Introduction

The existence of a large number of multigene families has become evident since the 1970s, largely as a result of the enormous amount of DNA sequence data available. Hood, Campbell & Elgin (1975) describe multigene families as groups of tandemly repeated genes, characterised by their multiplicity, linkage in clusters, sequence homology and related or overlapping phenotypic function. Searches for members of a multigene family can be made using DNA probes to search for related sequences within the genomic DNA, or, alternatively, the Gene Bank data base for that particular species can be searched to see whether related sequences have already been discovered. Multigene families are generally considered to have arisen by processes such as gene duplication and gene conversion. Two terms are used to describe homologies between genes and gene products. Genes are said to be orthologous if their sequences have diverged as the result of a speciation event, but they are paralogous if they have diverged as the result of gene duplication or some related process. For example, α-globin from domestic fowl and dog are orthologous proteins, whereas α-globin and β-globin both from domestic fowl are paralogous proteins. In this chapter we are mainly concerned with paralogous proteins.

Dayhoff (1978) has suggested that the proteins (≈ 500,000) expressed by the human genome can be grouped into about 500 superfamilies, each containing about 100 sequences. The term ‘superfamily’ is used here to distinguish distantly related proteins from closely related proteins.