计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (6A): 250700116-10.doi: 10.11896/jsjkx.250700116

• 大数据&数据科学 • 上一篇    下一篇

基于二阶多通道有向图卷积的基因调控推断方法

沈亚婕1, 王基书2, 金魁1, 字桐1, 唐明靖1,3   

  1. 1 云南师范大学信息学院 昆明 650500
    2 云南大学信息学院 昆明 650500
    3 云南特色生物资源高值化利用教育部工程中心(云南师范大学) 昆明 650500
  • 出版日期:2026-06-16 发布日期:2026-06-12
  • 通讯作者: 唐明靖(tmj@ynnu.edu.cn)
  • 作者简介:(1834722187@qq.com)
  • 基金资助:
    国家自然科学基金项目(61862067);云南省基础研究专项重点项目(202501AS070007)

Second-order Multi-channel Directed Graph Convolution for Gene Regulatory Inference

SHEN Yajie1, WANG Jishu2, JIN Kui1, ZI Tong1, TANG Mingjing1,3   

  1. 1 School of Information Science and Technology,Yunnan Normal University,Kunming 650500,China
    2 School of Information Science and Engineering,Yunnan University,Kunming 650500,China
    3 Engineering Research Center of High-Value Utilization of Yunnan Characteristic Biological Resources,Ministry of Education(Yunnan Normal University),Kunming 650500,China
  • Published:2026-06-16 Online:2026-06-12
  • About author:SHEN Yajie,born in 2001,master,is a member of CCF(No.Z9515G).Her main research interests include deep learning and bioinformatics.
    TANG Mingjing,born in 1978,Ph.D,professor,master's supervisor.His main research interests include machine lear-ning,computational intelligence and bioinformatics.
  • Supported by:
    National Natural Science Foundation of China(61862067) and Key Project of Basic Research in Yunnan Province(202501AS070007).

PDF (PC)

37

摘要： 单细胞RNA(scRNA-seq)测序技术的快速发展使得单细胞基因表达数据呈指数级增长,从而使研究人员获取了大量的基因表达数据集。因此,需要开发能够利用这些数据从而发现基因间潜在调控关系的计算方法。近年来,随着深度学习的发展及已知调控关系数据的增长,大量基于深度学习的有监督推断方法被提出,尤其是基于图神经网络的方法。然而,目前的这些方法大多将先验调控网络建模为一个无向图,忽略了基因间调控关系的有向性,导致有向信息无法被提取。此外,由于已知的调控信息有限,很多具有相同或相近表达模式的基因间不一定存在已知的直接联系。而目前的方法通常仅提取一阶邻域的信息,这可能导致无法充分捕获先验网络和表达数据中更丰富的信息。针对以上问题,提出一种基于二阶多通道有向图卷积的基因调控推断方法,其基于先验调控网络,构建了一阶邻近矩阵、二阶入度邻近矩阵和二阶出度邻近矩阵,并利用方向性拉普拉斯矩阵来表征有向图的结构,最终进一步提升了网络推断的性能和对复杂调控模式的建模能力。在多个数据集上进行的多指标的实验结果表明,与现有工作相比,所提出的方法能够更准确地预测基因间潜在的调控关系。同时,充分的消融实验也进一步证明了,所提出的方法中包含的不同模块均有助于提升模型的性能。

关键词: 图, 有向图神经网络, 基因调控网络, 二阶邻近, 图卷积

Abstract: The rapid development of single-cell RNA sequencing(scRNA-seq) technology has led to an exponential increase in single-cell gene expression data,thereby resulting in the accumulation of extensive gene expression datasets.Therefore,there is a pressing need for computational methods capable of leveraging these datasets to uncover potential regulatory relationships between genes.In recent years,the advancements in deep learning and the expansion of known regulatory relationship datasets have facilitated the development of numerous supervised inference methods,particularly those based on graph neural networks(GNNs).However,most of these current methods model the prior regulatory network as an undirected graph,neglecting the directed nature of regulatory relationships between genes,which makes it impossible to extract directional information.In addition,due to the limited availability of known regulatory information,genes with similar or correlated expression patterns may not ne-cessarily have known direct connections.Most current methods rely solely on extracting first-order neighborhood information,which may hinder the ability to fully capture the richer information contained in prior networks and expression data.To address these challenges,this paper proposes a gene regulatory inference method based on second-order multi-channel directed graph convolution.By leveraging prior regulatory networks,the method constructs a first-order adjacency matrix,a second-order in-degree adjacency matrix,and a second-order out-degree adjacency matrix.Additionally,it employs a directional Laplacian matrix to more accurately represent the structure of directed graphs,thereby enhancing the performance of network inference and the ability to model complex regulatory patterns.Experimental results with multiple datasets and evaluation metrics demonstrate that the proposed method can more accurately predict potential regulatory relationships between genes compared to existing work.Meanwhile,extensive ablation studies confirm that these different modules of the proposed method contribute to improving model performance.

Key words: Graph, Directed graph neural network, Gene regulatory network, Second-order neighborhood, Graph convolution

中图分类号: 

  • TP311.13
[1] DAVIDSON E H.The regulatory genome:gene regulatory networks in development and evolution [M].Amsterdam:Elsevier,2010.
[2] DONG J,LI J,WANG F.Deep learning in gene regulatory network inference:a survey [J].IEEE/ACM Transactions on Computational Biology and Bioinformatics,2024,21(6):2089-2101.
[3] CHEN G,NING B,SHI T.Single-cell RNA-seq technologiesand related computational data analysis [J].Frontiers in Gen-etics,2019,10:317.
[4] WANG J,MA A,MA Q,et al.Inductive inference of gene regulatory network using supervised and semi-supervised graph neural networks [J].Computational and Structural Biotechnology Journal,2020,18:3335-3343.
[5] HOU L,LIU J H,YU X,et al.Review of Graph Neural Net-works [J].Computer Science,2024,51(6):282-298.
[6] CHEN B,LI J L.Network Representation Learning ModelBased on Attention Mechanism for Fusing Multi-order Neighborhood Information [J].Journal of Chinese Computer Systems,2021,42(4):761-765.
[7] TONG Z,LIANG Y,SUN C,et al.Directed graph convolutional network [J].arXiv:2004.13970,2020.
[8] PENG L,PENG M,LIAO B,et al.The advances and challenges of deep learning application in biological big data processing [J].Current Bioinformatics,2018,13(4):352-359.
[9] HUYNH-THU V A,IRRTHUM A,WEHENKEL L,et al.In-ferring regulatory networks from expression data using tree-based methods [J].PLoS One,2010,5(9):e12776.
[10] PARK S,KIM J M,SHIN W,et al.BTNET:boosted tree based gene regulatory network inference algorithm using time-course measurement data [J].BMC Systems Biology,2018,12:69-77.
[11] MOERMAN T,AIBAR SANTOS S,BRAVO GONZÁLEZ-BLAS C,et al.GRNBoost2 and Arboreto:efficient and scalable inference of gene regulatory networks [J].Bioinformatics,2019,35(12):2159-61.
[12] HAURY A C,MORDELET F,VERA-LICONA P,et al.TI-GRESS:trustful inference of gene regulation using stability selection [J].BMC Systems Biology,2012,6:1-17.
[13] LIU F.Reconstruction Algorithm of Gene Regulatory Network Based on Gene Expression Profiles [D].Xi'an:Northwestern Polytechnical University,2018.
[14] STEINBACH M,TAN P N.kNN:k-nearest neighbors [M]//The Top Ten Algorithms in Data Mining.2009:165-76.
[15] BEN-GAL I.Bayesian networks [M]//Encyclopedia of Statistics in Quality and Reliability.2008.
[16] LANGFELDER P,HORVATH S.WGCNA:an R package for weighted correlation network analysis [J].BMC Bioinformatics,2008,9:1-13.
[17] YU P,XIAO S,XIN X,et al.Spatiotemporal clustering of the epigenome reveals rules of dynamic gene regulation [J].Genome Research,2013,23(2):352-364.
[18] ZHOU L,PAN S,WANG J,et al.Machine learning on big data:Opportunities and challenges [J].Neurocomputing,2017,237:350-361.
[19] GROSSBERG S.Recurrent neural networks [J].Scholarpedia,2013,8(2):1888.
[20] VASWANI A.Attention is all you need;proceedings of the Advances in Neural Information Processing Systems [C]//PMLR.2017.
[21] ZHAO Y,JOSHI P,SHIN D G.Recurrent neural network for gene regulation network construction on time series expression data [C]//Proceedings of the 2019 IEEE International Confe-rence on Bioinformatics and Biomedicine(BIBM).IEEE,2019.
[22] XU J,ZHANG A,LIU F,et al.STGRNS:an interpretabletransformer-based method for inferring gene regulatory networks from single-cell transcriptomic data [J].Bioinformatics,2023,39(4):btad165.
[23] YUAN Y,BAR-JOSEPH Z.Deep learning for inferring gene relationships from single-cell expression data [J].Proceedings of the National Academy of Sciences,2019,116(52):27151-27158.
[24] CHEN J,CHEONG C,LAN L,et al.DeepDRIM:a deep neural network to reconstruct cell-type-specific gene regulatory network using single-cell RNA-seq data [J].Briefings in Bioinformatics,2021,22(6):bbab325.
[25] XU Y,CHEN J,LYU A,et al.dynDeepDRIM:a dynamic deep learning model to infer direct regulatory interactions using time-course single-cell gene expression data [J].Briefings in Bioinformatics,2022,23(6):bbac424.
[26] CHEN G,LIU Z P.Graph attention network for link prediction of gene regulations from single-cell RNA-sequencing data [J].Bioinformatics,2022,38(19):4522-4529.
[27] YUAN L,ZHAO L,JIANG Y,et al.scMGATGRN:a multiview graph attention network-based method for inferring gene regulatory networks from single-cell transcriptomic data [J].Briefings in Bioinformatics,2024,25(6):bbae526.
[28] MAO G,PANG Z,ZUO K,et al.Predicting gene regulatorylinks from single-cell RNA-seq data using graph neural networks [J].Briefings in Bioinformatics,2023,24(6):bbad414.
[29] PRATAPA A,JALIHAL A P,LAW J N,et al.Benchmarking algorithms for gene regulatory network inference from single-cell transcriptomic data [J].Nature Methods,2020,17(2):147-154.
[30] GARCIA-ALONSO L,HOLLAND C H,IBRAHIM M M,et al.Benchmark and integration of resources for the estimation of human transcription factor activities [J].Genome Research,2019,29(8):1363-1375.
[31] LIU Z P,WU C,MIAO H,et al.RegNetwork:an integrated database of transcriptional and post-transcriptional regulatory networks in human and mouse [J].Database,2015,2015:bav095.
[32] HAN H,CHO J W,LEE S,et al.TRRUST v2:an expanded re-ference database of human and mouse transcriptional regulatory interactions [J].Nucleic Acids Research,2018,46(D1):D380-D386.
[33] XU H,BAROUKH C,DANNENFELSER R,et al.ESCAPE:database for integrating high-content published data collected from human and mouse embryonic stem cells [J].Database,2013,2013:bat045.
[34] KC K,LI R,CUI F,et al.GNE:a deep learning framework for gene network inference by aggregating biological information [J].BMC Systems Biology,2019,13:1-14.
[35] SHU H,ZHOU J,LIAN Q,et al.Modeling gene regulatory networks using neural network architectures [J].Nature Computational Science,2021,1(7):491-501.
[36] HONG Y,DOWNEY T,EU K W,et al.A ‘metastasis-prone' signature for early-stage mismatch-repair proficient sporadic colorectal cancer patients and its implications for possible therapeutics[J].Clinical & Experimental Metastasis,2010,27(2):83-90.
[37] LIU Z P,WU C,MIAO H,et al.RegNetwork:an integrated database of transcriptional and post-transcriptional regulatory networks in human and mouse[J].Database:The Journal of Biolo-gical Databases and Curation,2015:bav095.
[38] HAN H,CHO J W,LEE S,et al.TRRUST v2:an expanded re-ference database of human and mouse transcriptional regulatory interactions[J].Nucleic Acids Research,2018,46(D1):D380-D386.
[39] WANG H,HUANG R,GUO W,et al.RNA-binding proteinCELF1 enhances cell migration,invasion,and chemoresistance by targeting ETS2 in colorectal cancer[J].Clinical Science,2020,134(14):1973-1990.
[40] LIU Y,WU H,LUO T,et al.The SOX9-MMS22L Axis Promotes Oxaliplatin Resistance in Colorectal Cancer[J].Frontiers in Molecular Biosciences,2021,8:646542.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发