Human-associated microorganisms (the “microbiome”) are present in numbers exceeding the quantities of human cells by at least 10-fold, and the collective genome (the “metagenome”) exceeds our human genome in terms of gene content by more than 150-fold. We and others have recently demonstrated that bacteria are detected in the placenta using a variety of culturable and non-culturable approaches. For several years we have developed and employed metagenomics to characterize the placental microbiome, and observed variation in its community membership and their function by virtue of gestational age at delivery. However, it remains unknown what the different maternal source(s) are of the placental microbiome, and whether it sinks to the fetus via the intra-amniotic cavity and thus is measureable in amniotic fluid. In response to the Human Placental Project, we propose to identify the sources and sinks for and of the placental microbiome employing two large, robust and well characterized existing datasets. We will generate unparalleled metagenomics and metabolomics data, in order to test our central hypothesis that the placenta is populated by commensal microbiota which largely arise from the maternal oral and GI communities, with a lesser contribution from the vagina. We further hypothesize that placental microbes populate the fetus and the intrauterine environment, and are detectable in mid-trimester amniotic fluid. Moreover, their metabolites serve as lasting signatures of the microbiotas functional presence. In order to prove this hypothesis, we will execute three essential aims in a total of over 1230 subjects samples from two existing data sets. The net result of the completion of these aims will be to first identify the maternal source of the placental microbiota, and validate these findings using state of the art single molecule fluorescent in situ hybridization and culturation. We will thereafter recapitulate these findings in early and mid- second trimester amniotic fluid and thus identify the early evidence of the placental microbiota sink. In a final aim, we will use LC/MS full spectral metabolomics on these same subjects samples to identify the stable and lasting metabolic footprint of the microbiome. We present our proof of concept work on intrahepatic cholestasis of pregnancy as evidence for the feasibility, significance and ready translational application of our approach. As a proven team of perinatal physician scientists with an emphasis and history of being at the forefront of big data (and notably metagenomics) science, we are uniquely poised to now undertake complex integration of these unique data sets in studies which are feasible, justifiable, and of likely long-term significance and high impact.