Multi-dimensional feature recognition model based on capsule network for ubiquitination site prediction

PeerJ

article

Weimin Li¹  Jie Wang¹  Yin Luo²  Tsigabu Teame Bezabih¹ 

[1] School of Computer Engineering and Science, Shanghai University;School of Life Sciences, East China Normal University

关键词: Ubiquitination site;    Capsule network;    Feature recognition;    Channel attention;   

DOI  :  10.7717/peerj.14427

学科分类：社会科学、人文和艺术（综合）

来源: Inra

【 摘 要 】

Ubiquitination is an important post-translational modification of proteins that regulates many cellular activities. Traditional experimental methods for identification are costly and time-consuming, so many researchers have proposed computational methods for ubiquitination site prediction in recent years. However, traditional machine learning methods focus on feature engineering and are not suitable for large-scale proteomic data. In addition, deep learning methods are mostly based on convolutional neural networks and fuse multiple coding approaches to achieve classification prediction. This cannot effectively identify potential fine-grained features of the input data and has limitations in the representation of dependencies between low-level features and high-level features. A multi-dimensional feature recognition model based on a capsule network (MDCapsUbi) was proposed to predict protein ubiquitination sites. The proposed module consisting of convolution operations and channel attention was used to recognize coarse-grained features in the sequence dimension and the feature map dimension. The capsule network module consisting of capsule vectors was used to identify fine-grained features and classify ubiquitinated sites. With ten-fold cross-validation, the MDCapsUbi achieved 91.82% accuracy, 91.39% sensitivity, 92.24% specificity, 0.837 MCC, 0.918 F-Score and 0.97 AUC. Experimental results indicated that the proposed method outperformed other ubiquitination site prediction technologies.

【 授权许可】

CC BY