thermodynamics).

-In Molecular Dynamics (MD), as the name implies, you apply energy to the molecular system by association of a velocity vector to each atom, that gives you an overall kinetic energy (E=(1/2) mv^2) which you have to calculate to make fit to the Maxwell-Boltzmann equation at the temperature you wish to simulate. Then you can calculate, using MM, all the forces acting on each atom, at very short time intervals (integration time, usually 1 or 2 fs), you calculate how much each atom has move and estimate its new position. You also determine how forces acting on each atom changed. This is done over and over and tells you how the molecule vibrates and moves. In theory, the system should move following a random walk towards an equilibrium, but even at this equilibrium atoms have kinetic energy and they show stochastic oscillations around the equilibrium conformation.

There are also Monte Carlo (MC) simulations, where the system is set as in MD, but the atom coordinates are moved randomly (not accordingly to their speeds), usually with very small moves each time and avoiding aberrant moves (such as one bond forming a knot). Each move is evaluated and, if "acceptable" it is taken, if rejected, we go back to the previous position. Then a new move is calculated and everything is repeated over and over.

I guess you could say that EM is a kind MD or and MC performed at zero K temperature.

EM seeks for the lowest energy conformation (usually a local minimum), while MC samples the equilibrium conformation and properties of the system at some temperature, and MD explores how the systems approaches the equilibrium state at a given temperature.

Best wishes,

Rogelio

There is heaven and earth difference between molecular dynamic simulation and energy minimization. Abalone software is a very common software which will help you in energy minimization studies but independent of molecular dynamic simulation data. You need to look for other soft ware for molecular dynamic simulation studies and then gel the data obtained from both the softwares to reach to some sensible conclusion.

thank you Rogelio. could you please introduce me a good reference (or book) about these topics?

I have a question .. If I take a energy minimized conformation and start dynamics with that conformation as a starting point ... what will be its potential energy ?

Mr. Motevalli ,

To the topic of your question there is better to pay attention to the following references. Because of "MM" and "MD" have encompassed not only one method to each of those terms. Furthermore there are developments to each of them. The shown refs contain basic theories and more recent trends to methods of computational chemistry, particularly highlighting "MM" and "MD" (refs 1 and 2) as well:

1. Computational Molecular Science, P. Schreiner, W. Allen, M. Orozco, P. Willett (Eds.), Vols 1 - 6 (pp. 1 - 3041), Wiley, Chichester, 2014.

2. Encyclopedia of Computational Chemistry, PvR Schleyer, N. Allinger, J. Gasteiger, P. Kollman, H. Schaefer III, P. Schreiner (Eds.), Vols. 1 - 5 (pp. 1 - 3375), Wiley, Chichester, 1998.

thank you for your helpful comment