The ability to control the substrate specificity and stereochemistry of enzymatic reactions is of increasing interest in biocatalysis. As this review highlights, this control can be achieved through various means, including mutagenesis of active site residues, alteration of physical variables like temperature and pressure, as well as through changing the reaction medium. While the focus of this article is on alcohol dehydrogenase reactions, each of these techniques can be readily applied towards other enzyme classes as well.In order to understand better the orientation of the substrate in the enzyme active site, it is important to have a map of the active site. In pioneering studies of the 1960's, Prelog proposed using a diamond lattice structure to visualize oxidoreductase active sites. 2 To validate his hypothesis, he tested the reaction with a series of increasingly larger substrates, and was able to determine the maximum spatial limits of the active site in each direction. By overlaying the molecules, which gave successful results, the active site was effectively mapped. It was important to have a means to map the active site, since this would give a measure of the limit of stereospecificity, and in the future, substrate specificity.In an extension of Prelog's diamond lattice model, Jones and coworkers further studied its utility towards horse liver alcohol dehydrogenase (HLADH) reactions. 3-7 In the early 1980's, Jones and Jakovac proposed a cubic space section model, due to the disadvantages of the diamond lattice for mapping an active site. 3 Among the disadvantages were the use of sp 3 hybridized carbon bond lengths and angles which could invariably limit researchers to use the diamond lattice model to choose substrates for analysis. With the cubic space section model, the cube sizes were flexible, which allowed researchers to use blocks that were as large or small as desired. The results for simpler examples, which worked adequately with Prelog's diamond lattice, also worked satisfactorily with the cubic model. Jones and coworkers highlighted further examples that fit nicely with the new cubic shaped model. [4][5][6][7] While mutagenesis can be accomplished routinely at present, it was not always the case. Hence, Jones studied the varying active sites of naturally occurring variants. Even though the cubic space model was useful, it was not widely adopted by other research groups due to the advent of more straightforward X-ray crystal structures, which give very clear information regarding the shape of the active site. Nonetheless, these early active site models have paved the way for