Skip to main content
. 2023 Jul 6;20(7):e1004247. doi: 10.1371/journal.pmed.1004247

Fig 4. Blood CpGs associated with prevalent or incident disease states showing effect sizes on interpretable scale.

Fig 4

Effect sizes were reestimated using logistic regression (prevalent disease, blue points) or Cox proportional hazards models (incident disease, violet points) to return more interpretable effect sizes. Effect sizes represent a per-1 SD increase in age-, sex-, and experimental batch-adjusted CpG methylation M-values (or age- and batch-adjusted for breast cancer). CpGs shown were significant in both basic and fully adjusted models. Odds ratios and hazard ratios are detailed in S22 Table. CI, confidence interval; CpG, cytosine-phosphate-guanine dinucleotide; SD, standard deviation.