Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: Multi-resBind: a residual network-based multi-label classifier for in vivo RNA binding prediction and preference visualization

Fig. 3

Attribution maps of several respective RBPs (MBNL1, PUM2, and QKI), utilizing the sequence contribution only. For each RBP, the three sequences with the highest prediction scores were used to calculate gradients and generate the corresponding contribution relative to the predicted probability. The y-axis represents the relative contribution of each base and the x-axis represents the corresponding nucleic acid (A, U, G, or C) in the RNA sequence. Attribution maps for model interpretation were constructed using the IG method, in which the reference point was generated by taking the average of 2000 samples with the lowest prediction score. The sequence logos corresponding to attribution maps were generated through the RCAS profiles of the RBP binding sites [33]. The sequence logo of PUM2 was obtained from a pulished study [10]

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us