The Hidden Biology of Wellbeing: A Multi-Omics Analysis Across Genomic, Epigenomic, and Transcriptomic Layers

Biological systems are composed of multiple molecular layers, including the genome, epigenome, transcriptome, and metabolome. These layers are interconnected and interact, but are often studied independently. Multi-omics approaches aim to integrate these layers to better understand shared biological processes and their roles in complex traits, such as wellbeing.

This project used data from 2,320 participants from the Netherlands Twin Register (NTR), which includes genome, epigenome, and transcriptome data from the same individuals, providing a rare opportunity to study interactions between omics layers. Multi-Omics Factor Analysis (MOFA) was applied, an unsupervised dimensionality reduction method, to identify latent factors that capture shared variation across omics layers. These factors may reflect underlying biological mechanisms not directly observed in individual omics layers. The phenotypic focus of this study is wellbeing, a multifactorial trait involving emotional, psychological, and social dimensions.

Following MOFA modeling, one latent factor showed to be statistically significantly associated with wellbeing. After studying the latent factor, it mainly seemed to capture epigenetic variation. The association between the latent factor and wellbeing did not remain significant after correcting for relevant covariates. The association was mainly driven by age, a well-known confounder of epigenetic signal.

By adopting a multi-omics framework, this study managed to move beyond traditional single-omics, phenotype-driven approaches and provided a more integrated approach of how biological layers may jointly influence (or not) complex traits like wellbeing.