Of the ~400 known human proteases, ~14% are under investigation as drug targets. Although the total is certain to rise during the finishing phase of the human genome project, the initial annotation of the ~30,000 human proteome set includes ~500 proteases. Bioinformatic analysis can now be performed on complete human protease families and will soon include comparisons with mice and fish. New sequences will require evaluation of their function in normal physiology and human disease. By revealing details such as splice variants and population polymorphisms, genomic sequence information will have a central role in the validation of protease drug targets.