Skip to main content
Fig. 3 | Genome Biology

Fig. 3

From: ChromNet: Learning the human chromatin network from all ENCODE ChIP-seq data

Fig. 3

Enrichment of BioGRID-supported edges between transcription factors in networks estimated by correlation (yellow), inverse correlation (red), and GroupGM (blue). In (a) and (b), light lines represent bootstrap resampling variability, and dark lines represent average performance over all resampled networks. a Fold enrichment for BioGRID-supported edges against a varying number of evaluated network edges. Top: Excluding edges between different cell types. Bottom: Only including edges between different cell types. b Enrichment against a varying number of evaluated network edges. Here, we compared between a joint model and a cell-type specific model in four different cell types. c Enrichment within cell types that have 25 supported edges or more, where the network density was set to match the number of BioGRID-supported edges in each cell type. Beneath each cell-type name is the number of data sets in that cell type

Back to article page