The effect of progesterone supplementation and Cox2 specific inhibitors on labor associated changes in biochemical markers as the end-point. Here we used a systems biology approach to test whether the presence of subclinical infection/inflammation would modulate progesterone or Cox2 inhibitor effect on the generation of a PRA to PRB ratio associated with labor. In order to do this we developed a dynamiccomputer simulation of myometrium at the molecular level using differential equations. We observed that only at the highest concentrations do progesterone and Cox2 specific inhibitors prevent the NF-kB induced PRA/PRB increase. In the case of progesterone there have been recent clinical data that support this observation. In the case of Cox2 inhibition, high concentrations of Cox2 inhibitor may have increased fetal toxicity. System Biology seeks to use mathematical modeling to integrate currently available genomic, proteomic, in vitro and in vivo data into functional models of biological systems. An 
effective model of a complex system has a number of potential benefits, notably it may be possible to use the model to predict the behavior of the system when disturbed by pathology or the response of the system to a therapeutic. The simulations are especially valuable to answer questions that cannot be readily tested, such as new treatments of preterm delivery. The regulation of human parturition is demonstrably different in many ways from that in other mammals. In particular, in most mammals parturition PF-4217903 c-Met inhibitor follows a rapid fall in circulating maternal concentrations of progesterone while in the human circulating progesterone levels show no signs of falling until after removal of the placenta. The consequence of the inter-species differences is that animal studies give only limited insight into the mechanisms of human labor. Experimental studies are also problematic in the setting of human labor for Cycloheximide ethical reasons. In this manuscript we have started to develop a model of the molecular events occurring in the human myometrial cell as it transitions at term from non-laboring to the laboring state. Data was obtained from the literature on the perceived critical variables. In this context the critical factors were considered to be the concentrations of progesterone receptors and estrogen receptors and associated factors. To generate the model a number of explicit assumptions were made where clinical or in vitro data were unavailable. These assumptions are described in the Methods section. The model was designed in a bottom-up fashion. Every change to a molecular species, interaction between two or more species, transportation of a species from one compartment to another, transcription and translation is counted as a reaction. The model includes 199 different molecules, 208 reactions, and 624 kinetic parameters. The model was designed such that activation of NF-kB led to an increase in PRA/PRB ratio to labor levels, reflecting the observation that infection/inflammation is a well known risk factor for preterm delivery. We have then explored how the model responds to a potential tocolytic in the form of a Cox2 inhibitor or progesterone, in the presence of subclinical infection/inflammation. We observed that neither a 10 fold increase in progesterone receptor nor a 2 fold increase in Cox 2 inhibition were effective in preventing the PRA/PRB increase at levels of NF-kB activation that might occur during subclinical infection. These results parallel a recent double-blind, placebo controlled human trial where treatment with a selective Cox2 inhibitor did not reduce the incidence of early preterm delivery.