Skip to main content

Table 2 RMSE of missing data imputation for hierarchical Gaussian processes (HGP) principal component analysis (PCA) and K-nearest neighbour (KNN)

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

  Replicate median Full model
HGP 0 . 1 3 ± 0 . 0 3 0 . 0 5 ± 0 . 0 1
PCA 0.15±0.03 0.10±0.02
KNN 0.17±0.04 0.13±0.02
  1. The first column shows the RMSE measured against the replicate-median, and the second column shows the RMSE against the model with no missing data.