Fig. 2
From: ExplaiNN: interpretable and transparent neural networks for genomics
Exploring the limitations of ExplaiNN in capturing nonlinear interactions. Cooperativity (residual fold change; y-axis) plotted as a function of distance (x-axis) between the motifs of the housekeeping TFs Dref (top row; yellow), Ohler1 (middle row; blue), and Ohler6 (bottom row; green) for DeepSTARR [30] (left column) and ExplaiNN (right column). The 5-mer GGGCT is provided as a negative control (light gray). TF, transcription factor