A Representation Learning Approach for Predicting circRNA Back-Splicing Event via Sequence-Interaction-Aware Dual Encoder

Chengxin He; Lei Duan; Huiru Zheng; Xinye Wang; Lili Guan; Jiaxuan Xu

doi:10.1109/TNB.2024.3454079

A Representation Learning Approach for Predicting circRNA Back-Splicing Event via Sequence-Interaction-Aware Dual Encoder

Chengxin He
, Lei Duan
, Huiru Zheng
, Xinye Wang
, Lili Guan
, Jiaxuan Xu

Research output: Contribution to journal › Article › peer-review

128 Downloads (Pure)

Abstract

Circular RNAs (circRNAs) play a crucial role in gene
regulation and association with diseases because of their unique
closed continuous loop structure, which is more stable and conserved than ordinary linear RNAs. As fundamental work to clarify
their functions, a large number of computational approaches for
identifying circRNA formation have been proposed. However, these
methods fail to fully utilize the important characteristics of backsplicing events, i.e., the positional information of the splice sites
and the interaction features of its flanking sequences, for predicting circRNAs. To this end, we hereby propose a novel approach
called SIDE for predicting circRNA back-splicing events using only
raw RNA sequences. Technically, SIDE employs a dual encoder
to capture global and interactive features of the RNA sequence,
and then a decoder designed by the contrastive learning to fuse
out discriminative features improving the prediction of circRNAs
formation. Empirical results on three real-world datasets show
the effectiveness of SIDE. Further analysis also reveals that the
effectiveness of SIDE.

Original language	English
Pages (from-to)	603-611
Number of pages	9
Journal	IEEE Transactions on Nanobioscience
Volume	23
Issue number	4
Early online date	3 Sept 2024
DOIs	https://doi.org/10.1109/TNB.2024.3454079
Publication status	Published (in print/issue) - 14 Oct 2024

Bibliographical note

Publisher Copyright:
© 2002-2011 IEEE.

Funding

This work was supported in part by the National Natural Science Foundation of China (61972268), and the Joint Innovation Foundation of Sichuan University and Nuclear Power Institute of China. (Corresponding author: Lei Duan) Chengxin He, Lei Duan, Xinye Wang, Lili Guan, and Jiax-uan Xu are with School of Computer Science, Sichuan University, Chengdu, China. (Email: [email protected], [email protected], [email protected], [email protected], [email protected]) Huiru Zheng is with School of Computing, Ulster University, Northern Ireland, United Kingdom. (Email: [email protected])

Funders	Funder number
Nuclear Power Institute of China
National Natural Science Foundation of China	61972268

Keywords

Back-splicing events prediction
CircRNA formation
Contrastive learning
Deep learning
Algorithms
Humans
RNA, Circular/genetics
RNA Splicing/genetics
Computational Biology/methods
Machine Learning
Sequence Analysis, RNA/methods

Access to Document

10.1109/TNB.2024.3454079

A_Representation_Learning_Approach_for_Predicting_circRNA_Back-Splicing_Event_via_Sequence-Interaction-Aware_Dual_EncoderFinal published version, 5.5 MB

Cite this

@article{72055f6cc6a043d4a7f5d5adc92405d7,

title = "A Representation Learning Approach for Predicting circRNA Back-Splicing Event via Sequence-Interaction-Aware Dual Encoder",

abstract = "Circular RNAs (circRNAs) play a crucial role in gene regulation and association with diseases because of their unique closed continuous loop structure, which is more stable and conserved than ordinary linear RNAs. As fundamental work to clarify their functions, a large number of computational approaches for identifying circRNA formation have been proposed. However, these methods fail to fully utilize the important characteristics of backsplicing events, i.e., the positional information of the splice sites and the interaction features of its flanking sequences, for predicting circRNAs. To this end, we hereby propose a novel approach called SIDE for predicting circRNA back-splicing events using only raw RNA sequences. Technically, SIDE employs a dual encoder to capture global and interactive features of the RNA sequence, and then a decoder designed by the contrastive learning to fuse out discriminative features improving the prediction of circRNAs formation. Empirical results on three real-world datasets show the effectiveness of SIDE. Further analysis also reveals that the effectiveness of SIDE.",

keywords = "Back-splicing events prediction, CircRNA formation, Contrastive learning, Deep learning, Algorithms, Humans, RNA, Circular/genetics, RNA Splicing/genetics, Computational Biology/methods, Machine Learning, Sequence Analysis, RNA/methods",

author = "Chengxin He and Lei Duan and Huiru Zheng and Xinye Wang and Lili Guan and Jiaxuan Xu",

note = "Publisher Copyright: {\textcopyright} 2002-2011 IEEE.",

year = "2024",

month = oct,

day = "14",

doi = "10.1109/TNB.2024.3454079",

language = "English",

volume = "23",

pages = "603--611",

journal = "IEEE Transactions on Nanobioscience",

issn = "1536-1241",

publisher = "IEEE",

number = "4",

}

TY - JOUR

T1 - A Representation Learning Approach for Predicting circRNA Back-Splicing Event via Sequence-Interaction-Aware Dual Encoder

AU - He, Chengxin

AU - Duan, Lei

AU - Zheng, Huiru

AU - Wang, Xinye

AU - Guan, Lili

AU - Xu, Jiaxuan

PY - 2024/10/14

Y1 - 2024/10/14

N2 - Circular RNAs (circRNAs) play a crucial role in gene regulation and association with diseases because of their unique closed continuous loop structure, which is more stable and conserved than ordinary linear RNAs. As fundamental work to clarify their functions, a large number of computational approaches for identifying circRNA formation have been proposed. However, these methods fail to fully utilize the important characteristics of backsplicing events, i.e., the positional information of the splice sites and the interaction features of its flanking sequences, for predicting circRNAs. To this end, we hereby propose a novel approach called SIDE for predicting circRNA back-splicing events using only raw RNA sequences. Technically, SIDE employs a dual encoder to capture global and interactive features of the RNA sequence, and then a decoder designed by the contrastive learning to fuse out discriminative features improving the prediction of circRNAs formation. Empirical results on three real-world datasets show the effectiveness of SIDE. Further analysis also reveals that the effectiveness of SIDE.

AB - Circular RNAs (circRNAs) play a crucial role in gene regulation and association with diseases because of their unique closed continuous loop structure, which is more stable and conserved than ordinary linear RNAs. As fundamental work to clarify their functions, a large number of computational approaches for identifying circRNA formation have been proposed. However, these methods fail to fully utilize the important characteristics of backsplicing events, i.e., the positional information of the splice sites and the interaction features of its flanking sequences, for predicting circRNAs. To this end, we hereby propose a novel approach called SIDE for predicting circRNA back-splicing events using only raw RNA sequences. Technically, SIDE employs a dual encoder to capture global and interactive features of the RNA sequence, and then a decoder designed by the contrastive learning to fuse out discriminative features improving the prediction of circRNAs formation. Empirical results on three real-world datasets show the effectiveness of SIDE. Further analysis also reveals that the effectiveness of SIDE.

KW - Back-splicing events prediction

KW - CircRNA formation

KW - Contrastive learning

KW - Deep learning

KW - Algorithms

KW - Humans

KW - RNA, Circular/genetics

KW - RNA Splicing/genetics

KW - Computational Biology/methods

KW - Machine Learning

KW - Sequence Analysis, RNA/methods

UR - https://www.scopus.com/pages/publications/85203545488

U2 - 10.1109/TNB.2024.3454079

DO - 10.1109/TNB.2024.3454079

M3 - Article

C2 - 39226209

SN - 1536-1241

VL - 23

SP - 603

EP - 611

JO - IEEE Transactions on Nanobioscience

JF - IEEE Transactions on Nanobioscience

IS - 4

ER -

A Representation Learning Approach for Predicting circRNA Back-Splicing Event via Sequence-Interaction-Aware Dual Encoder

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this