Skip to main content

Table 2 Genome-wide performance of different algorithms on predicting methylation values, averaged across tissues and samples

From: BoostMe accurately predicts DNA methylation values in whole-genome bisulfite sequencing of multiple human tissues

Algorithm

RMSE (all)

RMSE (int.)

AUROC

AUPRC

Accuracy

Resources

Time (hrs)

BoostMe

0.09 ± 0.005

0.13 ± 0.006

0.99 ± 0.002

0.99 ± 0.0005

0.96 ± 0.005

16 CPUs

0.50 ± 0.15

Random Forests

0.09 ± 0.005

0.13 ± 0.006

0.99 ± 0.002

0.99 ± 0.0005

0.96 ± 0.005

16 CPUs

14 ± 2

DeepCpG

0.17 ± 0.007

0.27 ± 0.014

0.94 ± 0.003

0.98 ± 0.002

0.91 ± 0.010

1 GPU

140 ± 30

  1. RMSE, root-mean-squared error; int., intermediate beta values, defined as having a sample average methylation between 0.2 and 0.8; AUROC, area under the receiver operating characteristic curve; AUPRC, area under the precision-recall curve. Time is the average number of computational hours it took to train on all samples within a tissue