Lipase B from Candida antarctica (CaLB) is one of the most largely employed biocatalysts
for the synthesis of chiral fine chemicals. The successful application of this enzyme has also been
promoted by advanced computational methods able to simulate enantiodiscrimination at molecular and
energy level. Quantitative prediction of enantioselectivity remains a challenging task, affordable by
means of sophisticated and rigorous QM/MM methods or by hybrid methods that combine molecular
mechanics with experimental data and regression analysis. Most of the methods reported in the
literature aim to predict CaLB enantiopreference and to understand the structural basis of
enantiodiscrimination. Various experimental problems, such as resolution of alcohols, amines and carboxylic acids,
solvent effect, entropic contribution of substrates, are expected to receive beneficial indications from novel advanced
computational methods. However, the choice of the appropriate strategy is crucial for success in solving specific problems
within a realistic time frame and with a convenient computational cost. In order to be competitive with experimental work,
the rational and computational approach should be ideally within a high throughput scheme. Therefore, automation of
computational procedures, software and scoring steps represents a new emerging and promising perspective to make the
planning of biotransformation more effective and rational.
