The following is from Animal Models in Light of Evolution p176-179.
The human genome has less than 30,000 genes. The mouse genome is about the same size as the human genome. That of Drosophila has about 14,000 genes. These genes perform a variety of roles in organisms. We can divide these genes into 3 broad groupings:
(a) housekeeping genes. These genes encode proteins that do essential biochemical work in the body such as metabolism and biosynthesis of macromolecules.
(b) specialist genes. These do specialized work such as oxygen transport, or immune defense.
(c) toolkit genes. These genes make the products governing the construction of the body (body plan, number and identity of parts).
(The handy metaphor of the toolkit is derived from Carroll et al. (Carroll, Grenier, and Weatherbee 2004).)
This division should not be thought of as a ranking in descending order of importance (toolkit genes are not luxury “add ons” relative to housekeeping genes, moreover, normal organismal development requires the activities of genes of all types. Nevertheless, developmental geneticists have a special interest in the toolkit genes.
It will not go amiss to summarize briefly what is known about the genes in the developmental toolkit (derived from Carroll et al. (Carroll, Grenier, and Weatherbee 2004):
 Only a small fraction of the genes in an animal genome appear to be devoted to development of body plan and body parts.
 Toolkit genes produce two classes of gene product of particular importance:
(a) transcription factors, i.e., proteins that regulate the expression of genes during development.
(b) proteins in signaling pathways that mediate interactions between cells.
 Toolkit genes are generally conserved among different animal phyla.
The developmental toolkit is very similar across morphologically different animal lineages. It is the unity underlying morphological diversity.
So what does this tell us about the differences between, say, humans and other mammals? At the genetic level of description there will certainly be some differences in the coding content of structural genes. But much more importantly, a common developmental genetic keyboard has evidently been used to play distinct developmental (phenotypic) tunes. As pointed out by Carroll et al.:
Although the expansion of the toolkit may be related to morphological complexity, no correlation appears to exist between toolkit expansion and animal diversity. The morphological diversity of vertebrates, from humans to hummingbirds, or from whales to snakes, evolved around a common set of developmental genes. For example mammals, birds and amphibians share the same set of 39 Hox genes. [(Carroll, Grenier, and Weatherbee 2004) p113]
A growing body of evidence now exists to support the claim that morphological diversity results from changes in gene regulation rather than (a) the acquisition of new genes for novel features, or (b) the evolution of functional changes in protein-coding sequences of structural genes. Humans and chimpanzees or mice are not the same animal dressed up differently, rather they are complex biological systems whose manifest differences reflect underlying regulatory differences with respect to the ties to a common set of developmental genes.
While the role of developmental evolution in processes giving rise to speciation itself is controversial (Wilkins 2001), morphological diversity is a demonstrated consequence of speciation. Accompanying the morphological changes that result from changes in the regulation of developmental genes—changes that, in part, enable organisms to insinuate themselves into a wide variety of ecological niches—are physiological changes, that while perhaps less visible to the naked eye, are of equal importance, and are also reflective of underlying regulatory changes. As Hochachka and Somero, commenting on the consequences of organizational complexity of organisms, observe:
Here we emphasized that the requirements for intercellular and interorgan coordination in complex metazoans create regulatory challenges not faced by unicellular species. These requirements for complex integration and coordination of biochemical adaptations appear to have played major selective roles in the proliferation of genes encoding proteins that play regulatory roles, for instance, protein kinases and phosphatases. The question, “Why do complex organisms contain so much DNA?” may be answered in part by the complexity entailed in regulating the adaptive responses found in metazoans. [(Hochachka and Somero 2002) p18]
In the course of this chapter, we have explored, using a wide variety of contemporary sources, the issue of similarities and differences between organisms in the context of our current understanding of evolutionary developmental biology. The similarities reflect profound evidence of descent from common ancestors, whereas the morphological and even biochemical differences between organisms reflect, in no small measure, regulatory modifications brought about by a variety of evolutionary mechanisms that affect the evolutionary trajectories taken by divergent lineages.
Hochachka and Somero draw our attention to the curious phenomenon (also seen in our discussion of morphological evolution) whereby differences between organisms with respect to adaptive biochemical modification are evident notwithstanding the existence of highly conserved biochemical elements (something normally discussed under the heading of the unity of biochemistry—a matter to be touched on in the next chapter). In the present context, differential biochemical adaptations seen in different organisms reflect regulatory differences that ensure the activation of appropriate genes to effect the adaptive changes in question (Hochachka and Somero, 2002:18). Another case, in fact, of genetic unity in the face of adaptive diversity.
What are we to make of this observation? Are the similarities so great and pervasive, and the differences so minor and superficial, to support biomedical inferences from test results in members of one species to the formation of reliable expectations about the effects of similar causes as applied to members of another species? In other words, do we have any good reason to suppose that animal models (whatever else they may be useful for in science) are predictive of human responses? We are not yet ready to discuss this, but the present chapter raises issues that prompt these questions.
It is an enduring theme in contemporary evolutionary biology that the evolution of metazoans proceeds in large measure through the re-use and modification of existing parts. Intuitively, these modifications, reflected in differential patterns of regulatory action in distinct lineages, ought to be relevant to a discussion of the prediction question with which we are concerned in this book. However, since many things deemed intuitively plausible turn out to be illusory, it behooves us to examine in more detail the nature of inter-individual differences as it applies to both intraspecific and interspecific variation. Consistent with our interest in events occurring in the lifecycles of organisms, our inquiries begin with a discussion of delegated complexity and its implications for biological individuality. We then turn to issues raised in the context of the metabolism of xenobiotics and their implications for the issue of prediction in the biomedical sciences.