Exploring Maternal Epigenomic and Metabolomic Influences on Gestational Age at Birth: Insights from Multi-Omics Integration and Mediation Analysis of PFAS Exposure Restricted; Files Only

Abstract

The overarching goal of this dissertation was to better understand the biological pathways that influence gestational age at birth through the application of novel multi-omics methods and an improved understanding of high-dimensional mediation methods to further the understanding of how changes in biological pathways associated with per- and polyfluoroalkyl substances (PFAS) exposure mediate the association between exposure to PFAS and gestational age at birth (GA).

 

Aim 1: Understanding the biological mechanisms underpinning preterm birth is key to improving maternal and neonatal outcomes. Using epigenetic and metabolomic data from the Atlanta African American Maternal Child Cohort, we integrated and identified highly correlated molecular signatures and pathways underlying preterm birth. Functional genes and metabolites identified across early and late pregnancy underscore roles in immune function, energy regulation, and lipid metabolism, revealing novel insights into molecular networks indicative of preterm birth. 

 

Aim 2: This study aimed to test several high-dimensional mediation analysis techniques using simulated metabolomics data to understand the functionality and performance of methods (including HIMA, HDMA, and “Meet-in-the-Middle” [MITM] approaches). Using simulated data, we examined the total indirect effect (TIE) through all metabolic features, the component indirect effect (CIE) through individual features, and the sensitivity and specificity of the captured metabolites. Overall, HIMA showed the strongest performance, but its complexity makes it harder to implement and interpret than MITM, suggesting that researchers may benefit from using multiple methods in high-dimensional mediation analyses.

 

Aim 3: PFAS are persistent environmental toxicants. Epidemiological studies have linked PFAS exposure to reduced GA. Maternal metabolic perturbations may mediate this association. Using data from the Atlanta African American Maternal Child Cohort, parallel methodologies (HIMA and MITM) were applied to identify the serum metabolomic signature underpinning the association between PFAS and GA. Metabolite mediators of the PFAS-GA association were closely involved in the tricarboxylic acid (TCA) cycle, bile acid biosynthesis, steroid hormone biosynthesis, and purine metabolism. 

 

Conclusion: This dissertation expanded our knowledge of the molecular mechanisms influencing preterm birth, improved our understanding of high-dimensional data analysis within metabolomics data, and provided novel insights on how biological perturbations from PFAS mediate the impact between PFAS and GA.

Table of Contents

Contents

Distribution Agreement

Abstract

Acknowledgments

Table of Figures

Table of Tables

Introduction

Data source

Aim 1

Abstract

Background

Methods

Study cohort

Data collection methods

Epigenomics and Metabolomics Profiling

Outcome ascertainment

Statistical analysis

Sensitivity analyses

Results

Study population

EWAS and MWAS

Multi-Omic analyses

Sensitivity analyses

Discussion

Conclusion

Figures

Tables

Supplementary Information

References

Aim 2

Abstract

Background

Methods

High-dimensional mediation analysis

Data generation

Simulation scenarios

Analysis

Results

CIE

TIE

Sensitivity and specificity

Discussion

Conclusion

Figures

Supplementary information

References

Aim 3

Abstract

Background

Methods

Study population

Biological sample collection and Exposure assessment

Metabolomics profiling

Outcome assessment

Statistical analysis

Results

Study population and exposure levels

PFAS exposure levels and birth outcomes

PFAS exposure, maternal metabolome, and birth outcomes

Discussion

Conclusion

Tables

Figures

Supplementary information

References

Summary of results and future directions

Summary of results

Aim 1

Aim 2

Aim 3

Future directions

References

About this Dissertation

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Epidemiology
Degree	Ph.D.
Submission	Dissertation
Language	English
Research Field	Health Sciences, Epidemiology Environmental Health
Keyword	PFAS Gestational age at birth Multiomics Metabolomics
Committee Chair / Thesis Advisor	Michele Marcus , Emory University Donghai Liang , Emory University
Committee Members	Todd Everson , Emory University Ashley Naimi, Emory University Anne Dunlop , Emory University Audrey Gaskins, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	File download under embargo until 22 May 2027	2025-04-03 09:38:50 -0400	File download under embargo until 22 May 2027

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions