The smooth muscle of the human myometrium remains relatively quiescent throughout most of pregnancy, but it undergoes a transformation at term that results in the development of powerful rhythmic contractions. These painful contractions, under normal circumstances, give rise to the birth of a bouncing baby boy or girl, but in pathological conditions, premature, delayed or abnormal contractions can threaten the health of both the mother and child. The mechanisms responsible for the myometrial transformation have been the subject of a great deal of research, much of which has focused on global changes in gene expression, leading to changes in the components of the smooth muscle cells that control their electrical excitability and hence their propensity to produce spontaneous contractions. A number of recent studies have used a promising functional genomics approach, taking advantage of DNA microchip technology to screen among tens of thousands of genes whose expression in the human myometrium might increase or decrease during the transition to labour, and therefore perhaps account for the physiological transformation from quiescent to contracting smooth muscle. Although this approach has the apparent advantage of being unbiased by any preconceived notion of the underlying mechanisms, it also suffers from a number of disadvantages. The latter include its cost, small sample size, non-homogeneous sample collection, and the complexity of measuring thousands of transcripts in a single patient sample that may result in both false positive and false negative interpretations (Breuiller-Fouche et al. 2007). Despite these limitations, results obtained using this approach have suggested that relatively few changes in gene expression occur during the transition to labour, compared with the number of changes detected among earlier stages of human pregnancy (Breuiller-Fouche et al. 2007). The use of animal models to investigate changes in myometrial smooth muscle cell phenotype during pregnancy has the advantage that homogeneous tissue samples collected at precise time points and tissue locations can be used for both functional and genetic screening. The article by Greenwood et al. in this issue of The Journal of Physiology (Greenwood et al. 2009) illustrates the utility of this approach, using a mouse model to provide more mechanistic insights into the cellular components that regulate smooth muscle contractility in the pregnant myometrium. The Greenwood et al. study addresses the nature of the ionic conductances involved and molecular mechanisms whereby these ionic conductances change during the transition from quiescent to contracting murine myometrium. Several previous studies had used rat or mouse models to implicate voltage-dependent potassium channels in the regulation of myometrial contractility (Song et al. 2001; Smith et al. 2007). A decrease in potassium current, whether due to reduced channel expression or conductance, would be expected to increase electrical excitability and contractility. From the super-family of mammalian voltage-dependent potassium channels, members of the Kv1, Kv4 and Kv7 subfamilies were found to be expressed in myometrial smooth muscle of rat or mouse (Song et al. 2001; Smith et al. 2007; McCallum et al. 2009). Of these, only expression of Kv4.3 was found to significantly diminish during pregnancy (Song et al. 2001; Smith et al. 2007). Greenwood et al. demonstrate for the first time expression and function of another subfamily of voltage-dependent potassium channels (erg; Kv11) in myometrial myocytes. Expression of erg1a and 1b (but not erg2 or erg3) was detected in murine myometrial smooth muscle. Furthermore, pharmacological treatments suggested that erg channels contribute to maintaining the relatively quiescent contractile state of non-pregnant myometrium, but the contribution of erg is apparently lost in late pregnancy. Electrophysiological measurements also suggested that a reduction of erg channel activity occurs during late pregnancy, when there is a corresponding increase in spontaneous myometrial contractility. Interestingly the decreased potassium conductance may result not from decreased expression of the pore-forming erg1 α-subunit, but rather from increased expression of regulatory β-subunits (KCNE2). Greenwood et al. found no change in erg1 expression (mRNA or protein) during pregnancy, but KCNE2 (MiRP1), which is known to suppress human Kv11.1 (HERG) currents, was dramatically upregulated in late pregnant myometrium. The Greenwood et al. study highlights the utility of an animal model, combining molecular expression measurements with functional studies of contractility and ionic currents at defined time points during pregnancy. However, there are some known discrepancies between human and rodent studies of myometrial excitation–contraction coupling (Shmygol et al. 2007), raising the question as to whether mechanisms identified using rodent models will ultimately be applicable to human myometrium. It remains to be determined whether changes in ion channel expression and function, comparable to those elucidated in murine myometrium, occur in the human myometrium at the onset of labour. The next step may then be to apply what has to date been a somewhat descriptive functional genomics approach in a more directed way to evaluate the proposed role of KCNE upregulation in late pregnancy in humans. Other likely candidate channels (e.g. Kv7.5 (KCNQ5)) (McCallum et al. 2009) identified in rodent models might also be specifically examined in human myometrial samples to more effectively translate the mechanism-focused animal studies to elucidate human physiology, examining to what extent the same mechanisms are operating and potentially identifying novel targets for human therapies.