Deep Convolution Recurrent Neural Network for Predicting RNA-Protein Binding Preference in mRNA UTR Region

mRNA translated region contains the genetic information needed to encode the protein, and the function of mRNA untranslated region is to regulate the process of protein translation. The binding of translation factors to specific sequences in the untranslated region can regulate the translation process. Predicting the binding between translation factor and mRNA sequences can help understand the mechanism of mRNA translation regulation. In this paper, We use CNN to extract local features, and LSTM to extract correlation features between different potential binding sites. By adjusting model structure, we tested the impact of CNN and LSTM on model performance. The results show that our model is better than the existing model.