Iterative enzymes are the least understood of the PKS classes where control of chain length, chemical stabilization of the poly b-ketone intermediate and “programming” of reduction/de-hydration and cyclization steps are largely a mystery. We define the term “deconstruction” in the context of multidomainal enzymes, especially iterative systems, as a conceptual and experimental approach to separately express and reassemble functional fragments of these proteins to isolate key reaction and control steps for detailed mechanistic and structural investigation. Soluble, catalytically active protein fragments have been obtained by applying UMA, an algorithm designed to identify linker regions in a family of homologous proteins based on their extent of primary sequence identity, predicted regional secondary structure and local hydrophobicity [J. Mol. Biol. 2002, 323, 585]. Use of this bioinformatics tool will be exemplified by PksA, an aromatic, fungal, iterative Type I PKS, and an associated yeast-like pair of FAS subunits, Hex A and Hex B, that jointly carry out the synthesis of norsolorinic acid, the first intermediate in the biosynthesis of the mycotoxin aflatoxin B1. Discovery and characterization of a “starter unit-ACP transacylase” (SAT) domain and “product template” (PT), or “cyclization template” (CT). domain will be described.