Additional file 2: Supplementary Table 1. AD case control EWAS, Oxford dataset, P < 1.00E-05; test statistics after adjustment of DNAm data with estimated cell type compositions. Supplementary Table 2. AD Braak stage EWAS, Oxford dataset, P < 1.00E-05; test statistics after adjustment of DNAm data with estimated cell type compositions. Supplementary Table 3. Braak stage variance explained by PES with different P-value thresholds. Supplementary Table 4. Genomic Inflation Factors λ for all EWAS. Supplementary Table 5. Look-up of CpGs in previous AD DNAm studies. Supplementary Table 6. Experiment-wide significant AD associated CpGs in Smith et al. EC; Smithmeta: Results from the Smith et al. publication; Braak stage results for London1, London2, Oxford, and the meta-analysis (meta) across three datasets. Supplementary Table 7. Look-up of AD case control and Braak stage EWAS results from the Piras et al. publication in our study. Supplementary Table 8. Spearman rank correlations of DNAm and gene expression for CpG-gene pairs. Supplementary Table 9. Association of gene expression with AD case control status. Supplementary Table 10. Association of gene expression with AD Braak stage. Supplementary Table 11. Age acceleration linear regression results including DNAm PCs and genetic PCs as covariates. Supplementary Table 12. Meta-EWAS Results, comparison of Smith et al. with meta-EWAS using our methodology on London1 and London2 datasets. Supplementary Table 13. Meta-analysis across Oxford, London1, London2, including random effect meta-analysis results, and I2. Supplementary Table 14. Meta-analysis across Oxford, London1, London2 - Comparison of results using a threshold of 65% and 80% for the bisulfite-conversion efficiency in the Oxford dataset.