ProALIGN: Directly Learning Alignments for Protein Structure Prediction via Exploiting Context-Specific Alignment Motifs

doi:10.1089/cmb.2021.0430

CORC > 理论物理研究所 > 中国科学院理论物理研究所 > 理论物理所科研产出 > 理论物理所1978-2010年知识产出

	ProALIGN: Directly Learning Alignments for Protein Structure Prediction via Exploiting Context-Specific Alignment Motifs
	Kong, Lupeng 2,3; Ju, Fusong 2; Zheng, Wei-mou ; Zhu, Jianwei 1; Sun, Shiwei 2; Xu, Jinbo 3; Bu, Dongbo 2
刊名	JOURNAL OF COMPUTATIONAL BIOLOGY
	2022
关键词	SERVER
ISSN号	1066-5277
DOI	10.1089/cmb.2021.0430
英文摘要	Template-based modeling (TBM), including homology modeling and protein threading, is one of the most reliable techniques for protein structure prediction. It predicts protein structure by building an alignment between the query sequence under prediction and the templates with solved structures. However, it is still very challenging to build the optimal sequence-template alignment, especially when only distantly related templates are available. Here we report a novel deep learning approach ProALIGN that can predict much more accurate sequence-template alignment. Like protein sequences consisting of sequence motifs, protein alignments are also composed of frequently occurring alignment motifs with characteristic patterns. Alignment motifs are context-specific as their characteristic patterns are tightly related to sequence contexts of the aligned regions. Inspired by this observation, we represent a protein alignment as a binary matrix (in which 1 denotes an aligned residue pair) and then use a deep convolutional neural network to predict the optimal alignment from the query protein and its template. The trained neural network implicitly but effectively encodes an alignment scoring function, which reduces inaccuracies in the handcrafted scoring functions widely used by the current threading approaches. For a query protein and a template, we apply the neural network to directly infer likelihoods of all possible residue pairs in their entirety, which could effectively consider the correlations among multiple residues. We further construct the alignment with maximum likelihood, and finally build a structure model according to the alignment. Tested on three independent data sets with a total of 6688 protein alignment targets and 80 CASP13 TBM targets, our method achieved much better alignments and 3D structure models than the existing methods, including HHpred, CNFpred, CEthreader, and DeepThreader. These results clearly demonstrate the effectiveness of exploiting the context-specific alignment motifs by deep learning for protein threading.
学科主题	Biochemistry & Molecular Biology ; Biotechnology & Applied Microbiology ; Computer Science ; Mathematical & Computational Biology ; Mathematics
语种	英语
内容类型	期刊论文
源URL	[http://ir.itp.ac.cn/handle/311006/27911]
专题	理论物理研究所_理论物理所1978-2010年知识产出
作者单位	1.Chinese Acad Sci, Inst Theoret Phys, Beijing, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing, Peoples R China 4.Microsoft Res Asia, Beijing, Peoples R China 5.Toyota Technol Inst, Chicago, IL 60637 USA
推荐引用方式 GB/T 7714	Kong, Lupeng,Ju, Fusong,Zheng, Wei-mou,et al. ProALIGN: Directly Learning Alignments for Protein Structure Prediction via Exploiting Context-Specific Alignment Motifs[J]. JOURNAL OF COMPUTATIONAL BIOLOGY,2022.
APA	Kong, Lupeng.,Ju, Fusong.,Zheng, Wei-mou.,Zhu, Jianwei.,Sun, Shiwei.,...&Bu, Dongbo.(2022).ProALIGN: Directly Learning Alignments for Protein Structure Prediction via Exploiting Context-Specific Alignment Motifs.JOURNAL OF COMPUTATIONAL BIOLOGY.
MLA	Kong, Lupeng,et al."ProALIGN: Directly Learning Alignments for Protein Structure Prediction via Exploiting Context-Specific Alignment Motifs".JOURNAL OF COMPUTATIONAL BIOLOGY (2022).