Fig. 5
From: A data driven approach reveals disease similarity on a molecular level
Statistically and biologically explaining a high-dimensional curated Symmetric Kullback–Leibler Divergence. a c-SKL versus the number of top k probe sets that best explain the similarity of two datasets (GSE37171 and GSE46474 measured by GPL570) is shown in red. Gray color corresponds to the c-SKL computed using the same number of randomly selected probe sets. b Jaccard similarity coefficient between two cliques of AML measured by different platforms, GPL570 (microarray) and GPL11154 (RNA-seq) is shown in red. Gray color corresponds to the Jaccard index of the same similarity computed based on a random selection of genes. c, d Enrichment maps of the 20 most statistically significantly enriched pathways by the genes that explain a breast cancer clique of ten datasets measured by GPL570 and the similarity of eight different pairs of datasets connecting Alzheimer’s disease with schizophrenia
