Deciphering the role of the vaginal microbiome in spontaneous preterm birth via consideration of host genetics

/Abstract In the U.S, preterm birth affects 1 in 10 pregnancies, and about half of preterm births are spontaneous as opposed to clinically-indicated inductions or cesarean sections. Despite the grave and widespread threat posed by spontaneous preterm birth (sPTB), its etiology is not fully understood, and diagnostic and therapeutic tools remain limited. Research has independently linked human genetics and vaginal microbes to preterm birth, but no study has explored links between all three entities. Furthermore, research on the associations between genetics and the vaginal microbiome has been constrained by small sample sizes, heterogeneous sample populations, and the use of 16S rRNA sequencing, which cannot classify bacterial taxa at the subspecies level. To define sPTB risk factors and develop effective diagnostics and therapeutics, we must gain a high-resolution understanding of microbial and genetic contributions to prematurity, including causal effects. The objective of this proposal is to investigate associations between host genetics, the vaginal microbiome, and sPTB in a large, diverse cohort with metagenomic sequencing. The nuMoM2b study collected genotyping, vaginal swabs, and extensive clinical data on >10,000 women from eight sites across the U.S. I will detect sample processing errors in this cohort by comparing genotypes inferred from metagenomic sequencing of vaginal swabs to those independently obtained in chip genotyping. I will generalize this method such that it may be applied to any study with both genotyping and metagenomic sequencing. After using my method for quality control in the nuMoM2b cohort, I will conduct Genome Wide Association Studies to detect genetic variants associated with the microbiome. I will investigate variants associated with microbiome characteristics imperceptible by 16S sequencing, such as the relative abundances of bacterial strains or their genomic profiles. I will then utilize microbiome-associated genetic variants in Mendelian Randomization analysis to probe causation of the microbiome on sPTB. Finally, I will train a predictive model on both genetics and microbiome data and benchmark it against models trained on either data source alone. This project will investigate causal effects of microbiome characteristics on sPTB, thus elucidating sPTB’s etiology and providing targets for novel interventions. The machine learning model will demonstrate the potential of diagnostic tools that predict sPTB risk from vaginal microbiota and genetics, which can be easily collected via vaginal swabs and blood testing, respectively. At Columbia, I have access to the facilities, equipment, and mentorship necessary to complete the proposed work. The F31 Ruth L. Kirschstein NRSA will support completion of this specific project while broadly encouraging my academic and professional development, including my progress towards a career as an independent investigator using computational tools for the study of reproductive biology. Project Number: 1F31HD115394-01A1 | Fiscal Year: 2024 | NIH Institute/Center: Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) | Principal Investigator: Julia Urban | Institution: COLUMBIA UNIVERSITY HEALTH SCIENCES, NEW YORK, NY | Award Amount: $48,974 | Activity Code: F31 | Study Section: Special Emphasis Panel[ZRG1 F08-L (20)] View on NIH RePORTER: https://reporter.nih.gov/project-details/1F31HD11539401A1