Fig 1.
Summary of the experimental design of the study.
The panels (above) represent the genetic and environmental components contributing to trait variation and used in the models (table below). Each cell shows the included random effects in each combination of model (row) and fixed effects (columns). G: Genomic, K: Kinship, GxSmk: Genome-by-Smoking, M: Methylation, GxM: Genome-by-Methylation, GxSmkxSex: Genome-by-Smoking-by-Sex, GxMxSex: Genome-by-Methylation-by-Sex. Models applied to different data sets varied depending on data availability.
Fig 2.
Proportion of trait variation explained by genetic and interaction sources in GS18K.
Proportion of BMI, fat percentage, and HDL variance (y-axis) explained by each of the genetic and interaction sources in the corresponding models (x-axis). G: Genomic, K: Kinship, GxSmk: Genome-by-Smoking, GxSmkxSex: Genome-by-Smoking-by-Sex.
Table 1.
Summary of interaction results for all cohorts.
Results of GKGxSmk model for all traits in GS18K and meta-analysis of the recruitment centre-based sub-cohorts in UK Biobank. The table shows, for each trait, the proportion of the phenotypic variance explained (Var), its standard error (SE), the log-likelihood ratio test P value (LRT P, only for the interaction), the meta-analysis P value (P), for each of the components in the model: Genetic (G), Kinship (K) and Genome-by-Smoking interaction (GxSmk). Highlighted P values indicate nominally significant results for the GxSmk component.
Fig 3.
Proportion of BMI variation explained by Genome-by-Smoking interactions across all cohorts and sub-cohorts.
The plot shows the proportion of BMI variance (the bars represent standard errors) explained by the genome-by-smoking interaction (x-axis) in the mixed model analyses across cohorts (y-axis). Panels from top to bottom represent cohorts: Generation Scotland (GS), UK Biobank (UKB), UK Biobank females (UKB_F) and UK Biobank males (UKB_M). Blue coloured data points show sub-cohort results, green coloured data points show meta-analyses of the corresponding panel sub-cohorts.
Fig 4.
Proportion of BMI variation explained by genetic, environmental and methylation sources in GS9K.
Proportion of BMI variance (y-axis) explained by each of the genetic, environmental and interaction sources in the corresponding models (x-axis). G: Genomic, K: Kinship, GxSmk: Genome-by-Smoking, M: Smoking associated methylation, GxM: Genome-by-Methylation, GxSmkxSex: Genome-by-Smoking-by-Sex, GxMxSex: Genome-by-Methylation-by-Sex.