Skip to main content

Table 6 Performance comparison with Bayesian multi-task multi-kernel learning (BEMKL) method

From: Deep learning and multi-omics approach to predict drug responses in cancer

Dataset

Our method

BEKML

MSE

\(\mathrm {R^2}\)

MSE

\(\mathrm {R^2}\)

Gene expression

2.37

0.24

2.37

0.31

Gene mutations

3.25

−0.02

3.31

0.02

CNV

3.32

−0.04

4.51

−0.22

RPPA

3.11

−0.02

2.52

0.33

Metabolites

2.97

0.07

3.25

0.10

Multi-omics

0.46

0.84

2.32

0.34