Fig. 1: Project overview.

The analysis was conducted using data from three British cohorts: Generation Scotland (GS), the Lothian Birth Cohort of 1936 (LBC1936) and the Avon Longitudinal Study of Parents and Children (ALSPAC). While blood-based DNA methylation (DNAm) data were available in all three cohorts, LBC1936 also contained information about the DNAm levels in post-mortem brain tissue, covering five brain regions from 14 individuals. A Bayesian Epigenome-Wide Association Study (EWAS) of smoking was performed in GS ( ~ 850k sites, Illumina EPIC array). For 23 pairs of age- and sex-matched smokers and non-smokers from GS, a high-resolution methylation measurement approach was implemented ( ~ 4 million sites, TWIST human methylome panel and ~21 million sites, Oxford Nanopore Technologies sequencing), followed by an EWAS analysis. An epigenetic biomarker of smoking, mCigarette, was developed in GS and tested as a predictor of self-reported smoking in LBC1936. The association between mCigarette and self-reported smoking was replicated in multiple age groups present in ALSPAC. Next, EWASs of smoking were run across five brain regions for 14 individuals using EPIC DNAm from LBC1936. Finally, Genome-Wide Association Studies (GWAS) were run in GS to compare genetic signal of self-reported and epigenetic smoking (GrimAge DNAm pack years score). EPIC – Illumina EPIC array, TWIST – TWIST Biosciences Human Methylome Panel, ONT – Oxford Nanopore Technologies Sequencing. Created in BioRender. Marioni, R. (2024) https://BioRender.com/h44z126.