| 30 | 0 | 100 |
| 下载次数 | 被引频次 | 阅读次数 |
长链非编码RNA(long non-coding RNA, lncRNA)能够与DNA相互作用,作为关键的调控元件,在癌症和代谢紊乱等复杂疾病的发生发展中起着核心作用。尽管基于深度学习的计算方法已开始应用于全基因组lnc RNA-DNA结合位点的表征,但现有模型难以同时捕获长程与短程染色质依赖关系,这仍是当前领域的瓶颈。本研究构建了一个深度学习框架Binder,可以通过整合卷积神经网络(convolutional neural networks, CNN)与多头注意力机制,准确预测全基因组范围内的lnc RNADNA结合位点。Binder的性能已在Ch IRP-seq数据集上得到了验证,其平均受试者工作特征曲线下面积(area under the receiver operating characteristic curve, AUC)达到0.870,优于CNN＿deeper等现有最先进的模型。此外, Binder自主捕捉到了与已知转录因子结合位点(如TAL1::TCF3)一致的调控基序。本研究突破了以往方法在建模全局序列上下文方面的局限,并提供了一个用户友好的Web应用工具,以便于研究lnc RNA调控机制,为辅助疾病治疗和药物研发打下基础。
Abstract：Long non-coding RNA( lnc RNA) serve as critical regulatory elements by interacting with DNA, playing central roles in the pathogenesis of complex diseases such as cancer and metabolic disorders. Although deep learning-based computational approaches have emerged for characterizing genome-wide lnc RNA-DNA binding sites, existing models struggle to simultaneously capture long-range and short-range chromatin dependencies, which remains a major bottleneck in the field. This study constructs Binder, a deep learning framework that accurately predicts genome-wide lnc RNA-DNA binding sites by integrating convolutional neural networks( CNNs) with a multi-head attention mechanism. The performance of Binder has been validated on the Ch IRP-seq dataset, achieving an average area under the area under the receiver operating characteristic curve( AUC) of 0. 870 and outperforming state-of-the-art models such as CNN＿deeper. Furthermore, Binder autonomously identified regulatory motifs highly consistent with known transcription factor binding sites, such as TAL1: : TCF3, demonstrating its ability to capture complex global sequence contexts. This study addresses the limitations of previous methods in modeling global sequence contexts and provides a user-friendly web application to facilitate research on lnc RNA regulatory mechanisms, thereby laying the foundation for aiding disease treatment and drug discovery.
黄浈琪,王治中,丘兆旺,等,2025.长链非编码RNASNHG16介导的海马神经元线粒体自噬对糖尿病认知障碍影响的研究[J].中国糖尿病杂志,33(9):692-700.[Huang Z Q,Wang Z Z,Qiu Z W,et al.,2025.Effect of long non-coding RNA SNHG16 mediated mitophagy on diabetesassociated cognitive impairment[J].Chinese Journal of Diabetes,33(9):692-700.]
刘苏慧,段丽娟,李超,等,2025.Lnc RNA CTBP1-AS2调节mi R-433-3p/DPP8信号轴对急性髓系白血病细胞恶性生物学行为的影响[J].中国免疫学杂志,41(11):2631-2636.[Liu S H,Duan L J,Li C,et al.,2025.Effect of Lnc RNA CTBP1-AS2 on malignant biological behavior of acute myeloid leukemia cells by regulating mi R-433-3p/DPP8 signaling axis[J].Chinese Journal of Immunology,41(11):2631-2636.]
杨峰,易凡,曹慧青,等,2014.长链非编码RNA研究进展[J].遗传,36(5):456-468.[Yang F,Yi F,Cao H Q,et al.,2014.The emerging landscape of long non-coding RNAs[J].Hereditas,36(5):456-468.]
Al-Selwi S M,Hassan M F,Abdulkadir S J,et al.,2023.LSTMinefficiency in long-term dependencies regression problems[J].J.Adv.Res.Appl.Sci.Eng.Technol.,30(3):16-31.
Alfeghaly C,Sanchez A,Rouget R,et al.,2021.Implication of repeat insertion domains in the trans-activity of the long noncoding RNA ANRIL[J].Nucleic Acids Res.,49 (9):4954-4970.
Alharbi W S,Rashid M,2022.A review of deep learning applications in human genomics using next-generation sequencing data[J].Hum.Genom.,16:26.
Alipanahi B,Delong A,Weirauch M T,et al.,2015.Predicting the sequence specificities of DNA-and RNA-binding proteins by deep learning[J].Nat.Biotechnol.,33(8):831-838.
Airlangga G,2024.Comparative analysis of deep learning architectures for DNA sequence classification:performance evaluation and model insights[J].J.Comput.Syst.Inform.(Josyc),5(3):709-718.
Andrabi S B A,Kalim U U,Palani S,et al.,2024.Long noncoding RNA LIRIL2R modulates FOXP3 levels and suppressive function of human CD4+regulatory T cells by regulating IL2RA[J].Proc.Natl.Acad.Sci.U.S.A.,121(23):e2315363121.
Chen K M,Cofer E M,Zhou J,et al.,2019.Selene:a Py Torchbased deep learning library for sequence data[J].Nat.Meth.,16(4):315-318.
Chen X,Yan C C,Zhang X,et al.,2017.Long non-coding RNAs and complex diseases:from experimental results to computational models[J].Brief.Bioinform.,18(4):558-576.
Chiu H S,Somvanshi S,Chang C T,et al.,2025.Coordinated regulation by lnc RNAs results in tight lnc RNA-target couplings[J].Cell Genom.,5(8):100927.
Chu C,Qu K,Zhong F L,et al.,2011.Genomic maps of long noncoding RNA occupancy reveal principles of RNA-chromatin interactions[J].Mol.Cell,44(4):667-678.
Chu C,Quinn J,Chang H Y,2012.Chromatin isolation by RNApurification (Ch IRP)[J].J.Vis.Exp.,(61):e3912.
Congrains A,Kamide K,Oguro R,et al.,2012.Genetic variants at the 9p21 locus contribute to atherosclerosis through modulation of ANRIL and CDKN2A/B[J].Atherosclerosis,220(2):449-455.
Cui F Y,Yue Y,Zhang Y,et al.,2020.Advancing biosensors with machine learning[J].ACS Sens.,5(11):3346-3364.
Derrien T,Johnson R,Bussotti G,et al.,2012.The GENCODEv7 catalog of human long noncoding RNAs:analysis of their gene structure,evolution,and expression[J].Genome Res.,22(9):1775-1789.
Ding Y,Robinson N,Tong C X,et al.,2024.LGGNet:learning from local-global-graph representations for brain-computer interface[J].IEEE Trans.Neural Netw.Learn.Syst.,35(7):9773-9786.
Dunn-Davies H,Dudnakova T,Nogara A,et al.,2024.Control of endothelial cell function and arteriogenesis by MEG3:EZH2 epigenetic regulation of integrin expression[J].Mol Ther.Nucleic Acids,35(2):102173.
Engreitz J M,Pandya-Jones A,Mc Donel P,et al.,2013.The Xist lnc RNA exploits three-dimensional genome architecture to spread across the X chromosome[J].Science,341(6147):1237973.
Esteller M,2011.Non-coding RNAs in human disease[J].Nat.Rev.Genet.,12(12):861-874.
Gil N,Ulitsky I,2020.Regulation of gene expression by cis-acting long non-coding RNAs[J].Nat.Rev.Genet.,21 (2):102-117.
Gupta R A,Shah N,Wang K C,et al.,2010.Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis[J].Nature,464(7291):1071-1076.
Gupta S,Stamatoyannopoulos J A,Bailey T L,et al.,2007.Quantifying similarity between motifs[J].Genome Biol.,8(2):R24.
Jeong C B,Cho H,Park D,2026.Deep learning application for genomic data analysis[J].BMB Rep.,59(1):60-68.
Jing X,Han J M,Zhang J H,et al.,2021.Long non-coding RNA MEG3 promotes cisplatin-induced nephrotoxicity through regulating AKT/TSC/m TOR-mediated autophagy[J].Int.J.Biol.Sci.,17(14):3968-3980.
Kang Z H,Li R M,Liu C,et al.,2024.M6A-modified cen RNAstabilizes CENPA to ensure centromere integrity in cancer cells[J].Cell,187(21):6035-6054.e27.
Kiranyaz S,Avci O,Abdeljaber O,et al.,2021.1D convolutional neural networks and applications:a survey[J].Mech.Syst.Signal Process.,151:107398.
Kumar S,Goel P,2025.Deep genomics:deep learning-based analysis of genome-sequenced data for identification of gene alterations[M]//Mandal S.Artificial intelligence (AI) in cell and genetic engineering.New York,NY:Springer US:335-367.
Kuo C C,Hänzelmann S,Sentürk Cetin N,et al.,2019.Detection of RNA-DNA binding sites in long noncoding RNAs[J].Nucleic Acids Res.,47(6):e32.
Li H,2013.Aligning sequence reads,clone sequences and assembly contigs with BWA-MEM[PP/OL].Ar Xiv.(2013-05-26)[2026-05-25].https://doi.org/10.48550/ar Xiv.1303.3997.
Li H,Handsaker B,Wysoker A,et al.,2009.The sequence alignment/map format and SAMtools[J].Bioinformatics,25(16):2078-2079.
Li W Q,Huo Y,Ren Y,et al.,2023.Deciphering the functional long non-coding RNAs derived from micro RNA loci[J].Adv.Sci.,10(33):2203987.
Marchese F P,Raimondi I,Huarte M,2017.The multidimensional mechanisms of long noncoding RNA function[J].Genome Biol.,18(1):206.
Mattick J S,Amaral P P,Carninci P,et al.,2023.Long noncoding RNAs:definitions,functions,challenges and recommendations[J].Nat.Rev.Mol.Cell Biol.,24(6):430-447.
Mercer T R,Dinger M E,Mattick J S,2009.Long non-coding RNAs:insights into functions[J].Nat.Rev.Genet.,10(3):155-159.
Núñez-Martínez H N,Tapia-Urzúa G,Cerecedo-CastilloÁJ,et al.,2025.The lnc RNA DUBR is regulated by CTCF and coordinates chromatin landscape and gene expression in hematopoietic cells[J].Nucleic Acids Res.,53(4):gkaf093.
Quang D,Xie X H,2016.Dan Q:a hybrid convolutional and recurrent deep neural network for quantifying the function of DNA sequences[J].Nucleic Acids Res.,44(11):e107.
Rauluseviciute I,Riudavets-Puig R,Blanc-Mathieu R,et al.,2024.JASPAR 2024:20th anniversary of the open-access database of transcription factor binding profiles[J].Nucleic Acids Res.,52(D1):D174-D182.
Rinn J L,Chang H Y,2012.Genome regulation by long noncoding RNAs[J].Annu.Rev.Biochem.,81:145-166.
Simon M D,Wang C I,Kharchenko P V,et al.,2011.The genomic binding sites of a noncoding RNA[J].Proc.Natl.Acad.Sci.USA,108(51):20497-20502.
Statello L,Guo C J,Chen L L,et al.,2021.Gene regulation by long non-coding RNAs and its biological functions[J].Nat.Rev.Mol.Cell Biol.,22(2):96-118.
Tsai M C,Manor O,Wan Y,et al.,2010.Long noncoding RNAas modular scaffold of histone modification complexes[J].Science,329(5992):689-693.
Valeri J A,Collins K M,Ramesh P,et al.,2020.Sequence-tofunction deep learning frameworks for engineered riboregulators[J].Nat.Commun.,11(1):5058.
Vaswani A,Shazeer N,Parmar N,et al.,2017.Attention is all you need[C]//Advances in Neural Information Processing Systems,30:5998-6008.
Wang F,Chainani P,White T,et al.,2018.Deep learning identifies genome-wide DNA binding sites of long noncoding RNAs[J].RNA Biol.,15(12):1468-1476.
Wang X H,Zhang L H,Liang Q Y,et al.,2022a.DUSP5P1promotes gastric cancer metastasis and platinum drug resistance[J].Oncogenesis,11(1):66.
Wang Z,Chen J F,Sun F J,et al.,2022b.Lnc RNA CRLM1 inhibits apoptosis and promotes metastasis through transcriptional regulation cooperated with hnRNPK in colorectal cancer[J].Cell Biosci.,12(1):120.
Warwick T,Brandes R P,Leisegang M S,2023.Computational methods to study DNA:DNA:RNA triplex formation by lnc RNAs[J].Non Coding RNA,9(1):10.
Wojciuk M,Swiderska-Chadaj Z,Siwek K,et al.,2024.Improving classification accuracy of fine-tuned CNN models:impact of hyperparameter optimization[J].Heliyon,10(5):e26586.
Zhai P S,Tong T,Wang X N,et al.,2024.Nuclear mi R-451a activates KDM7A and leads to cetuximab resistance in head and neck squamous cell carcinoma[J].Cell.Mol.Life Sci.,81(1):282.
Zhang C,Wang D P,Hao Y J,et al.,2022.Lnc RNA CCTT-mediated RNA-DNA and RNA-protein interactions facilitate the recruitment of CENP-C to centromeric DNA during kinetochore assembly[J].Mol.Cell,82(21):4018-4032.e9.
Zhang J,Ataei L,Mittal K,et al.,2025.LINE1 and PRC2 control nucleolar organization and repression of the 8C state in human ESCs[J].Dev.Cell,60(2):186-203.e13.
Zhang Y,Liu T,Meyer C A,et al.,2008.Model-based analysis of Ch IP-seq (MACS)[J].Genome Biol.,9(9):R137.
Zhang Y N,2023.Encoder-decoder models in sequence-tosequence learning:a survey of RNN and LSTM approaches[J].Appl.Comput.Eng.,22(1):218-226.
基本信息:
DOI：10.13417/j.gab.045.000887
中图分类号:Q811.4
引用信息:
[1]田雨欣,张立木,杨亦轩,等.面向长短程依赖发现的全基因组lncRNA结合位点预测深度学习框架[J].基因组学与应用生物学,2026,45(04):887-901.DOI:10.13417/j.gab.045.000887.
基金信息:
广西大学引进人才科研启动基金(ZX01080031324004); 广西科技基地与人才专项——亚热带国家重点实验室项目(AD 25069107); 广西壮族自治区大学生创新训练计划项目(S2025105 93187)共同资助
2026-07-16
2026-07-16
2026-07-16