Simple tricks of convolutional neural network architectures improve DNA-protein binding prediction

doi:10.1093/bioinformatics/bty893

CSpace > 应用数学研究所

	Simple tricks of convolutional neural network architectures improve DNA-protein binding prediction
	Cao, Zhen 1,2; Zhang, Shihua1,2,3
	2019-06-01
发表期刊	BIOINFORMATICS
ISSN	1367-4803
卷号	35 期号:11 页码:1837-1843
摘要	Motivation: With the accumulation of DNA sequencing data, convolution neural network (CNN) based methods such as DeepBind and DeepSEA have achieved great success for predicting the function of primary DNA sequences. Previous studies confirm the importance of utilizing the reverse complement and flanking DNA sequences, which has a natural connection with data augmentation. However, it is not fully understood how these DNA sequences work during model training and testing. Results: In this study, we proposed several CNN tricks to improve the DNA sequence related prediction tasks and took the DNA-protein binding prediction as an illustrative task for demonstration. Different from the DeepBind, we treated the reverse complement DNA sequence as another sample, which enables the CNN model to automatically learn the complex relationships between the double strand DNA sequences. This trick promotes the using of deeper CNN models, improving the prediction performance. Next, we augmented the training sets by extending the DNA sequences and cropping each one to three shorter sequences. This approach greatly improves the prediction due to more environmental information from extending step and strong regularization effect of the cropping step. Moreover, this practice fits well with wider CNN models, which also increases the prediction accuracy. On the basis of DNA sequence augmentation, we integrated the results of different effective CNN models to mine the prediction potential of primary DNA sequences. On 156 datasets of predicting DNA-protein binding, our final prediction significantly outperformed the state-of-the-art results with an average AUC increase of 0.057 (P-value = 6 x 10(-62)).
DOI	10.1093/bioinformatics/bty893
语种	英语
资助项目	National Natural Science Foundation of China[11661141019] ; National Natural Science Foundation of China[61621003] ; National Natural Science Foundation of China[61422309] ; National Natural Science Foundation of China[61379092] ; Strategic Priority Research Program of the Chinese Academy of Sciences (CAS)[XDB13040600] ; National Ten Thousand Talent Program for Young Topnotch Talents ; Key Research Program of the Chinese Academy of Sciences[KFZD-SW-219] ; National Key Research and Development Program of China[2017YFC0908405] ; CAS Frontier Science Research Key Project for Top Young Scientist[QYZDB-SSW-SYS008]
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:000476567500005
出版者	OXFORD UNIV PRESS
引用统计
文献类型	期刊论文
条目标识符	http://ir.amss.ac.cn/handle/2S8OKBNM/35248
专题	应用数学研究所
通讯作者	Zhang, Shihua
作者单位	1.Chinese Acad Sci, Acad Math & Syst Sci, RCSDS, NCMIS,CEMS, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Math Sci, Beijing 100049, Peoples R China 3.Chinese Acad Sci, Ctr Excellence Anim Evolut & Genet, Kunming 650223, Yunnan, Peoples R China
推荐引用方式 GB/T 7714	Cao, Zhen,Zhang, Shihua. Simple tricks of convolutional neural network architectures improve DNA-protein binding prediction[J]. BIOINFORMATICS,2019,35(11):1837-1843.
APA	Cao, Zhen,&Zhang, Shihua.(2019).Simple tricks of convolutional neural network architectures improve DNA-protein binding prediction.BIOINFORMATICS,35(11),1837-1843.
MLA	Cao, Zhen,et al."Simple tricks of convolutional neural network architectures improve DNA-protein binding prediction".BIOINFORMATICS 35.11(2019):1837-1843.