Fig. 7
From: Towards in silico CLIP-seq: predicting protein-RNA interaction via sequence-to-signal learning
Predicting SARS-CoV-2 binding with RBPNet. The top shows target track predictions using a model trained on CNBP eCLIP data, together with the normalized observed signal taken from Schmidt et al. [55]. Shown below are predictions from a RBPNet-QKI model trained on ENCODE eCLIP data, as well as IG attribution maps around the top-5 positions with highest predicted probability