PROJECT SUMMARY Perinatal depression (PND), defined as depression during pregnancy and up to one year postpartum, affects more than 20% of pregnancies and disproportionately impacts Black Women and Latinas. PND increases risk of preterm birth and infant neurodevelopment deficits. Yet we still do not fully understand the pathobiology of PND, which limits efforts to improve its prevention, identification and treatment. Interactions between the host and microbial communities that reside in the gut are essential for human health. Gut microorganisms play an important role in producing beneficial metabolites, including the neurotransmitters serotonin and gamma- aminobutyric acid (GABA). The gut microbiota bidirectionally communicate with the brain, an interaction mediated by the neurological, immune, and endocrine systems and coined as the microbiota-gut-brain axis (MGBA). Our initial pilot results from a longitudinal study of low-income women of color (n=42) early in pregnancy, which showed that multiple attributes of the MGBA were associated with depressive symptom severity, including production of short chain fatty-acids, metabolism of tryptophan and GABA, and systemic inflammation mediated by bile acid metabolism. Although our initial data points to new MGBA signatures linked to depressive symptom severity in the first and second trimesters, additional work is needed to determine whether these signatures and their interactions extend beyond the second trimester. Further, microbial metabolism is driven both by the interactions between metabolites and by the interplay between microbial metabolic systems with maternal inflammatory system and metabolism. Thus, determining the causal influence of these new MGBA signatures on depressive symptom severity during the perinatal period requires using approaches that can establish the effect of systems (networks) coupling in MGBA functioning. Here, we propose a clinical translational study to determine the role of MGBA in PND by using a systems biology framework and an experimental animal model to assess causality. To test this, we will draw from our research infrastructure to recruit 158 women (55% Black, 30% Latina) early in pregnancy (<16 gestational weeks) and follow them bimonthly for up to 6 weeks postpartum. We will assess mood; lifestyle (diet, physical activity, sleep); and heart-rate variability (a proxy for stress) and will measure microbial genome and meta-metabolome and maternal blood transcriptome and metabolome. In Aim 1, we will employ interpretable machine learning models to predict depressive symptom severity concurrently and prospectively. In Aim 2, we will establish coupling mechanisms that regulate symptom severity by modeling the interplay between microbial and maternal metabolic, genetic and regulatory systems using network theory. In Aim 3, we will determine the causal role of gut microbiota in symptom severity in a female pregnant germ-free mouse model using fecal microbiota transplant...