A semi-supervised approach for the integration of multi-omics data based on transformer multi-head self-attention mechanism and graph convolutional networks

Table 1 Dataset Overview

Dataset	Categories	Number of features for mRNA	Number of features for methylation	Number of Features for microRNA	Number of labeled nodes	Number of unlabeled nodes
BRCA	Normal-like:115, Basal-like: 131, HER2-enriched:46, LuminalA:436, Luminal B: 147	1000	1000	503	612	263
ROSMAP	NC:169, AD:182	200	200	200	245	106
GBM	Classical:71, Mesenchymal:47, Proneura:84, Neural:72	3613	1500	534	191	83

ISSN: 1471-2164