Computer modelling and simulation
When studying biophysical phenomena it is not always possible to directly measure events happening at the nanoscale. These limitations are due to the sensitivity of the equipment used and the size scale of the system investigated. Likewise the complexity of the system and the timescale of the biophysical phenomena investigated also put restraints on what is measureable and what is not. A solutions to these problems is applying molecular dynamics, specialized computer modelling and simulations techniques, which can be used to characterise biophysical events on the nanoscale. By using the laws of physics such as Newtonian mechanics, thermodynamics, statistical mechanics etc, combined with experimental data it becomes possible to approximate the interactions in the system using simplified models. Molecular dynamics acts as a link between laboratory experiments and theory, thus enabling a scientist to examine properties of for instance a nanodevice or system that has not been produced yet.
Although molecular dynamics is a powerful scientific tool it also has its limitations. The models used for the calculations are often based on classical physics instead of quantum mechanics for sake of simplicity and lack of computing power. Therefore the model attained represents a simplification of the system. The complexity of the simulated system is limited by the computing power so only necessary system parameters are used for the simulation. Furthermore computing power restricts the time span of the simulation – most scientific simulations go from spanning nanoseconds to microseconds. Even though this might seem to be minute simulation times these simulations can take from several CPU days to CPU years to obtain.
See examples of computer simulations in the video gallery. Many of the processes illustrated here are hard or impossible to see in experiments because they are too small or occur too fast. This is why computer simulations are useful in that they can reveal details about dynamical processes that we cannot observe.
Water is present in ALL simulations but is invisible for clarity. Also, the movies are typically between a few hundred nanoseconds to a few microseconds long, but an exact time-scale is hard to measure. It takes about a week of computer time on a single processor to generate each movie.

- Computer modelling of chain reversal of oxidized lipids in a lipid bilayer composed of POPC lipids and oxidized PazePC lipids. The image shows a simulation snapshot of the chain reversal of a carboxyl-group containing lipid acyl chain reversing into the aqueous phase [Computer modelling : Himanshu Khandelia].