DIFFUSE: predicting isoform functions from sequences and expression profiles via deep learning

doi:10.1093/bioinformatics/btz367

	DIFFUSE: predicting isoform functions from sequences and expression profiles via deep learning
	Chen, Hao 1; Shaw, Dipan 1; Zeng, Jianyang 2; Bu, Dongbo 3,4; Jiang, Tao 1,5
	2019-07-15
发表期刊	BIOINFORMATICS
ISSN	1367-4803
卷号	35 期号:14 页码:I284-I294
摘要	Motivation Alternative splicing generates multiple isoforms from a single gene, greatly increasing the functional diversity of a genome. Although gene functions have been well studied, little is known about the specific functions of isoforms, making accurate prediction of isoform functions highly desirable. However, the existing approaches to predicting isoform functions are far from satisfactory due to at least two reasons: (i) unlike genes, isoform-level functional annotations are scarce. (ii) The information of isoform functions is concealed in various types of data including isoform sequences, co-expression relationship among isoforms, etc. Results In this study, we present a novel approach, DIFFUSE (Deep learning-based prediction of IsoForm FUnctions from Sequences and Expression), to predict isoform functions. To integrate various types of data, our approach adopts a hybrid framework by first using a deep neural network (DNN) to predict the functions of isoforms from their genomic sequences and then refining the prediction using a conditional random field (CRF) based on co-expression relationship. To overcome the lack of isoform-level ground truth labels, we further propose an iterative semi-supervised learning algorithm to train both the DNN and CRF together. Our extensive computational experiments demonstrate that DIFFUSE could effectively predict the functions of isoforms and genes. It achieves an average area under the receiver operating characteristics curve of 0.840 and area under the precision-recall curve of 0.581 over 4184 GO functional categories, which are significantly higher than the state-of-the-art methods. We further validate the prediction results by analyzing the correlation between functional similarity, sequence similarity, expression similarity and structural similarity, as well as the consistency between the predicted functions and some well-studied functional features of isoform sequences. Availability and implementation https://github.com/haochenucr/DIFFUSE. Supplementary information Supplementary data are available at Bioinformatics online.
DOI	10.1093/bioinformatics/btz367
收录类别	SCI
语种	英语
资助项目	National Science Foundation[IIS-1646333] ; National Natural Science Foundation of China[61772197] ; National Natural Science Foundation of China[31671369] ; National Natural Science Foundation of China[31770775] ; National Natural Science Foundation of China[61872216] ; National Natural Science Foundation of China[61472205] ; National Natural Science Foundation of China[81630103] ; National Key Research and Development Program of China[2018YFC0910404] ; National Key Research and Development Program of China[2018YFC0910405]
WOS研究方向	Biochemistry & Molecular Biology ; Biotechnology & Applied Microbiology ; Computer Science ; Mathematical & Computational Biology ; Mathematics
WOS类目	Biochemical Research Methods ; Biotechnology & Applied Microbiology ; Computer Science, Interdisciplinary Applications ; Mathematical & Computational Biology ; Statistics & Probability
WOS记录号	WOS:000477703600033
出版者	OXFORD UNIV PRESS
引用统计	被引频次：26[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/4546
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Jiang, Tao
作者单位	1.Univ Calif Riverside, Dept Compute Sci & Engn, Riverside, CA 92521 USA 2.Tsinghua Univ, Inst Interdisciplinary Informat Sci, Beijing 100084, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 4.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 5.Tsinghua Univ, Dept Comp Sci & Technol, BNRIST, Bioinformat Div, Beijing 100084, Peoples R China
推荐引用方式 GB/T 7714	Chen, Hao,Shaw, Dipan,Zeng, Jianyang,et al. DIFFUSE: predicting isoform functions from sequences and expression profiles via deep learning[J]. BIOINFORMATICS,2019,35(14):I284-I294.
APA	Chen, Hao,Shaw, Dipan,Zeng, Jianyang,Bu, Dongbo,&Jiang, Tao.(2019).DIFFUSE: predicting isoform functions from sequences and expression profiles via deep learning.BIOINFORMATICS,35(14),I284-I294.
MLA	Chen, Hao,et al."DIFFUSE: predicting isoform functions from sequences and expression profiles via deep learning".BIOINFORMATICS 35.14(2019):I284-I294.