Integrating Multi-Omics with environmental data for precision health: A novel analytic framework and case study on prenatal mercury induced childhood fatty liver disease

Abstract

Background: Precision Health aims to revolutionize disease prevention by leveraging information across multiple omic datasets (multi-omics). However, existing methods generally do not consider personalized environmental risk factors (e.g., environmental pollutants).

Objective: To develop and apply a precision health framework which combines multiomic integration (including early, intermediate, and late integration, representing sequential stages at which omics layers are combined for modeling) with mediation approaches (including high-dimensional mediation to identify biomarkers, mediation with latent factors to identify pathways, and integrated/quasi-mediation to identify high-risk subpopulations) to identify novel biomarkers of prenatal mercury induced metabolic dysfunction-associated fatty liver disease (MAFLD), elucidate molecular pathways linking prenatal mercury with MAFLD in children, and identify high-risk children based on integrated exposure and multiomics data.

Methods: This prospective cohort study used data from 420 mother-child pairs from the Human Early Life Exposome (HELIX) project. Mercury concentrations were determined in maternal or cord blood from pregnancy. Cytokeratin 18 (CK-18; a MAFLD biomarker) and five omics layers (DNA Methylation, gene transcription, microRNA, proteins, and metabolites) were measured in blood in childhood (age 6-10 years).

Results: Each standard deviation increase in prenatal mercury was associated with a 0.11 [95% confidence interval: 0.02-0.21] standard deviation increase in CK-18. High dimensional mediation analysis identified 10 biomarkers linking prenatal mercury and CK-18, including six CpG sites and four transcripts. Mediation with latent factors identified molecular pathways linking mercury and MAFLD, including altered cytokine signaling and hepatic stellate cell activation. Integrated/quasi-mediation identified high risk subgroups of children based on unique combinations of exposure levels, omics profiles (driven by epigenetic markers), and MAFLD.

Conclusions: Prenatal mercury exposure is associated with elevated liver enzymes in childhood, likely through alterations in DNA methylation and gene expression. Our analytic framework can be applied across many different fields and serve as a resource to help guide future precision health investigations.

Keywords: Bioinformatics; Biomarkers; Epigenetics; Multiomics; Precision health; Prenatal exposures.

Fig. 1.

Conceptual diagram illustrating the analytic framework for mediation analysis with multiple omic layers.

Fig. 2.

Heatmap illustrating the correlation of molecular features within and between different omics layers. The top 28 features per omic layer were selected for inclusion in the heatmap using a modified version of sure independence screening with prenatal mercury and childhood CK-18.

Fig. 3.

High dimensional mediation analysis with multiple omic layers identifies individual molecular features linking prenatal mercury with childhood liver injury. Panels A, B, and C present the results for early integration (described in Fig. 1A), intermediate integration (described in Fig. 1B), and late integration (described in Fig. 1C), respectively, with each column representing a single omics feature. Alpha represents the coefficient estimates of the exposure to the mediator, Beta indicates the coefficient estimates of the mediators to the outcome, and TE (%) represents the percent total effect mediated calculated as alpha*beta/gamma. The triangular heatmap at the bottom of the figure shows the correlation between individual omics features.

Fig. 4.

Mediation analysis with latent factors identifies alterations in critical molecular pathways linking prenatal mercury exposure with childhood CK-18 (Fig. 1, Column 2). Panel A illustrates the mediation effects for each of the three different omics integration approaches (early, intermediate, and late), where alpha is the coefficient estimate of the exposure to the mediator, beta is the coefficient estimate of the mediators to the outcome (PCs to childhood liver enzymes), and % TE indicates the percent of total effect mediated, calculated as alpha*beta/total effect and scaled to 100%. Each column represents either a joint variance component (describing common variance across all omics) or an individual variance component (describing variance specific to that omic layer). Panel B illustrates the top three molecular pathways associated with each individual or joint variance component.

Fig. 5.

Associations of individual omic features with joint and individual variance components mediating the association of prenatal mercury exposure with childhood MAFLD were determined using mediation analysis with latent factors (Fig. 1, Column 2). Panels A-C represent the correlation of each omic feature with each variance component calculated using three different omics integration approaches. For omics specific components (i.e., methylome components and transcriptome components), grey colors indicate omics features not associated with that specific omic layer. The triangular heatmap at the bottom of the figure shows the correlation between individual omics features. Vertical magenta lines in panels A-C separate features from different omics layers. * indicates joint components which mediated greater than 15% of the total effect. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6.

Quasi-mediation analysis with intermediate integration identifies eight distinct groups of children at high risk of MAFLD based on prenatal mercury exposure and distinct omics profiles (Fig. 1, Column 3). Panel A shows the associations of prenatal mercury exposure (on left) with different omics profiles (in the middle). The red line connecting mercury exposure with each of the omics profiles indicates positive associations between prenatal mercury exposure, with the width of the lines being proportional to the magnitude of the association. The red line connecting each of the omics profiles with the outcome indicates that these omics profiles are associated with higher risk of liver injury in childhood. The dark green, dark purple, and dark gold lines indicate positive associations between the omic profile and the omic feature. In contrast, the light green, light purple, and light gold lines indicate negative associations. Panel B illustrates the eight groups of individuals with unique exposure, omics, and outcome profiles. Points indicate individuals and lines connect individuals with similar exposure, omic, and outcome profiles. Groups are positioned in order of increasing exposure and childhood CK-18. Panel C shows the exposure levels, childhood CK-18, and omics profiles for each of the eight groups in panel B. For example, group 1 has low exposure and low childhood CK-18 and is characterized by methylation profile 0, transcriptome profile 0, and miRNA profile 1. In contrast, group 8 has high exposure, moderate to high risk of MAFLD, and methylation profile 1, transcriptome profile 1, and miRNA profile 1. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)