Skip to main content
Fig. 7 | Genome Biology

Fig. 7

From: An interpretable bimodal neural network characterizes the sequence and preexisting chromatin predictors of induced transcription factor binding

Fig. 7

Induced TFs display a wide range of relative dependencies on preexisting chromatin features. a Distribution of model performance (auPRC) for a sequence-only network and Bichrom (sequence and preexisting NIH-3T3 ATAC-seq) for 9 TFs induced in mouse NIH-3T3 fibroblasts. The boxplots represent data from 5 independent training sets; each consisting of a separate held-out test chromosome. b Distribution of model performance (auPRC) for a sequence-only network and Bichrom (sequence and preexisting fibroblast ATAC-seq, H3K27ac, H3K4me2, and H3K27me3) for FOXA2, GATA4, and OCT4 induced in human BJ fibroblasts

Back to article page

Genome Biology

ISSN: 1474-760X

Contact us