Socioeconomic inequalities in molecular risk for chronic diseases observed in young adulthood.
Michael J ShanahanSteven W ColeSudharshan RaviJustin ChumbleyWenjia XuCecilia PotenteBrandt LevittJulien BodeletAllison E AielloLauren GaydoshKathleen Mullan HarrisPublished in: Proceedings of the National Academy of Sciences of the United States of America (2022)
Many common chronic diseases of aging are negatively associated with socioeconomic status (SES). This study examines whether inequalities can already be observed in the molecular underpinnings of such diseases in the 30s, before many of them become prevalent. Data come from the National Longitudinal Study of Adolescent to Adult Health (Add Health), a large, nationally representative sample of US subjects who were followed for over two decades beginning in adolescence. We now have transcriptomic data (mRNA-seq) from a random subset of 4,543 of these young adults. SES in the household-of-origin and in young adulthood were examined as covariates of a priori -defined mRNA-based disease signatures and of specific gene transcripts identified de novo . An SES composite from young adulthood predicted many disease signatures, as did income and subjective status. Analyses highlighted SES-based inequalities in immune, inflammatory, ribosomal, and metabolic pathways, several of which play central roles in senescence. Many genes are also involved in transcription, translation, and diverse signaling mechanisms. Average causal-mediated effect models suggest that body mass index plays a key role in accounting for these relationships. Overall, the results reveal inequalities in molecular risk factors for chronic diseases often decades before diagnoses and suggest future directions for social signal transduction models that trace how social circumstances regulate the human genome.
Keyphrases
- genome wide
- mental health
- young adults
- healthcare
- depressive symptoms
- body mass index
- dna methylation
- public health
- endothelial cells
- single cell
- middle aged
- copy number
- physical activity
- electronic health record
- early life
- rna seq
- big data
- health information
- single molecule
- oxidative stress
- dna damage
- binding protein
- gene expression
- machine learning
- social media
- risk assessment
- deep learning
- artificial intelligence