Talk:Molecular mechanics

From Wikipedia, the free encyclopedia

WikiProject Physics This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.
??? This article has not yet received a rating on the assessment scale. [FAQ]
??? This article has not yet received an importance rating within physics.

Help with this template Please rate this article, and then leave comments to explain the ratings and/or to identify its strengths and weaknesses.

Polarizability

I noticed that Biophys wrote about polarizability being included into MM. I always conceived of polarizability as inherently dynamic. But of course its timescales should be faster than molecular motion. There are MD force fields that explicitly account for polarizability. How do these methods work in MM? Jorgenumata 10:09, 20 November 2006 (UTC)

Of course you are right that polarizability (and especially its orientational comonent) is dynamic. That is why there is no way to reproduce it in MM. My personal opinion is that polarizable force fields do not work too well at the moment. But I did not do any serious research here. If you find any publications were people were able to reproduce (by MD simulations!) the temperature dependence of macroscopic dielectric constant for different materials (this dependence is opposite for liquids and solids), that would be an indication that polarizable force fields are working. Biophys 14:43, 20 November 2006 (UTC)

I think this article is redundant and should be merged into molecular dynamics and force field (chemistry). It is completely equivalent, even from the content. The term "molecular dynamics" is nowadays more common than "molecular mechanics", except in the fixed expression QM/MM (quantum mechanics/molecular mechanics). Jorgenumata 21:17, 9 October 2006 (UTC)

I disagree in principle, although I agree that there is a lot of redundancy between the articles and that they would benefit from some reorganization.
Molecular mechanics (MM) and molecular dynamics (MD) are two completely different things, and a force field is only one part of molecular mechanics. MD is a simulation of how a system evolves with time. Although the most popular form of MD uses an MM force field for computing the forces and energies, there are other types of MD that use ab initio calculations or DFT. Therefore, not all MD is MM. Conversely, not all MM is MD either, because MM can also be used for doing Monte Carlo simulations, or for doing energy minimizations, conformational searching, and potential energy mapping, among other things. The term "force field" refers specifically to the functional form and parameters that are used for computing the energy, while the term "molecular mechanics" has, in my opinion, a looser definition that includes the various tools and techniques. Itub 23:11, 9 October 2006 (UTC)
These articles have been on my "sometime soon" list for a while now for exactly this reason; they don't adequately explain the differences. If you can get it, see this review paper for an overview, though Itub has summed it up pretty well. It is possible to have MD that does not use MM, and it is possible to use MM for applications besides MD. Opabinia regalis 00:54, 10 October 2006 (UTC)
Thank you for the review article. I'll be sure to check it out. I originally suggested the merge, but am now convinced of the separateness of the topics. It is clear that MD can be more neatly defined (for example as "the solution of the equations of motion for a molecular system and the extraction of properties from the resulting trajectories". MM is referring to a set of tools, and different people would put different models and analytical tools into this category. We really need to include some text explaining the differences between MD and MM in both articles. Jorgenumata 21:38, 22 October 2006 (UTC)
Molecular mechanics includes two different parts: (a) a model for calculating conformational energy, i.e. "force field" (the way of calculating this energy actually assumes that the molecule is a "mechanical system"), and (b) methods for global and local optimization of this energy, including such as Monte Carlo. So, "Force fields" and "energy optimization methods" are two separate subsections of MM. One could consider MD just as another method of global energy optimization, but this is more about modeling of molecular motions. So, I agree that MD should be a completely separate topic, which would refer to MM and force fields. Part "Integration and propagation" in the current MM article belongs completely to MD! It must be moved. As for "solvation", it may be present in both MM and MD, although this is usually only implicit model in the case of MM (for example, in FANTOM). Biophys 04:36, 25 October 2006 (UTC)
OK, so now I deleted the section on "Integration and Propagation" as it belongs in article Molecular Dynamics. Also added section on Areas of application. I think it reflects the discussion we've had here over the last days. Jorgenumata 08:57, 25 October 2006 (UTC)
I edited the "Areas of application" a little. Biophys 18:54, 31 October 2006 (UTC)
Hello Biophys! I think this page really needed a revamp to make make the distinction between MD and MM more clear. I did like the tone of my previous text about enthalpy vs. entropy. I said that "One pitfall of local optimization methods is the assumption that the enthalpic component of free energy (i.e., the evaluation of the potential or force field) is the dominant contribution to equilibrium. MM methods that neglect entropy assume that the minimum enthalpy conformation is most populated in equilibrium, which may not always be the case." And you took out all references to the force field being only the enthalpic contribution and reformulated it to "the analysis of equilibrium between different states requires also conformational entropy be included. The calculations involving conformational entropy are rather common." Now, is it really the case that most MM calculations include conformational entropy? I think the warning was justified. What do you think? Jorgenumata 09:36, 1 November 2006 (UTC)
I have made a couple more changes. I am sorry if I excluded any references. Could you put them back please? I agree that conformational entropy is important, and you are right that most of MM simulations do not include it. Only some do. So, "rather common" was probably wrong. I just wanted to emphasize that it is possible to include conformational entropy in molecular mechanics or other molecular modeling methods, and some people do just that (basically using the equation in the conformational entropy article). I do not think it was clear from your text. Most people do not really think that the minimum enthalpy conformation is most populated in equilibrium; they know about entropy. They simply do not care. Biophys 19:08, 1 November 2006 (UTC)

I like the new formulation. It is actually a strange state of affairs, to ignore entropy completely. Actually the formula you find in the article on conformational entropy for side chain rotameric entropy is problematic. Further discussion about conformational entropy was moved to Talk:Conformational_entropy 87.123.31.4 21:00, 5 November 2006 (UTC)

Suggest the term "Force field" should occur somewhere in this article, and that there be a link to the Force field (chemistry) page

It is mentioned and linked, although perhaps not prominently enough. Itub 11:38, 28 June 2006 (UTC)
I fixed this. I also did not like second phrase (removed). It mixes MM and MD. Besides, Monte Carlo can be applied for ligand docking. The phrase could be O'K if formulated better. Biophys 23:20, 4 November 2006 (UTC)