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

Abstract

Motivation: Circular RNAs (circRNAs) have been found to have the potential to code proteins. Internal ribosome entry sites (IRESs) are key RNA regulatory elements for the translation of proteins by circRNAs through a cap-independent mechanism. IRES can be identified by bicistronic assay, but the method is time-consuming and laborious. Therefore, it is important to develop computational methods for facilitating IRES identification, evaluation, and design in circRNAs. Results: In this study, we proposed DeepCIP, a multimodal deep learning approach for circRNA IRES prediction, by exploiting both sequence and structure information. As far as we know, DeepCIP is the first predictor for circRNA IRESs, which consists of an RNA processing module, an S-LSTM module, a GCN module, a feature fusion module, and an ensemble module. The comparative studies show that DeepCIP outperforms other comparative methods and justify the effectiveness of the sequence model and structure model of DeepCIP for extracting features. We found that the integration of structural information on the basis of sequence information effectively improves predictive performance. For the real circRNA IRES prediction, DeepCIP also outperforms other methods. DeepCIP may facilitate the study of the coding potential of circRNAs as well as the design of circRNA drugs. DeepCIP as a standalone program is freely available at https://github.org/zjupgx/DeepCIP.

A 5’ UTR Language Model for Decoding Untranslated Regions of mRNA and Function Predictions

“…Several baseline methods exist for the task of predicting IRES, such as IRESfinder 13 , IRESpred 12 , IRESpy 11 , and DeepCIP 14 . IRESfinder is a logit model with framed k -mer features for cellular IRES prediction but may not be effective for viral IRES.…”

“…The model generalized well to unseen data, especially human 5' UTRs with varying lengths. Additionally, we adapted the UTR-LM to identify unannotated internal ribosome entry sites (IRESs) [11][12][13][14] , which are specific sequences within some mRNAs that enable ribosomes to initiate translation internally, bypassing the traditional cap-dependent mechanism. The UTR-LM outperforms IRESpy 11 by 0.15 in terms of the AUPR score.…”

Research progress on circular RNA vaccines

“…The selection of effective IRES structural elements is of great importance as it is a major determinant of protein translation by circRNA vaccines ( 24 , 51 , 58 ). Recently, it was reported that DeepCIP, the world’s first CircRNA IRES prediction tool, could predict the IRES that are more suitable for CircRNAs, so that CircRNAs can be adapted to different scenarios, such as vaccines and antitumor therapy ( 118 ). CircRNAs require delivery systems for encapsulation to form vaccines.…”

Advances in the protein‑encoding functions of circular RNAs associated with cancer (Review)

“…The majority of the aforementioned IRES prediction methods are based on traditional machine-learning algorithms, which are still limited as far as linear RNA IRES predictions are concerned, and at present, there are no prediction methods specifically suited for circRNA IRESs. To address this, Zhou et al (45) of Zhejiang University used DeepCIP, which, through a multi-modal deep learning approach, has been developed as a tool dedicated to the prediction of circRNA IRESs: This tool has enabled the study of the encoding potential of circRNAs and to more effectively capture the characteristics of circRNA IRESs. The SRAMP software is able to identify mammalian m 6 A sites with single-nucleotide resolution; users input mammalian circRNA sequences in order to predict the m 6 A motif.…”