The study of oncogenes has illuminated many aspects of cellular signaling. The delineation and characterization of protein modules exemplified by Src Homology domains has revolutionized our understanding of the molecular events underlying signal transduction pathways. Several well characterized intracellular modules which mediate protein-protein interactions, namely SH2, SH3, PH, PTB, EH, PDZ, EVH1 and WW domains, are directly involved in the multitude of membrane, cytoplasmic and nuclear processes in multicellular and/or unicellular organisms. The modular character of these protein domains and their cognate motifs, the universality of their molecular function, their widespread occurrence, and the specificity as well as the degeneracy of their interactions have prompted us to propose the concept of the `protein recognition code'. By a parallel analogy to the universal genetic code, we propose here that there will be a finite set of precise rules to govern and predict protein-protein interactions mediated by modules. Several rules of the `protein recognition code' have already emerged.