Fully Atomic Simulations

The most thorough representation of folding would involve describing both the protein and its environment in explicit atomic detail. This level of detail comes at a significant computational cost, with typical simulations consisting of hundreds of protein atoms and thousands of water molecules. 

Systems of this size prohibit the use of a computational approach based solely on quantum mechanics, a method that can only provide an exact solution for systems with limited number of electrons. To study the dynamics of proteins, an approximate solution to the Schrodinger equations, such as the one given by molecular dynamics using force fields, is required. This type of molecular dynamics is based on three approximations  

The Bom-Oppenheimer approximation that allows decoupling of the motions of the electrons and the nuclei. 

An approximation of the potential energy surface by a potential energy function (PEE) describing the physical interactions between the particles. The PEE permits the calculation of the potential energy and interatomic forces as a function of the coordinates of the system. In the case of proteins, the PEE are atom-based rather than nuclei-based. 

A typical PEE used in biomolecular simulations consists of bonded and nonbonded interaction terms  

---

Fully atomistic simulations are the most realistic of the three simulation methods. They include a fully detailed description of the amino acids comprising the protein, and they are thus much more true to life than the other models. In addition, solvent molecules may be added explicitly or implicitly to the simulation. Because of this extreme detail, a simulation of a small protein may require the treatment of thousands of atoms. Fully atomic simulations are thus extremely computationally expensive, and only short time scales can be explored. As computational power continues to increase, so do the time scales accessible with this method. Nevertheless, fully atomic simulations still cannot capture kinetic information they are, however, useful in understanding important local interactions that drive protein folding. 

In the next section, we present a tutorial on simplified and fully atomic simulation methodologies that can be used to investigate the features of energy landscapes as well as to reveal the nature of the thermodynamics and kinetics of folding. 

The protein folding problem is of fundamental importance in modern structural biology. Recent advances in experimental techniques have helped to elucidate thermodynamic and kinetic mechanisms that underlie different stages of the folding process [1-6]. Computer simulations performed at various levels of molecular detail have played a central role in the interpretation of experimental studies. Molecular simulations using models based on fully atomic representations are becoming more accurate and more practical and are increasingly... 

Figure 5.1 Anticipated results from the fully coupled simulation of gas-liquid flow. The pictures show the interface as it evolves, from image (a) to (d) during primary atomization. Bcised qualitatively on [9].

As in any molecular level simulation one of the first decisions to make is what inter- and intramolecular force field to use. We have basically two choices. Firstly, we can set about bringing together as much information as possible from experiment and quantum mechanical calculations to develop good force fields and in this way to aim for quantitatively accurate modeling. With this approach there is usually little alternative, but to employ a fully atomic representation with, for instance, the GROMOS force field. ° It must be remembered however that no force field will be completely accurate and all of them have limitations. 

Example If a drug molecule interacts with a receptor molecule through hydrogen bonds, then yon might restrain the distance between the donor and acceptor atoms involved in the hydrogen bonds. During a molecular dynamics simulation, these atoms would slay near an ideal value, while the rest of the molecular system fully relaxes. 

Molecular dynamics simulations have also been used to interpret phase behavior of DNA as a function of temperature. From a series of simulations on a fully solvated DNA hex-amer duplex at temperatures ranging from 20 to 340 K, a glass transition was observed at 220-230 K in the dynamics of the DNA, as reflected in the RMS positional fluctuations of all the DNA atoms [88]. The effect was correlated with the number of hydrogen bonds between DNA and solvent, which had its maximum at the glass transition. Similar transitions have also been found in proteins. 

Figure 10 Elastic incoherent structure factors for lipid H atoms obtained from an MD simulation of a fully hydrated DPPC bilayer, and quasielastic neutron scattering experiments on DPPC bilayers at two hydration levels for (a) motion in the plane of the bilayer and (b) motion m the direction of the bilayer normal.

This chapter has given an overview of the structure and dynamics of lipid and water molecules in membrane systems, viewed with atomic resolution by molecular dynamics simulations of fully hydrated phospholipid bilayers. The calculations have permitted a detailed picture of the solvation of the lipid polar groups to be developed, and this picture has been used to elucidate the molecular origins of the dipole potential. The solvation structure has been discussed in terms of a somewhat arbitrary, but useful, definition of bound and bulk water molecules. 

---