DeepCIP: A multimodal deep learning method for the prediction of internal ribosome entry sites of circRNAs

• Published in: Computers in biology and medicine 2023-09, Vol.164, p.107288-107288, Article 107288
• Main Authors: Zhou, Yuxuan, Wu, Jingcheng, Yao, Shihao, Xu, Yulian, Zhao, Wenbin, Tong, Yunguang, Zhou, Zhan
• Format: Article
• Language: English
• Subjects: Cap-dependent translation; Chemical bonds; Circular RNA; Circular RNAs; Datasets; Deep learning; Drug development; Feature extraction; Hydrogen bonds; Internal ribosome entry sites; Machine learning; Multimodal model; Predictions; Proteins; Regulatory sequences; Translation
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2023.107288

Circular RNAs (circRNAs) have been found to have the ability to encode proteins through internal ribosome entry sites (IRESs), which are essential RNA regulatory elements for cap-independent translation. Identification of IRES elements in circRNA is crucial for understanding its function. Previous studies have presented IRES predictors based on machine learning techniques, but they were mainly designed for linear RNA IRES. In this study, we proposed DeepCIP (Deep learning method for CircRNA IRES Prediction), a multimodal deep learning approach that employs both sequence and structural information for circRNA IRES prediction. Our results demonstrate the effectiveness of the sequence and structure models used by DeepCIP in feature extraction and suggest that integrating sequence and structural information efficiently improves the accuracy of prediction. The comparison studies indicate that DeepCIP outperforms other comparative methods on the test set and real circRNA IRES dataset. Furthermore, through the integration of an interpretable analysis mechanism, we elucidate the sequence patterns learned by our model, which align with the previous discovery of motifs that facilitate circRNA translation. Thus, DeepCIP has the potential to enhance the study of the coding potential of circRNAs and contribute to the design of circRNA-based drugs. DeepCIP as a standalone program is freely available at https://github.org/zjupgx/DeepCIP. [Display omitted] •Given the variations in IRES activity between linear RNA and circular RNA, it is imperative to devise methods for circRNA IRES prediction.•A multimodal deep learning framework together with RNA sequence and secondary structure features used to predict circular RNA internal ribosome entry sites is proposed.•Experiments demonstrated that DeepCIP outperformed other comparative methods on the test set and experimentally validated circRNA IRES data, and can predict circRNA IRES more effectively.•DeepCIP is the first deep learning method specifically designed for circRNA IRES prediction, which is expected to contribute to the research field of circRNA translation.