Table 4 Search for epistatic effects based on SNPs exhibiting main-effects.

Per-CpG model in 3.5 MB window	N unique CpGs	Discovery sample Average variance explained by SNPs			Replication sample Average variance explained by SNPs
Per-CpG model in 3.5 MB window	N unique CpGs	Most-signif. main effect	All signif. main effects	All signif. main effects and interaction effects	Most-signif. main effect	All signif. main effects	All signif. main effects and interaction effects
- CpGs showing at least one significant main effect	59,134	16%	17.7%	17.7%	16%	18.1%	18.1%
- CpGs showing at least two significant main effect	17,938	22.6%	28.2%	28.3%	22.3%	29%	29.1%
- CpGs showing significant interaction effects	281	31.2%	41.9%	46.8%	31.2%	43.3%	49.1%

Average variance explained by main effects of SNPs or interaction effects of SNP-pairs. Results are shown for three different filtering steps, which were based on the number of significant main effects or interaction effects per CpG, identified with a forward-linear regression approach. Most-signif. main effect: average variance explained by the main effect that exhibited the smallest p-value. All signif. main effects: average variance explained by all main effects that were kept in the final model. All signif. main effects and interaction effects: average variance explained by all main effects and interaction effects that were kept in the final model. Signif.: significant.

Quick links