Skip to main content
Fig. 3 | BMC Genomics

Fig. 3

From: Deep neural networks for inferring binding sites of RNA-binding proteins by using distributed representations of RNA primary sequence and secondary structure

Fig. 3

Performance comparison of the models with or without distributed representation of sequences and secondary structural profiles. The performance was evaluated in terms of AUROC on RBP-24 and RBP-31 dataset. DeepRKE is our proposed model, DeepRKE- model is without RNA secondary structure, and DeepRKE- - is without RNA secondary structure and distributed representation of sequence, using one-hot encoding instead. a-b Performance comparison between DeepRKE and DeepRKE- on two datasets. c-d Performance comparison between DeepRKE- and DeepRKE- - on two datasets. e-f Performance comparison between models with only CNN laryer and CNN+BiLSTM layer on two datasets

Back to article page