Fig. 6

Workflow of the evaluation on cancer multi-omics datasets. a Three kinds of omics data were used as input. b Supervised DL methods are evaluated in the classification tasks. The performance of these methods was based on 4-fold cross-validation and was evaluated by three metrics: accuracy, F1 macro, and F1 weighted score. c Unsupervised DL methods were first applied to fuse the cancer multi-omics data to obtain the fused 10-dimensional embeddings. Then k-means algorithm was used to cluster the multi-omics dimensionality reduction results into several categories. We employed C-index, silhouette score and Davies Bouldin score as the evaluation indexes of clustering. Furthermore, the associations of the embeddings with survival and clinical annotations were evaluated