If we include beta-lactam antibiotics on the grounds that they have the same biosynthetic origin, peptides remain commercially the most important group of pharmaceuticals. However, our increasing knowledge of the genetic and enzymic background to biosynthesis, and of the regulation of metabolite production, will eventually bring a more unified approach to bioactive compounds. Mixing of structural types will become important, and we will be able to use our knowledge of biosynthetic genes and their regulatory networks. We will also benefit from an appreciation of the modular organization of catalytic functions, substrate transfer mechanisms and signalling between interacting enzymes. Since all of this is, in fact, the basis for enzymic synthesis of complex natural products in vivo, the exploitation of living cells requires mastery of a formidable network of cellular controls and compartments. For the present we are able to see fascinating connections emerging between genes in a variety of reaction sequences, not only in biosynthetic but also in degradative pathways. Peptide synthetases show surprising similarities to acylcoenzyme A synthetases, which are key enzymes in forming polyketides as well as in generating the CoA-derivatives that serve as substrates in degradative pathways. 4'-Phosphopantetheine, the functional half of CoA, plays a key role as the intrinsic transfer cofactor in various multienzyme systems. The comparatively small catalogue of reactions modifying natural products, notably epimerization, methylation, hydroxylation, decarboxylation (of peptides) and reduction/dehydration (of polyketides) can be found within or amongst biosynthetic proteins, generally as modules and organized in a specified order. The biochemist is coming close to the synthetic chemist's recipes, and may soon be recruiting proteins to carry them out.