The Application of Metagenomic and Bioinformatics approach in Maternal and Children's Health

Abstract

The human microbiota, the vast consortium of microorganisms inhabiting our bodies, is increasingly recognized as a central regulator of physiological homeostasis and a critical determinant of health and disease. This is particularly true during sensitive developmental windows such as pregnancy, infancy, and childhood, where the establishment and maturation of these microbial communities can have profound and lasting impacts. Moving beyond simple descriptive cataloging, this thesis presents a comprehensive investigation into the intricate and dynamic roles of microbial ecosystems in maternal and child health. By employing an integrative, multi-omics framework that combines high-resolution metagenomics, metabolomics, and advanced bioinformatics, this work dissects complex host-microbe interactions and pathogen evolution across diverse clinical cohorts. The primary objective is to translate high-dimensional biological data into mechanistic insights that can inform the development of novel diagnostics, targeted interventions, and precision medicine strategies to improve clinical outcomes across the maternal-child health continuum.

Our investigation begins with the microecology of the female reproductive tract, a niche where microbial balance is paramount for health. We provide robust evidence demonstrating that the composition of the cervicovaginal microbiota is a significant predictor of outcomes in high-risk human papillomavirus (hrHPV) infection. Specifically, communities dominated by Lactobacillus crispatus are strongly and consistently associated with the spontaneous clearance of hrHPV. In stark contrast, dysbiotic microbial states, characterized by a decrease in Lactobacillus abundance and an enrichment of anaerobic bacteria such as Gardnerella, Prevotella, and Sneathia, are significantly correlated with persistent hrHPV infection and the subsequent progression to high-grade cervical intraepithelial neoplasia (CIN). Delving deeper into the functional consequences of these microbial shifts, we analyzed the vaginal microenvironment following surgical treatment for CIN. This analysis revealed a pivotal finding: the vaginal metabolome, particularly dynamic shifts in biogenic amines and phospholipid derivatives, serves as a more consistent and sensitive indicator of micro-environmental recovery and HPV clearance than microbial composition alone. This underscores the profound value of functional readouts in assessing ecosystem health. Furthermore, we extend our investigation to endometriosis, providing evidence for the translocation of specific bacteria like Fannyhessea and Sneathia from the lower to the upper reproductive tract, linking their presence to the up-regulation of functional pathways involved in bacterial motility and inflammation.

The thesis then transitions to the crucial maternal–infant axis, a period defined by the vertical transmission and initial establishment of the neonatal microbiome. We characterize the profound shifts that occur in the maternal gut microbiota during pregnancy and identify distinct microbial signatures in women with gestational diabetes mellitus (GDM). In these women, the enrichment of Ruminococcus gnavus and Bacteroides vulgatus is significantly correlated with the upregulation of microbial pathways for lipopolysaccharide (LPS) biosynthesis and host pathways related to insulin resistance, suggesting a direct microbial contribution to metabolic dysregulation. We also investigate the impact of common clinical interventions, revealing that perinatal antibiotic exposure leads to a dramatic depletion of maternal vaginal Lactobacillus populations, altering vertical microbial transmission with potential implications for neonatal sepsis risk. Importantly, our research highlights a critical modifiable factor: exclusive breastfeeding plays a remarkable restorative role, particularly for infants delivered by Cesarean section. Breastfeeding actively enriches the infant gut with beneficial Bifidobacterium species, which is significantly associated with a reduced incidence of respiratory infections and diarrhea during the first year of life.

In pediatric populations, this work uncovers the multifaceted roles of the microbiota in a range of childhood diseases. In respiratory infections, we identify pathogen-specific dysbiosis patterns: Mycoplasma pneumoniae infection is linked to reduced diversity and Staphylococcus dominance, whereas influenza virus infection correlates with increased diversity and Prevotella enrichment. Extending our analysis to neurological and hepatic conditions, we demonstrate that the baseline gut microbiota can predict clinical outcomes, such as the response to the ketogenic diet in West Syndrome and jaundice clearance after Kasai portoenterostomy in biliary atresia. Furthermore, in a multi-center, randomized controlled trial on antibiotic-associated diarrhea (AAD), we establish that a child's baseline gut enterotype is a key determinant of both the microbiome's stability under antibiotic pressure and its responsiveness to a probiotic intervention. This finding strongly advocates for personalized, microbiome-informed approaches to probiotic therapy.

In conclusion, this thesis establishes a robust and multifaceted evidence base for the profound clinical relevance of microorganisms in women's and children's health. By systematically bridging high-resolution multi-omics data with sophisticated computational methods, our findings move beyond association to uncover the functional roles of microbial communities and the evolutionary dynamics of critical pathogens. The insights generated—from identifying predictive microbiome signatures in HPV and GDM to understanding pathogensis—pave the way for a new generation of clinical tools. These include microbiome-informed diagnostics for early risk stratification, targeted interventions like personalized probiotics, and advanced genomic surveillance strategies, all aimed at modulating host-microbe interactions to prevent disease and improve health outcomes across the entire maternal–child continuum.

Date of Award	15 Apr 2026
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Shuaicheng LI (Supervisor)