Skip to main content

Table 1 RMSE for missing data imputation for differing numbers of randomly removed arrays

From: Hierarchical Bayesian modelling of gene expression time series across irregularly sampled replicates and clusters

  1 of 56 5 of 56 10 of 56 20 of 56
HGP 0 . 3 0 ± 0 . 2 7 0 . 3 2 ± 0 . 0 9 0 . 3 4 ± 0 . 0 5 0 . 3 8 ± 0 . 0 6
GP 0.46±0.23 0.44±0.09 0.43±0.08 0.46±0.07
Mean 0.52±0.24 0.48±0.12 0.48±0.11 0.48±0.08
Median 0.50±0.25 0.46±0.11 0.47±0.11 0.48±0.08
  1. The rows correspond to Hierarchical GPs (HGP), non-hierarchical GPs (GP), and imputation using the mean and median of available replicates. Columns represent increasing amounts of missing data.