Conceptual representation of a genetic cis‐regulatory domain. The gene is represented by a horizontal line and the transcription start site by a bent open arrow. Letter sequences below the line represent specific sites of protein–DNA interaction, while shapes above the line represent proteins bound at those sites. Curved lines terminated in either arrows or bars represent positive and negative regulatory functions, respectively. The activity of the basal transcription apparatus (BTA) is positively affected by the proteins bound at sites TGTGGT and CCAAT, and negatively affected by the protein bound at site TAACGG. The proteins bound at sites GGGG interact with each other, causing the DNA to loop and thereby facilitating the interactions between the other regulatory proteins and the BTA. While the cartoon is imaginary, the sites and their functions are based on known examples.

Computational model for endo16 module A regulatory functions. (a) Schematic diagram of interrelations and functions. Interrelations between upstream modules G–B and specific module A target sites are indicated beneath the line representing the DNA. The region from module G to module B is not to scale. Each circle or node represents the locus in the system of a specific quantitative operation, conditional on the state of the system; operations are specified for all relevant states in (b). Operations at each node are carried out on inputs designated by the arrows incident on each circle, and produce outputs designated by arrows emergent from each circle. Open arrowheads indicate inputs to the indicated node that are constant through time, the values of which are specified according to the logic sequence in (b); closed arrowheads indicate time‐varying inputs. The terminated bar indicates a boolean repression function that under given conditions extinguishes activity at node h. Solid line, time‐varying influence; dashed line, scalar factor; dotted line, inhibitory switch. (b) Logic sequence for operation of model shown in (a). The value 0 denotes that a given site or module site has been mutationally destroyed or is inactive because its factor (or factors) is missing or inactive; the value 1 indicates that the site or module is present and productively occupied by its cognate transcription factor. For the case of modules F, E and DC, a boolean representation is chosen because ectopic expression is essentially zero when these modules (together with module A) are present; otherwise, ectopic expression occurs. Sites within module A are designated as above. The logic sequence specifies the values attained at each operation locus (circles in (a)), either as constants determined experimentally and conditional on the state of the relevant portions of the system, or in terms of time‐varying, continuous inputs designated by the symbol (t). The final output, q(t), can be thought of as the factor by which, at any point in time, the endogenous transcriptional activity of the BTA is multiplied as a result of the interactions mediated by the cis‐regulatory control system. Reprinted with permission from Yuh C‐H, Bolouri H and Davidson EH (1998) Genomic cis‐regulatory logic: experimental and computational analysis of a sea urchin gene. Science279: 1896–1902. Copyright 1998 American Association for the Advancement of Science.

A sector of an imaginary developmental gene regulatory network. (a) Network sector. Three genes encoding transcription factors are shown at the top. These are a spatial regulatory (orange), a temporal regulatory (green) and a signal‐mediated regulator (blue). Genes encoding other transcription factors that originate off the diagram are indicated in black type on open backgrounds. A battery of six genes encoding some differentiation proteins (P1–P6) is shown below. Connections between the three genes are indicated by respectively coloured bent arrows and the transcription factors as solid circles. The spatial regulatory gene is controlled by positive and negative interactions, which establish the limited spatial domain where it will be expressed, and it utilizes a ubiquitous ancillary activator to achieve an appropriate level of expression. This gene would be expressed only at certain stages due to requirement for the factor produced by the green temporal regulator, shown below the line binding to its target sites in the cis‐regulatory DNA. The cis‐regulatory system of the temporal regulator responds to its own transcription factor, and also depends on a factor appearing only after a certain stage of development, and on another ubiquitous ancillary activator. The signal‐mediated regulator produces a factor that is activated by signals. For example, if this were a short‐range signal produced by cells adjacent to the domain of expression of the spatial orange regulator, P1–P6 would be expressed only near the boundary. The cis‐regulatory system controlling expression of the signal‐mediated transcription factor includes target sites for the product of the orange spatial regulator, shown binding below the line representing the DNA, and also for two factors that work together to promote transcription during growth, one imagined as a regulator produced when cells are cycling, the other as a ubiquitous cofactor. The arrows at the right indicate that each of the three genes encoding transcription factors have many downstream targets besides the P1–P6 gene battery. Any resemblance between this network sector and known regulatory element is purely coincidental. (b) A single relationship extracted from the network. A causal diagram is shown portraying the multilevel function of the orange spatial regulator, which controls both the gene encoding the blue signal‐mediated regulator and the P1 gene; the latter, however, is also directly responsive to the spatial regulator. Reprinted with permission from Arnone MI and Davidson EH (1997) The hardwiring of development: organization and function of genomic regulatory systems. Development124: 1851–1864. Copyright 1997 Company of Biologists Ltd.

Two well‐defined developmental gene regulatory networks. Genes are represented by short horizontal coloured lines that terminate in bent arrows, which represent the gene's expression (transcription and translation). Connections leading from each gene to other genes in the network are represented by lines of the same colour. Inputs are positive (activation) if they end in arrows; negative (repression) if they end in bars. (a) Sea urchin endomesoderm network. The network is continuously updated at http://sugp.caltech.edu/endomes/. Genes in the bottom rectangles encode differentiation proteins, of which only small samples are included; most remaining genes in the network encode transcription factors; a few encode components of signalling systems, as indicated by the names beneath the genes. From left to right the areas of the diagram represent interactions occurring in the skeletogenic domain (lavender); in the endoderm and remainder of the mesodermal cell types (light green area; genes on blue backgrounds function ultimately in mesoderm, those on yellow backgrounds in endoderm) and in the late invaginating ‘veg1’ endoderm (orange background). At the top are relevant maternal inputs (maternal mRNA or protein). Small open circles in this diagram represent cytoplasmic interactions, e.g. biochemical transactions among signal transduction pathway components, and small solid black circles represent junctions, where multiple sources of given inputs merge. In addition, the diagram indicates nodes at which the cis‐regulatory modules executing the functions specified in the network are known (red dots), and the role of individual target sites has been established by mutation (green dots). (b) Fruitfly dorsal‐ventral patterning network. On left (pink area) is a summary of interactions leading to the ventral‐to‐dorsal gradient of nuclearized Dorsal transcription factor within the syncytial embryo. These interactions begin during oogenesis in follicle cells, and following expression of the pipe gene in the ventral follicle cells, they occur in the perivitelline space between the egg membrane and the chorion. Their result is the proteolytic activation of the Spatzle (Spz) ligand, which is deposited in a graded cline on the ventral‐to‐lateral surface of the egg, and results in graded activation of the Toll receptor, generating the Dorsal nuclearization function. The central region of the network (light green, dorsal ectoderm and dorsal neurogenic ectoderm; light yellow, ventral neurogenic ectoderm; light blue, ventral, mesodermal territory) show regulatory interactions among genes responding to the Dorsal input during the syncytial phase of development, from about 2 to 3 h postfertilization. On the right (darker green, dorsal epidermis, dorsal neurogenic and amnioserosal territories; tan, ventral epidermis and ventral neurogenic territories; darker blue, dorsal mesoderm, here heart, territory) are shown regulatory interactions occurring after cellularization and into gastrulation, up to about 5 h postfertilization. Reproduced with the permision from The Regulatory Genome, by Eric Davidson Copyright Elsevier (2006).

Examples of GRN subcircuits. In each the output is an alteration of regulatory state, as indicated in black at the bottom of each panel, and in each case the biological consequence is to accomplish a given developmental ‘job’, as indicated in red. Genes (designate by capital letters) and their interactions are depicted and colour‐coded as in Figure . All genes in these diagrams encode transcription factors. The outputs of these genes, i.e. the protein factors translated from their mRNA, are inputs to the cis‐regulatory modules of other genes in the subcircuits, depicted as in the actual GRNs shown in Figure . Initial inputs from outside the subcircuits, signals from other cells (S) and subcircuit outputs, are shown in black. (a) Subdivision of spatial domains by subcircuit AND logic; the circle containing ‘+’ is an example of a logic function executed by the cis‐regulatory module, here an ‘AND’ logic operation. (b) Transformation of transient spatial regulatory input into stable regulatory state; (c) Eventual specification of confined spatial expression beginning with broad domain; (d) Exclusion of alternative regulatory state on specification; (e) Spatial and temporal peak of expression. Reproduced with permision from The Regulatory Genome, by Eric Davidson Copyright Elsevier (2006).