Table 3 Prediction accuracy from the average of cross-validation AUC using SNPs and methylation data simultaneously.

From: Integration of single nucleotide variants and whole-genome DNA methylation profiles for classification of rheumatoid arthritis cases from controls

Integrated data

Prediction accuracy

1% of methylation selected based on BRR method

+All SNPs as random effects except rs660895 as fixed

0.927 (0.032)b

+All SNPs as random effects

0.924 (0.034)b

+Subset of 41,673 LD-pruned SNPs + rs660895 as fixed

0.975 (0.014)a

+rs660895 as fixed

0.906 (0.037)c

  1. Average AUCs with different superscript(s) are significantly different (p value < 0.05).
Back to article page