TY - JOUR
T1 - RSMBSP-DON: RNA-Small Molecule Binding Sites Prediction by Dual-path feature extraction and One-dimensional multi-scale feature fusion Network
AU - Yang, Xiao
AU - Sun, Zhan Li
AU - Liu, Mengya
AU - Zeng, Zhigang
AU - Lam, Kin Man
AU - Wang, Xin
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2025/4
Y1 - 2025/4
N2 - Due to the significant differences between the structural and sequence information of RNA, accurately predicting RNA-small molecule binding sites by utilizing these two attributes remains a challenging task. This study introduces a novel network for predicting RNA-small molecule binding sites, employing a two-stage approach that integrates feature extraction and fusion processes. On one hand, in order to capture the diverse characteristic information of RNA, a dual-path feature extraction module is proposed to extract features from both short-range and long-range perspectives, by incorporating convolutional and attention networks. On the other hand, a one-dimensional multi-scale feature fusion module, consisting of parallel one-dimensional convolutional kernels, is proposed to extract feature information at multiple granularities and to effectively integrate the features of nucleotides on the RNA chain and their neighboring nucleotides. Experimental results demonstrate that RSMBSP-DON is competitive with some recently reported methods.
AB - Due to the significant differences between the structural and sequence information of RNA, accurately predicting RNA-small molecule binding sites by utilizing these two attributes remains a challenging task. This study introduces a novel network for predicting RNA-small molecule binding sites, employing a two-stage approach that integrates feature extraction and fusion processes. On one hand, in order to capture the diverse characteristic information of RNA, a dual-path feature extraction module is proposed to extract features from both short-range and long-range perspectives, by incorporating convolutional and attention networks. On the other hand, a one-dimensional multi-scale feature fusion module, consisting of parallel one-dimensional convolutional kernels, is proposed to extract feature information at multiple granularities and to effectively integrate the features of nucleotides on the RNA chain and their neighboring nucleotides. Experimental results demonstrate that RSMBSP-DON is competitive with some recently reported methods.
KW - Binding sites prediction
KW - deep learning
KW - feature extraction
KW - feature fusion
UR - http://www.scopus.com/inward/record.url?scp=105003634778&partnerID=8YFLogxK
U2 - 10.1109/TAI.2025.3564243
DO - 10.1109/TAI.2025.3564243
M3 - Journal article
AN - SCOPUS:105003634778
SN - 2691-4581
JO - IEEE Transactions on Artificial Intelligence
JF - IEEE Transactions on Artificial Intelligence
ER -