Neutrality and Selection in the Evolution of Gene Families

Abstract

Evolutionary relationships among genes, as revealed by sequence similarity, are used to characterize gene families. Surprisingly, a power-law can reasonably describe the distribution of sizes of a genomes gene families. Evolutionary models are able to reproduce the size distribution with simulations of a set of genes growing through duplications and modifications. Most conspicuously, positive selection is not included in the models, suggesting per-haps, that neutral forces determine gene family sizes. Here I advocate this notion with comparative genomic analyses and a review of recent research on the evolution of gene duplicates. I show that a power-law also relates the sizes of orthologous gene families across 66 known microbial genomes. Furthermore, singletons (gene families of size = 1) in one genome have orthologs that are themselves power-law distributed in other genomes. The signature of positive selection, however, is revealed in the fact that gene families of size six and more have a more skewed family sizes distribution across other genomes. The general pleiotropy of genes and the notion that gene duplicates may rapidly subfunctionalize support the conception of gene family growth without positive selection. Such a model runs contrary to Susumu Ohno’s famous dictum that only “redundancy created” and suggests a novel view of the evolution of functional novelty.