The process of parturition involves the transformation of the quiescent myometrium (uterine smooth muscle) to the highly contractile laboring state. This is thought to be driven by changes in gene expression in myometrial cells. Despite the existence of multiple myometrial gene expression studies, the transcriptional programs that initiate labor are not known. Here, we integrated three transcriptome datasets, one novel (NCBI Gene Expression Ominibus: GSE80172) and two existing, to characterize the gene expression changes in myometrium associated with the onset of labor at term. Computational analyses including classification, singular value decomposition, pathway enrichment, and network inference were applied to individual and combined datasets. Outcomes across studies were integrated with multiple protein and pathway databases to build a myometrial parturition signaling network. A high-confidence (significant across all studies) set of 126 labor genes were identified and machine learning models exhibited high reproducibility between studies. Labor signatures included both known (interleukins, cytokines) and unknown (apoptosis, MYC, cell proliferation/differentiation) pathways while cyclic AMP signaling and muscle relaxation were associated with non-labor. These signatures accurately classified and characterized the stages of labor. The data-derived parturition signaling networks provide new genes/signaling interactions to understand phenotype-specific processes and aid in future studies of parturition.