Skip to main content

Table 7 Global imputation performances across all datasets (healthy and disease) and all missingness mechanisms

From: Methylation data imputation performances under different representations and missingness patterns

 

MAE

RMSE

Method

M-value

B-value

M-value

B-value

mean

0.041 ±0.001

0.040 ±0.001

0.068 ±0.001

0.066 ±0.001

impute.knn

0.052 ±0.008

0.068 ±0.011

0.095 ±0.014

0.119 ±0.016

softImpute

0.042 ±0.003

0.043 ±0.008

0.072 ±0.006

0.077 ±0.020

imputePCA

0.035 ±0.001

0.035 ±0.001

0.062 ±0.002

0.059 ±0.001

SVDmiss

0.046 ±0.001

0.037 ±0.001

0.079 ±0.003

0.065 ±0.002

missForest

0.036 ±0.001

0.036 ±0.001

0.061 ±0.002

0.059 ±0.002

methyLImp

0.037 ±0.001

0.033 ±0.001

0.062 ±0.002

0.058 ±0.002

  1. For each method, the symbol indicates the measure (either M-value or β-value) for which the Wilcoxon signed-rank test p-value is <0.05. Best results per metric with respect to the Wilcoxon signed-rank test are highlighted in bold