Skip to main content

Table 3 Performance of Various Modules of MOSAE

From: A multi-omics supervised autoencoder for pan-cancer clinical outcome endpoints prediction

Methods OS DSS PFI DFI
Plain autoencoder 0.7632
(±0.0058)
0.7660
(±0.0229)
0.6999
(±0.0103)
0.6615
(±0.0340)
MO + Cat 0.7644
(±0.0135)
0.7709
(±0.0292)
0.7030
(±0.0115)
0.6634
(±0.0366)
MO + Ave 0.7682
(±0.0134)
0.7753
(±0.0291)
0.7189
(±0.0136)
0.6942
(±0.0368)
MO + Ave + Sup 0.7721
(±0.0112)
0.7793
(±0.0252)
0.7227
(±0.0124)
0.6960
(±0.0388)
MO + Ave + Sup + Spec 0.7830
(±0.0081)
0.7870
(±0.0293)
0.7325
(±0.0123)
0.7061
(±0.0393)