Computer modeling is widely used in biological disciplines such as pharmacology. One example of this is the `lock and key' model of the interaction of a ligand with a receptor. The molecular structure of the receptor defines the shape of the `lock' which the ligand, the `key', must fit in order to bind. This concept is often used to aid in the design of candidate drug molecules which must bind to a specific receptor. Computational simulations may be used to evaluate the quality of the fit of the `key' within the `lock'.
Molecular simulations are not a replacement for conventional experiments. Instead, the two approaches may be used to complement one another. In common with any new experimental method, a computational modeling technique must be validated by comparing the results it generates with those of conventional experiments which are trusted. Simulations may be used to aid in the interpretation of experimental results and `computational experiments' may be used to provide supplementary information. These `computational experiments' can have many advantages over their more conventional brethren. They offer a very high degree of control over experimental conditions, a single parameter may be changed and the result observed. In addition, an enormous range of chemical and physical observables may be calculated with only one piece of `apparatus', the computer. The careful use of computational modeling can substantially aid in our understanding of biological processes.
Perhaps the most valuable rôle of computer simulations will be their use as predictive tools. If a technique has been validated, it may be used to predict the structure of a molecule or outcome of a reaction. Used in this way, computer simulations can provide large savings in time and money by permitting candidate molecules to be tested without the need to formulate the compound in the laboratory.