Fig. 2
From: A data driven approach reveals disease similarity on a molecular level
Towards a landscape of the biological dataome. a Problem definition: identify statistical similarities on a molecular level among public -omics datasets. b Compute all pairwise similarities based on the curated Symmetric Kullback–Leibler (c-SKL) divergence and the similarity of the covariance matrices. c The network of similarities among datasets of the same platform is visualized and explored for novel biological findings. The dataset similarity networks lead to a disease similarity network. d To gain intuition on the molecular underpinnings of interesting similarities, the molecular quantities that influence the c-SKL metric the most, are reported; these correspond to the same rows and columns in the covariance matrices not grayed out in the matrices on the right. They are used further as input for (pathway, gene ontology) enrichment analysis
