Advances in applications of deep learning for predicting sequence-based protein interactions

doi:10.12211/2096-8280.2023-074

Synthetic Biology Journal ›› 2024, Vol. 5 ›› Issue (1): 88-106.DOI: 10.12211/2096-8280.2023-074

• Invited Review • Previous Articles Next Articles

Advances in applications of deep learning for predicting sequence-based protein interactions

Jingyong ZHU¹^,²^,³, Junxiang LI³^,⁴, Xuhui LI³^,⁵, Jin ZHANG², Wenjing WU²

^1.College of Life Sciences and Medicine，Zhejiang Sci-tech University，Hangzhou 310018，Zhejiang，China
^2.College of Biological Chemical Sciences and Engineering，Jiaxing University，Jiaxing 314000，Zhejiang，China
^3.Agecode R&D Center，Yangtze Delta Region Institute of Tsinghua University，Jiaxing 341001，Zhejiang，China
^4.Harvest Biotech. Co. ，Ltd. ，Jiaxing 341001，Zhejiang，China
^5.Zhejiang Provincial Key Laboratory of Applied Enzymology，Yangtze Delta Region Institute of Tsinghua University，Jiaxing 314006，Zhejiang，China

Received:2023-10-24 Revised:2023-11-28 Online:2024-03-20 Published:2024-02-29
Contact: Jin ZHANG

深度学习在基于序列的蛋白质互作预测中的应用进展

朱景勇¹^,²^,³, 李钧翔³^,⁴, 李旭辉³^,⁵, 张瑾², 毋文静²

^1.浙江理工大学生命科学与医药学院，浙江杭州 310018
^2.嘉兴学院生物与化学工程学院，浙江嘉兴 314000
^3.浙江清华长三角研究院，衰老科学创新研发中心，浙江嘉兴 341001
^4.禾美生物科技（浙江）有限公司，浙江嘉兴 341001
^5.浙江清华长三角研究院，浙江省应用酶学重点实验室，浙江嘉兴 314006

通讯作者: 张瑾
作者简介:朱景勇（1998—），男，硕士研究生。研究方向为深度学习预测蛋白质互作。 E-mail：jingyongzhu2016@163.com
张瑾（1976—），男，教授，硕士生导师。研究方向为代谢疾病的致病机理及药物靶点发掘。 E-mail：zhangjin7688@163.com
基金资助:
国家自然科学基金(32172708);浙江省自然科学基金重点项目(LZ23C170002)

Abstract

Abstract:

Protein-protein interactions play a crucial role in biological processes such as cell signal transduction, gene expression and metabolic regulation, and thus their identification is essential for understanding these complex biological processes. Predicting protein-protein interactions is a hot topic of great significance, which can provide assistances in areas such as drug discovery and protein function research and design as well. In recent years, with the development of artificial intelligence, machine learning technologies have been applied gradually to the prediction of protein-protein interactions, which has shown good potentials. However, when processing a large amount of protein information, traditional machine learning methods are difficult to mine the intrinsic patterns and potential features, and deep learning techniques are needed. Compared with the three-dimensional structure of proteins, sequence information is easier to obtain, and the development of high-throughput sequencing technology provides abundant protein sequence information, which greatly facilitates the development of sequence-based deep learning technologies. Sequence-based deep learning models predict protein-protein interactions by learning intrinsic patterns and features from protein sequence information, which greatly improves prediction efficiency and accuracy. In this review, we focus on progress of deep learning in predicting sequence-based protein interactions, categorize, which is summarized according to the algorithmic framework and timeline, briefly describing the construction methods of datasets and the evaluation metrics of the models, discussing in detail the sequence encoding methods and common algorithmic architectures, and demonstrating the computational models based on various types of algorithms and their features and advantages. Finally, we analyze current challenges in predicting protein-protein interactions using deep learning methods, and discuss possible solutions. With the development of deep learning technology, the efficiency of predicting protein-protein interactions has increased dramatically. As a result, there is a need to develop models with stronger generalization and more robust prediction capabilities to aid the prediction of protein-protein interactions in the future.

Key words: protein interactions, deep learning, artificial intelligence, sequence encoding, neural network

摘要：

蛋白质-蛋白质相互作用在细胞信号转导、基因表达和代谢调控等生物过程中发挥重要作用，鉴定蛋白质间的相互作用对于理解复杂生物过程至关重要。预测蛋白质间的相互作用可以为药物发现、蛋白质功能研究和设计等领域提供帮助。近年来，随着人工智能技术的蓬勃发展，深度学习技术在预测蛋白质互作领域做出巨大贡献，其中基于序列的深度学习模型通过学习蛋白质序列信息的深层特征进行互作预测。本文综述了深度学习在基于序列的蛋白质互作预测中的应用，按照算法框架和时间线对该领域进展进行分类归纳，介绍了数据处理、序列编码方法、算法架构以及模型的评估指标等内容，并分析了当前面临的问题以及未来的发展方向。随着深度学习技术的发展，预测蛋白质互作的效率大幅提高，未来需要发展泛化能力更强的预测模型，助力蛋白质互作的预测。

关键词: 蛋白质互作, 深度学习, 人工智能, 序列编码, 神经网络

CLC Number:

Q816

Jingyong ZHU, Junxiang LI, Xuhui LI, Jin ZHANG, Wenjing WU. Advances in applications of deep learning for predicting sequence-based protein interactions[J]. Synthetic Biology Journal, 2024, 5(1): 88-106.

朱景勇, 李钧翔, 李旭辉, 张瑾, 毋文静. 深度学习在基于序列的蛋白质互作预测中的应用进展[J]. 合成生物学, 2024, 5(1): 88-106.

Figures/Tables 9

Fig. 1 General workflow for predicting protein-protein interactions by sequence-based deep learning

Table 1 Public databases and basic information

数据库名称Database name	简介 Description	链接 URL	最近更新 Last update	参考文献Reference
BioGRID	以蛋白质为核心的互作数据库	https://thebiogrid.org	2023	[25]
DIP	通过实验验证和文献确认的PPI互作信息	https://dip.doe-mbi.ucla.edu/dip	2020	[27]
HIPPIE	人工整合的超过60 000条PPI互作数据	https://cbdm.uni-mainz.de/hippie	2023	[30]
HPRD	人类PPI数据库，包括41 327对互作信息	https://hprd.org	2010	[26]
HVIDB	HVIDB重点介绍了48 643个经过实验验证的人与病毒PPI	http://zzdlab.com/hvidb/	2020	[31]
Intact	从22 954份文献中提取1 194 594份互作数据信息	https://www.ebi.ac.uk/intact/home	2022	[32]
Negatome	通过整理文献和分析蛋白质复合物的三维结构得到的非相互作用信息	http://mips.helmholtz-muenchen.de/proj/ppi/negatome	2014	[29]
STRING	蛋白质互作数据库，涉及14 094种生物的67 592 464个蛋白质	https://cn.string-db.org	2023	[33]

Table 2 Protein encoding methods in sequence-based PPI prediction models

类型 Type	编码方法 Encoding method	形式 Form	向量长度 Vector length	参考文献 Reference
基于序列成分	二肽组成 dipeptide composition	一维向量	400	[36]
	氨基酸组成 amino acid composition	一维向量	20	[37]
	伪氨基酸组成 pseudo-amino acid composition	一维向量	20+L，L为最大滞后值	[38]
	联合三元组 conjoint triad	一维向量	343	[16]
基于自相关	自协方差 auto covariance	一维向量	L×理化性质个数，L为最大滞后值	[39]
	交叉协方差 cross covariance	一维向量	L×N×(N-1)，L为最大滞后值，N为理化性质个数	[40]
	自交叉协方差 auto-cross covariance	一维向量	L×N×N，L为最大滞后值，N为理化性质个数	[41]
	Moran自相关 Moran autocorrelation	一维向量	L×理化性质个数，L为最大滞后值	[42]
	Geary自相关 Geary autocorrelation	一维向量	L×理化性质个数，L为最大滞后值	[43]
	物理化学距离变换 physicochemical distance transformation	一维向量	531×β，β为最大间隔距离	[44]
基于进化信息	位置特异性得分矩阵PSSM	二维矩阵	20×N，N为序列长度	[45]
	ACC-PSSM	一维向量	400×L，L为最大滞后值	[46]
	Top-n-grams	一维向量	L×N，L为序列长度，N为阈值	[47]
	BLOSUM62	二维矩阵	20×N，N为序列长度	[48]
	伪位置特异性得分矩阵Pseudo-PSSM	一维向量	40	[49]

Fig.2 Converting amino acid sequences into vector matrices using the one-hot strategy

Table 3 Computational models developed within the past 5 years

类型 Type	发表时间 Published time	模型名称 Model name	算法框架 Algorithm framework	预测类型 Prediction type	参考文献 Reference
基于DNN	2018	—	DNN	PPI	[73]
	2019	EnsDNN	DNN	PPI	[74]
	2019	—	DNN	PPI	[75]
	2019	—	DNN	PPI	[76]
	2019	DeepFEPPI	DNN	PPI	[77]
	2019	DNN-PPI	DNN	PPI	[78]
	2020	—	DNN	PPI	[79]
	2020	—	DNN	PPI	[80]
	2022	DNN-XGB	DNN，XGB	PPI	[81]
	2022	DWPPI	DNN	PPI	[82]
	2022	—	DNN	PPI	[83]
	2022	CT-DNN	DNN	PPI	[84]
	2022	—	DNN	PPI	[85]
基于CNN	2018	DPPI	CNN	PPI	[86]
	2019	—	CNN-RF	PPI	[87]
	2019	—	CNN	PPI	[88]
	2020	Visual	CNN	PPIsite	[89]
	2020	DeepPPISP	TextCNN	PPIsite	[90]
	2020	EnAmDNN	CNN	PPI	[91]
	2020	—	CNN	PPIsite	[92]
	2021	TransPPI	CNN	PPI	[93]
	2021	—	CNN	PPI	[94]
	2021	D-script	CNN	PPI	[95]
	2021	CAMP	CNN	PepPI	[57]
	2021	DeepViral	CNN	PPI	[96]
	2022	DeepTrio	CNN	PPI	[97]
	2023	EResCNN	ResCNN+RF	PPI	[98]
	2023	ProtInteract	CNN	PPI	[99]
基于RNN及变体	2019	—	RNN	PPI	[100]
	2019	DLPred	LSTM	PPIsite	[101]
	2019	—	LSTM	PPI	[102]
	2020	—	LSTM	PPI	[103]
	2021	—	GRNN	PPI	[104]
	2021	LSTM-PHV	LSTM	PPI	[105]
	2023	—	RNN	PPI	[106]
基于注意力机制和Transformer	2021	HANPPIS	Stratified attention	PPIsite	[107]
	2022	Cross-attention PHV	Cross-attention	PPI	[108]
	2022	ADH-PPI	Self-attention	PPI	[109]
	2022	SDNN	Self-attention	PPI	[110]
	2023	EnsemPPIS	Transformer	PPIsite	[111]
基于混合网络	2018	DNN-PPI	CNN+LSTM	PPI	[112]
	2019	PIPR	RNN+CNN	PPI	[113]
	2019	IPPI	DNN+LSTM	PPI	[114]
	2021	DELPHI	RNN+CNN	PPIsite	[115]
	2021	OR-RCNN	RNN+CNN	PPI	[116]
	2023	MM-StackEns	SNN+GNN	PPI	[117]

Fig. 3 Basic network structure of DNN, RNN, and CNN(a) DNN contains an input layer, multiple hidden layers, and an output layer; (b) RNN consists of an input unit (Input), a hidden unit (Hidden), and an output unit (Output) to unfold the RNN in time, with the subscript of the symbols representing the time, indicating that H t receives the inputs from I t and H t-1, and then propagates the results of the computation to O t and H t+1; (c) CNNs include an input layer, multiple convolutional and pooling layers, and fully connected and output layers

Fig. 4 Basic structure of the basic structure of Attention mechanism and Transformer(a) Providing a detailed depiction of the core operations in the attention mechanism, such as Query, Key, and corresponding value for each Key. The similarity scores (S1 to S4) between each Query and Key are first computed using the function F (Q, K), which are then normalized through the Softmax function to derive weights for each Key (A1 to A4). Every Key is associated with a "value" (Value 1 to Value 4). The final attention value is derived by summing up the product of each value and its respective weight. (b) Adapted from reference [72], showcasing the fundamental architecture of the transformer model that is split into encoder and decoder components. The encoder on the left initially receives inputs and pre-processes them through "input embedding" and "positional encoding", and then repeatedly passes them through structural units containing the "multi-head attention" mechanism and feed-forward neural networks. The objective is to transform the input sequence into a context-rich continuous vector representation. On the right, the decoder is responsible for producing an output sequence based on the context provided by the encoder. Its initial input is labeled as "Output", ensuring that the output at the current position in the decoding process only depends on prior information. The decoder also undergoes multiple rounds of "multi-head attention" and feed-forward network processing, ultimately yielding an output probability distribution via linear transformation and a Softmax layer, representing a probability for the potential output

Table 4 Four basic prediction results and their specific meanings

预测结果 Prediction result	英文名称 English name	描述 Description
真阳性	true positive, TP	对阳性样本预测为阳性
假阳性	false positive, FP	对阴性样本预测为阳性
真阴性	true negative, TN	对阴性样本预测为阴性
假阴性	false negative, FN	对阳性样本预测为阴性

Table 5 Six evaluation metrics and their calculation formulas

评估指标 Evaluation index	计算公式 Calculation formula
准确率(accuracy)	$T P + T N T P + T N + F P + F N$
精准率(precision)	$T P T P + F P$
敏感性(sensitivity)	$T P T P + F N$
特异性(specificity)	$T N T N + F P$
F1值(F1 score)	$2 T P 2 T P + F N + F P$
马修斯相关系数(Matthews correlation coefficient)	$T P × T N - F P × F N T P + F P T P + F N T N + F P T N + F N$

Table 5 Six evaluation metrics and their calculation formulas

评估指标 Evaluation index	计算公式 Calculation formula
准确率(accuracy)	$T P + T N T P + T N + F P + F N$
精准率(precision)	$T P T P + F P$
敏感性(sensitivity)	$T P T P + F N$
特异性(specificity)	$T N T N + F P$
F1值(F1 score)	$2 T P 2 T P + F N + F P$
马修斯相关系数(Matthews correlation coefficient)	$T P × T N - F P × F N T P + F P T P + F N T N + F P T N + F N$

References 127

1	HOSSEINI S, ILIE L. PITHIA: protein interaction site prediction using multiple sequence alignments and attention[J]. International Journal of Molecular Sciences, 2022, 23(21): 12814.
2	FIELDS S, STERNGLANZ R. The two-hybrid system: an assay for protein-protein interactions[J]. Trends in Genetics, 1994, 10(8): 286-292.
3	RIGAUT G, SHEVCHENKO A, RUTZ B, et al. A generic protein purification method for protein complex characterization and proteome exploration[J]. Nature Biotechnology, 1999, 17(10): 1030-1032.
4	ZHU H, BILGIN M, BANGHAM R, et al. Global analysis of protein activities using proteome chips[J]. Science, 2001, 293(5537): 2101-2105.
5	ZHAN X K, XIAO M, YOU Z H, et al. Predicting protein-protein interactions based on ensemble learning-based model from protein sequence[J]. Biology, 2022, 11(7): 995.
6	LEE H, DENG M H, SUN F Z, et al. An integrated approach to the prediction of domain-domain interactions[J]. BMC Bioinformatics, 2006, 7(1): 269.
7	HSIN LIU C, LI K C, YUAN S. Human protein-protein interaction prediction by a novel sequence-based co-evolution method: co-evolutionary divergence[J]. Bioinformatics, 2013, 29(1): 92-98.
8	KOVÁCS I A, LUCK K, SPIROHN K, et al. Network-based prediction of protein interactions[J]. Nature Communications, 2019, 10: 1240.
9	SMITH G R, STERNBERG M J E. Prediction of protein-protein interactions by docking methods[J]. Current Opinion in Structural Biology, 2002, 12(1): 28-35.
10	ZHOU J, CUI G Q, HU S D, et al. Graph neural networks: a review of methods and applications[J]. AI open, 2020, 1: 57-81.
11	PENG X Q, WANG J X, PENG W, et al. Protein-protein interactions: detection, reliability assessment and applications[J]. Briefings in Bioinformatics, 2017, 18(5): 798-819.
12	NORTHEY T C, BAREŠIĆ A, MARTIN A C R. IntPred: a structure-based predictor of protein-protein interaction sites[J]. Bioinformatics, 2018, 34(2): 223-229.
13	BARANWAL M, MAGNER A, SALDINGER J, et al. Struct2Graph: a graph attention network for structure based predictions of protein-protein interactions[J]. BMC Bioinformatics, 2022, 23(1): 370.
14	LEE A C L, HARRIS J L, KHANNA K K, et al. A comprehensive review on current advances in peptide drug development and design[J]. International Journal of Molecular Sciences, 2019, 20(10): 2383.
15	WEI L Y, ZOU Q A. Recent progress in machine learning-based methods for protein fold recognition[J]. International Journal of Molecular Sciences, 2016, 17(12): 2118.
16	SHEN J W, ZHANG J, LUO X M, et al. Predicting protein-protein interactions based only on sequences information[J]. Proceedings of the National Academy of Sciences of the United States of America, 2007, 104(11): 4337-4341.
17	ZHOU C, YU H A, DING Y J, et al. Multi-scale encoding of amino acid sequences for predicting protein interactions using gradient boosting decision tree[J]. PLoS One, 2017, 12(8): e0181426.
18	LIN X T, CHEN X W. Heterogeneous data integration by tree-augmented naïve Bayes for protein-protein interactions prediction[J]. Proteomics, 2013, 13(2): 261-268.
19	LI J Q, YOU Z H, LI X, et al. PSPEL: In silico prediction of self-interacting proteins from amino acids sequences using ensemble learning[J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2017, 14(5): 1165-1172.
20	BAI Q F, MA J, LIU S, et al. WADDAICA: a webserver for aiding protein drug design by artificial intelligence and classical algorithm[J]. Computational and Structural Biotechnology Journal, 2021, 19: 3573-3579.
21	CHAI J Y, ZENG H, LI A M, et al. Deep learning in computer vision: a critical review of emerging techniques and application scenarios[J]. Machine Learning with Applications, 2021, 6: 100134.
22	LOPEZ M M, KALITA J. Deep learning applied to NLP[EB/OL]. arXiv, 2017: 1703.03091[2023-10-01]. .
23	TANG B H, PAN Z X, YIN K, et al. Recent advances of deep learning in bioinformatics and computational biology[J]. Frontiers in Genetics, 2019, 10: 214.
24	CONSORTIUM T U. UniProt: a hub for protein information[J]. Nucleic Acids Research, 2015, 43(D1): D204-D212.
25	STARK C, BREITKREUTZ B J, REGULY T, et al. BioGRID: a general repository for interaction datasets[J]. Nucleic Acids Research, 2006, 34(): D535-D539.
26	KESHAVA PRASAD T S, GOEL R, KANDASAMY K, et al. Human protein reference database—2009 update[J]. Nucleic Acids Research, 2009, 37(): D767-D772.
27	XENARIOS I, SALWÍNSKI L, DUAN X J, et al. DIP, the Database of Interacting Proteins: a research tool for studying cellular networks of protein interactions[J]. Nucleic Acids Research, 2002, 30(1): 303-305.
28	KANDEL D, MATIAS Y, UNGER R, et al. Shuffling biological sequences[J]. Discrete Applied Mathematics, 1996, 71(1/2/3): 171-185.
29	BLOHM P, FRISHMAN G, SMIALOWSKI P, et al. Negatome 2.0: a database of non-interacting proteins derived by literature mining, manual annotation and protein structure analysis[J]. Nucleic Acids Research, 2014, 42(D1): D396-D400.
30	ALANIS-LOBATO G, ANDRADE-NAVARRO M A, SCHAEFER M H. HIPPIE v2.0: enhancing meaningfulness and reliability of protein-protein interaction networks[J]. Nucleic Acids Research, 2017, 45(D1): D408-D414.
31	YANG X D, LIAN X Y, FU C, et al. HVIDB: a comprehensive database for human-virus protein-protein interactions[J]. Briefings in Bioinformatics, 2021, 22(2): 832-844.
32	KERRIEN S, ARANDA B, BREUZA L, et al. The IntAct molecular interaction database in 2012[J]. Nucleic Acids Research, 2012, 40(D1): D841-D846.
33	SZKLARCZYK D, GABLE A L, LYON D, et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets[J]. Nucleic Acids Research, 2019, 47(D1): D607-D613.
34	HE J A, WU Y L, PU X M, et al. A transfer-learning-based deep convolutional neural network for predicting leukemia-related phosphorylation sites from protein primary sequences[J]. International Journal of Molecular Sciences, 2022, 23(3): 1741.
35	LIU B, LIU F L, WANG X L, et al. Pse-in-One: a web server for generating various modes of pseudo components of DNA, RNA, and protein sequences[J]. Nucleic Acids Research, 2015, 43(W1): W65-W71.
36	BHASIN M, RAGHAVA G P S. Classification of nuclear receptors based on amino acid composition and dipeptide composition[J]. The Journal of Biological Chemistry, 2004, 279(22): 23262-23266.
37	ZHANG C T, CHOU K C. An optimization approach to predicting protein structural class from amino acid composition[J]. Protein Science, 1992, 1(3): 401-408.
38	CHOU K C. Prediction of protein cellular attributes using pseudo-amino acid composition[J]. Proteins: Structure, Function, and Bioinformatics, 2001, 43(3): 246-255.
39	GUO Y Z, YU L Z, WEN Z N, et al. Using support vector machine combined with auto covariance to predict protein-protein interactions from protein sequences[J]. Nucleic Acids Research, 2008, 36(9): 3025-3030.
40	XIAO X, WANG P, CHOU K C. iNR-PhysChem: a sequence-based predictor for identifying nuclear receptors and their subfamilies via physical-chemical property matrix[J]. PLoS One, 2012, 7(2): e30869.
41	DONG Q W, ZHOU S G, GUAN J H. A new taxonomy-based protein fold recognition approach based on autocross-covariance transformation[J]. Bioinformatics, 2009, 25(20): 2655-2662.
42	FENG Z P, ZHANG C T. Prediction of membrane protein types based on the hydrophobic index of amino acids[J]. Journal of Protein Chemistry, 2000, 19(4): 269-275.
43	SOKAL R R, THOMSON B A. Population structure inferred by local spatial autocorrelation: an example from an Amerindian tribal population[J]. American Journal of Physical Anthropology, 2006, 129(1): 121-131.
44	LIU B, WANG X L, CHEN Q C, et al. Using amino acid physicochemical distance transformation for fast protein remote homology detection[J]. PLoS One, 2012, 7(9): e46633.
45	ZAHIRI J, YAGHOUBI O, MOHAMMAD-NOORI M, et al. PPIevo: protein-protein interaction prediction from PSSM based evolutionary information[J]. Genomics, 2013, 102(4): 237-242.
46	LIU B, WU H, CHOU K C. Pse-in-One 2.0: an improved package of web servers for generating various modes of pseudo components of DNA, RNA, and protein sequences [J]. Natural Science, 2017, 9(4): 67.
47	LIU B, WANG X L, LIN L, et al. A discriminative method for protein remote homology detection and fold recognition combining Top-n-grams and latent semantic analysis[J]. BMC Bioinformatics, 2008, 9: 510.
48	LI A, WANG L R, SHI Y Z, et al. Phosphorylation site prediction with a modified k-nearest neighbor algorithm and BLOSUM62 matrix[C/OL]//2005 IEEE Engineering in Medicine and Biology 27th Annual Conference. January 17-18, 2006, Shanghai. IEEE, 2006: 6075-6078 [2023-10-01]. .
49	SHEN H B, CHOU K C. Nuc-PLoc: a new web-server for predicting protein subnuclear localization by fusing PseAA composition and PsePSSM[J]. Protein Engineering, Design & Selection, 2007, 20(11): 561-567.
50	YANG K K, WU Z, BEDBROOK C N, et al. Learned protein embeddings for machine learning[J]. Bioinformatics, 2018, 34(15): 2642-2648.
51	DUNKER A K, LAWSON J D, BROWN C J, et al. Intrinsically disordered protein[J]. Journal of Molecular Graphics and Modelling, 2001, 19(1): 26-59.
52	FONG J H, SHOEMAKER B A, GARBUZYNSKIY S O, et al. Intrinsic disorder in protein interactions: insights from a comprehensive structural analysis[J]. PLoS Computational Biology, 2009, 5(3): e1000316.
53	TOMPA P, FUXREITER M, OLDFIELD C J, et al. Close encounters of the third kind: disordered domains and the interactions of proteins[J]. BioEssays, 2009, 31(3): 328-335.
54	WEATHERITT R J, GIBSON T J. Linear motifs: lost in (pre)translation[J]. Trends in Biochemical Sciences, 2012, 37(8): 333-341.
55	DYSON H J, WRIGHT P E. Coupling of folding and binding for unstructured proteins[J]. Current Opinion in Structural Biology, 2002, 12(1): 54-60.
56	DUNKER A K, CORTESE M S, ROMERO P, et al. Flexible nets: the roles of intrinsic disorder in protein interaction networks[J]. The FEBS Journal, 2005, 272(20): 5129-5148.
57	LEI Y P, LI S Y, LIU Z Y, et al. A deep-learning framework for multi-level peptide-protein interaction prediction[J]. Nature Communications, 2021, 12: 5465.
58	MÉSZÁROS B, ERDŐS G, DOSZTÁNYI Z. IUPred2A: context-dependent prediction of protein disorder as a function of redox state and protein binding[J]. Nucleic Acids Research, 2018, 46(W1): W329-W337.
59	BASU S, GSPONER J, KURGAN L. DEPICTER2: a comprehensive webserver for intrinsic disorder and disorder function prediction[J]. Nucleic Acids Research, 2023, 51(W1): W141-W147.
60	DEL CONTE A, BOUHRAOUA A, MEHDIABADI M, et al. CAID prediction portal: a comprehensive service for predicting intrinsic disorder and binding regions in proteins[J]. Nucleic Acids Research, 2023, 51(W1): W62-W69.
61	HANSON J, PALIWAL K K, LITFIN T, et al. SPOT-Disorder2: improved protein intrinsic disorder prediction by ensembled deep learning[J]. Genomics, Proteomics & Bioinformatics, 2019, 17(6): 645-656.
62	HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C/OL]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. June 18-23, 2018, Salt Lake City, UT, USA. IEEE, 2018: 7132-7141 [2023-10-01]. .
63	GRAVES A. Long short-term memory[M/OL]// Supervised sequence labelling with recurrent neural networks. Berlin, Heidelberg: Springer, 2012: 37-45 [2023-10-01]. .
64	KAWASHIMA S, POKAROWSKI P, POKAROWSKA M, et al. AAindex: amino acid index database, progress report 2008[J]. Nucleic Acids Research, 2008, 36(): D202-D205.
65	ALTSCHUL S F, MADDEN T L, SCHÄFFER A A, et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs[J]. Nucleic Acids Research, 1997, 25(17): 3389-3402.
66	EDDY S R. Where did the BLOSUM62 alignment score matrix come from?[J]. Nature Biotechnology, 2004, 22(8): 1035-1036.
67	CHOU K C, SHEN H B. MemType-2L: a Web server for predicting membrane proteins and their types by incorporating evolution information through Pse-PSSM[J]. Biochemical and Biophysical Research Communications, 2007, 360(2): 339-345.
68	SUN T L, ZHOU B, LAI L H, et al. Sequence-based prediction of protein protein interaction using a deep-learning algorithm[J]. BMC Bioinformatics, 2017, 18(1): 1-8.
69	MIIKKULAINEN R, LIANG J, MEYERSON E, et al. Evolving deep neural networks[M/OL]//Artificial intelligence in the age of neural networks and brain computing. Amsterdam: Elsevier. 2019: 293-312 [2023-10-01]. .
70	GU J X, WANG Z H, KUEN J, et al. Recent advances in convolutional neural networks[J]. Pattern Recognition, 2018, 77: 354-377.
71	SHERSTINSKY A. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network[J]. Physica D: Nonlinear Phenomena, 2020, 404: 132306.
72	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C/OL]//Proceedings of the 31st International Conference on Neural Information Processing Systems. December 4-9, 2017, Long Beach, California, USA. New York: ACM, 2017: 6000-6010 [2023-10-01]. .
73	GUI Y M, WANG R J, WEI Y Y, et al. Construction of protein-protein interactions model by deep neural networks[C/OL]//Proceedings of the 2018 International Workshop on Bioinformatics, Biochemistry, Biomedical Sciences (BBBS 2018). April 14-15, 2018. Hangzhou City, China. Paris, France: Atlantis Press, 2018: 221-229 [2023-10-01]. .
74	ZHANG L, YU G X, XIA D W, et al. Protein-protein interactions prediction based on ensemble deep neural networks[J]. Neurocomputing, 2019, 324: 10-19.
75	WANG X E, WU Y J, WANG R J, et al. A novel matrix of sequence descriptors for predicting protein-protein interactions from amino acid sequences[J]. PLoS One, 2019, 14(6): e0217312.
76	WANG X, WANG R J, WEI Y Y, et al. A novel conjoint triad auto covariance (CTAC) coding method for predicting protein-protein interaction based on amino acid sequence[J]. Mathematical Biosciences, 2019, 313: 41-47.
77	YAO Y, DU X Q, DIAO Y Y, et al. An integration of deep learning with feature embedding for protein-protein interaction prediction[J]. PeerJ, 2019, 7: e7126.
78	GUI Y M, WANG R J, WEI Y Y, et al. DNN-PPI: a large-scale prediction of protein-protein interactions based on deep neural networks[J]. Journal of Biological Systems, 2019, 27(1): 1-18.
79	HANGGARA F S, ANAM K. Sequence-based protein-protein interaction prediction using greedy layer-wise training of deep neural networks[C/OL]//AIP Conference Proceedings. Novosibirsk, Russia: AIP Publishing, 2020, 2278(1): 020050 [2023-10-01]. .
80	GUI Y M, WANG R J, WANG X E, et al. Using deep neural networks to improve the performance of protein-protein interactions prediction[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2020, 34(13): 2052012.
81	MAHAPATRA S, GUPTA V R, SAHU S S, et al. Deep neural network and extreme gradient boosting based hybrid classifier for improved prediction of protein-protein interaction[J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2022, 19(1): 155-165.
82	PAN J, YOU Z H, LI L P, et al. DWPPI: a deep learning approach for predicting protein-protein interactions in plants based on multi-source information with a large-scale biological network[J]. Frontiers in Bioengineering and Biotechnology, 2022, 10: 807522.
83	KHANH LE N Q, KHA Q H. Prediction of protein-protein interactions through deep learning based on sequence feature extraction and interaction network[C/OL]//2022 IEEE Biomedical Circuits and Systems Conference (BioCAS). October 13-15, 2022, Taipei, China. IEEE, 2022: 539-543 [2023-10-01]. .
84	WANG J H, WANG X D, CHEN W T. Prediction of protein interactions based on CT-DNN[C/OL]//Proceedings of the 2022 9th International Conference on Biomedical and Bioinformatics Engineering. November 10-13, 2022, Kyoto, Japan. New York: ACM, 2022: 81-87 [2023-10-01]. .
85	MEWARA B, LALWANI S. Strengthening auto-feature engineering of deep learning architecture in protein-protein interaction prediction[C/OL]//SHARMA H, SHRIVASTAVA V, KUMARI BHARTI K, et al. Communication and intelligent systems: Proceedings of ICCIS 2021. Singapore: Springer, 2022: 1205-1216 [2023-10-01]. .
86	HASHEMIFAR S, NEYSHABUR B, KHAN A A, et al. Predicting protein-protein interactions through sequence-based deep learning[J]. Bioinformatics, 2018, 34(17): i802-i810.
87	WANG L, WANG H F, LIU S R, et al. Predicting protein-protein interactions from matrix-based protein sequence using convolution neural network and feature-selective rotation forest[J]. Scientific Reports, 2019, 9: 9848.
88	WANG Y B, YOU Z H, YANG S, et al. A high efficient biological language model for predicting protein-protein interactions[J]. Cells, 2019, 8(2): 122.
89	WARDAH W, DEHZANGI A, TAHERZADEH G, et al. Predicting protein-peptide binding sites with a deep convolutional neural network[J]. Journal of Theoretical Biology, 2020, 496: 110278.
90	ZENG M, ZHANG F H, WU F X, et al. Protein-protein interaction site prediction through combining local and global features with deep neural networks[J]. Bioinformatics, 2020, 36(4): 1114-1120.
91	LI F F, ZHU F, LING X H, et al. Protein interaction network reconstruction through ensemble deep learning with attention mechanism[J]. Frontiers in Bioengineering and Biotechnology, 2020, 8: 390.
92	XIE Z Y, DENG X Y, SHU K X. Prediction of protein-protein interaction sites using convolutional neural network and improved data sets[J]. International Journal of Molecular Sciences, 2020, 21(2): 467.
93	YANG X D, YANG S P, LIAN X Y, et al. Transfer learning via multi-scale convolutional neural layers for human-virus protein-protein interaction prediction[J]. Bioinformatics, 2021, 37(24): 4771-4778.
94	WANG Y, LI Z C, ZHANG Y F, et al. Performance improvement for a 2D convolutional neural network by using SSC encoding on protein-protein interaction tasks[J]. BMC Bioinformatics, 2021, 22(1): 184.
95	SLEDZIESKI S, SINGH R, COWEN L, et al. D-SCRIPT translates genome to phenome with sequence-based, structure-aware, genome-scale predictions of protein-protein interactions[J]. Cell Systems, 2021, 12(10): 969-982.e6.
96	LIU-WEI W, KAFKAS Ş, CHEN J, et al. DeepViral: prediction of novel virus-host interactions from protein sequences and infectious disease phenotypes[J]. Bioinformatics, 2021, 37(17): 2722-2729.
97	HU X T, FENG C, ZHOU Y C, et al. DeepTrio: a ternary prediction system for protein-protein interaction using mask multiple parallel convolutional neural networks[J]. Bioinformatics, 2022, 38(3): 694-702.
98	GAO H L, CHEN C, LI S Y, et al. Prediction of protein-protein interactions based on ensemble residual convolutional neural network[J]. Computers in Biology and Medicine, 2023, 152: 106471.
99	SOLEYMANI F, PAQUET E, VIKTOR H L, et al. ProtInteract: a deep learning framework for predicting protein-protein interactions[J]. Computational and Structural Biotechnology Journal, 2023, 21: 1324-1348.
100	GONZALEZ-LOPEZ F, MORALES-CORDOVILLA J A, VILLEGAS-MORCILLO A, et al. End-to-end prediction of protein-protein interaction based on embedding and recurrent neural networks[C/OL]//2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). December 3-6, 2018, Madrid, Spain. IEEE, 2019: 2344-2350 [2023-10-01]. .
101	ZHANG B Z, LI J Y, QUAN L J, et al. Sequence-based prediction of protein-protein interaction sites by simplified long short-term memory network[J]. Neurocomputing, 2019, 357: 86-100.
102	ALAKUS T B, TURKOGLU I. Prediction of protein-protein interactions with LSTM deep learning model[C/OL]//2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT). October 11-13, 2019, Ankara, Turkey. IEEE, 2019: 1-5 [2023-10-01]. .
103	YANG L, HAN Y K, ZHANG H X, et al. Prediction of protein-protein interactions with local weight-sharing mechanism in deep learning[J]. BioMed Research International, 2020, 2020: 5072520.
104	XU H X, XU D, ZHANG N Q, et al. Protein-protein interaction prediction based on spectral radius and general regression neural network[J]. Journal of Proteome Research, 2021, 20(3): 1657-1665.
105	TSUKIYAMA S, HASAN M M, FUJII S, et al. LSTM-PHV: prediction of human-virus protein-protein interactions by LSTM with word2vec[J]. Briefings in Bioinformatics, 2021, 22(6): bbab228.
106	MEWARA B, LALWANI S. Sequence-based prediction of protein-protein interaction using auto-feature engineering of RNN-based model[J]. Research on Biomedical Engineering, 2023, 39(1): 259-272.
107	TANG M L, WU L X, YU X Y, et al. Prediction of protein-protein interaction sites based on stratified attentional mechanisms[J]. Frontiers in Genetics, 2021, 12: 784863.
108	TSUKIYAMA S, KURATA H. Cross-attention PHV: prediction of human and virus protein-protein interactions using cross-attention-based neural networks[J]. Computational and Structural Biotechnology Journal, 2022, 20: 5564-5573.
109	ASIM M N, IBRAHIM M ALI, MALIK M I, et al. ADH-PPI: an attention-based deep hybrid model for protein-protein interaction prediction[J]. iScience, 2022, 25(10): 105169.
110	LI X, HAN P F, WANG G, et al. SDNN-PPI: self-attention with deep neural network effect on protein-protein interaction prediction[J]. BMC Genomics, 2022, 23(1): 474.
111	MOU M J, PAN Z Q, ZHOU Z M, et al. A transformer-based ensemble framework for the prediction of protein-protein interaction sites[J]. Research, 2023, 6: 0240.
112	LI H, GONG X J, YU H A, et al. Deep neural network based predictions of protein interactions using primary sequences[J]. Molecules, 2018, 23(8): 1923.
113	CHEN M H, JU C J T, ZHOU G Y, et al. Multifaceted protein-protein interaction prediction based on Siamese residual RCNN[J]. Bioinformatics, 2019, 35(14): i305-i314.
114	GUO Y, CHEN X. A deep learning framework for improving protein interaction prediction using sequence properties[EB/OL]. bioRxiv, 2019: 843755[2023-10-01]. .
115	LI Y W, GOLDING G B, ILIE L. DELPHI: accurate deep ensemble model for protein interaction sites prediction[J]. Bioinformatics, 2021, 37(7): 896-904.
116	XU W X, GAO Y Y, WANG Y, et al. Protein-protein interaction prediction based on ordinal regression and recurrent convolutional neural networks[J]. BMC Bioinformatics, 2021, 22(): 485.
117	ALBU A I, BOCICOR M I, CZIBULA G. MM-StackEns: a new deep multimodal stacked generalization approach for protein-protein interaction prediction[J]. Computers in Biology and Medicine, 2023, 153: 106526.
118	DU X Q, SUN S W, HU C L, et al. DeepPPI: boosting prediction of protein-protein interactions with deep neural networks[J]. Journal of Chemical Information and Modeling, 2017, 57(6): 1499-1510.
119	YANG L, XIA J F, GUI J E. Prediction of protein-protein interactions from protein sequence using local descriptors[J]. Protein & Peptide Letters, 2010, 17(9): 1085-1090.
120	CHUNG J, GULCEHRE C, CHO K, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[EB/OL]. arXiv, 2014: 1412.3555[2023-10-01]. .
121	NAMBIAR A, LIU S, HEFLIN M, et al. Transformer neural networks for protein family and interaction prediction tasks[J]. Journal of Computational Biology, 2023, 30(1): 95-111.
122	JAISWAL A, BABU A R, ZADEH M Z, et al. A survey on contrastive self-supervised learning[J]. Technologies, 2020, 9(1): 2.
123	HU X, CHU L Y, PEI J, et al. Model complexity of deep learning: a survey[J]. Knowledge and Information Systems, 2021, 63(10): 2585-2619.
124	OUYANG L, WU J, JIANG X, et al. Training language models to follow instructions with human feedback[C/OL]//Advances in Neural Information Processing Systems 35 (NeurIPS 2022), 2022, 35: 27730-27744 [2023-10-01]. .
125	MIKOLOV T, CHEN K, CORRADO G, et al. Efficient estimation of word representations in vector space[EB/OL]. arXiv, 2013: 1301.3781[2023-10-01]. .
126	ELNAGGAR A, HEINZINGER M, DALLAGO C, et al. ProtTrans: toward understanding the language of life through self-supervised learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(10): 7112-7127.
127	JUMPER J, EVANS R, PRITZEL A, et al. Highly accurate protein structure prediction with AlphaFold[J]. Nature, 2021, 596(7873): 583-589.

[1]	Ru LEI, Hui TAO, Tiangang LIU. Deep genome mining boosts the discovery of microbial terpenoids [J]. Synthetic Biology Journal, 2024, 5(3): 507-526.
[2]	Mengyu XI, Yiling HU, Yucheng GU, Huiming GE. Genome mining-directed discovery for natural medicinal products [J]. Synthetic Biology Journal, 2024, 5(3): 447-473.
[3]	Sheng WANG, Zechen WANG, Weihua CHEN, Ke CHEN, Xiangda PENG, Fafen OU, Liangzhen ZHENG, Jinyuan SUN, Tao SHEN, Guoping ZHAO. Design of synthetic biology components based on artificial intelligence and computational biology [J]. Synthetic Biology Journal, 2023, 4(3): 422-443.
[4]	Liqi KANG, Pan TAN, Liang HONG. Enzyme engineering in the age of artificial intelligence [J]. Synthetic Biology Journal, 2023, 4(3): 524-534.
[5]	Qiaozhen MENG, Fei GUO. Applications of foldability in intelligent enzyme engineering and design: take AlphaFold2 for example [J]. Synthetic Biology Journal, 2023, 4(3): 571-589.
[6]	Yiming TANG, Yifei YAO, Zhongyuan YANG, Yun ZHOU, Zichao WANG, Guanghong WEI. Pathological aggregation and liquid-liquid phase separation of proteins associated with neurodegenerative diseases [J]. Synthetic Biology Journal, 2023, 4(3): 590-610.
[7]	Zhihang CHEN, Menglin JI, Yifei QI. Research progress of artificial intelligence in desiging protein structures [J]. Synthetic Biology Journal, 2023, 4(3): 464-487.
[8]	Qilong LAI, Shuai YAO, Yuguo ZHA, Hong BAI, Kang NING. Microbiome-based biosynthetic gene cluster data mining techniques and application potentials [J]. Synthetic Biology Journal, 2023, 4(3): 611-627.
[9]	Yidong SONG, Qianmu YUAN, Yuedong YANG. Application of deep learning in protein function prediction [J]. Synthetic Biology Journal, 2023, 4(3): 488-506.
[10]	Jingwei LYU, Zixin DENG, Qi ZHANG, Wei DING. Identification of RiPPs precursor peptides and cleavage sites based on deep learning [J]. Synthetic Biology Journal, 2022, 3(6): 1262-1276.
[11]	Jiahao BIAN, Guangyu YANG. Artificial intelligence-assisted protein engineering [J]. Synthetic Biology Journal, 2022, 3(3): 429-444.
[12]	Ye WANG, Haochen WANG, Minghao YAN, Guanhua HU, Xiaowo WANG. Design of biomolecular sequences by artificial intelligence [J]. Synthetic Biology Journal, 2021, 2(1): 1-14.

Advances in applications of deep learning for predicting sequence-based protein interactions

深度学习在基于序列的蛋白质互作预测中的应用进展

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 127

Related Articles 12

Recommended Articles

Metrics

Comments