Self-Attention Subtype Recognition Neural Network for Classification of Kidney Renal Clear Cell Carcinoma Data

doi:10.3778/j.issn.1002-8331.2304-0280

Abstract

Abstract: To analyse transcriptomic data of kidney renal clear cell carcinoma（KIRC）, a modified classification model is constructed using the self-attention mechanism. A new self-attention subtype recognition neural network（SSRNN） is constructed, which includes an encoder and classifier and takes the self-attention mechanism as the main improvement. After screening 358 survival-related protein-coding genes, three subtypes are identified by cluster analysis. Comparison of clinical information and survival analysis for the three cancer subtypes of C1, C2 and C3 reveales differences in survival outcomes among the groups. SSRNN achieves the best classification performance, achieving an area under the curve of 93.44%. Gene expression heatmaps reveale differences in gene expression among the three subtypes, suggesting that low gene expression indicates better survival prognosis. By analysing the differences among the three subtypes and drawing the volcano map, 266 differentially expressed genes are obtained. GO and KEGG enrichment analyses and node mapping are helpful to reveal cancer-related functions and pathways. Therefore, the SSRNN has high prediction accuracy and robustness, can effectively use omics data to predict the survival of KIRC patients and screen reasonable biomarkers, and has high methodological significance and application value.

Key words: self-attention, deep learning, omics data, kidney renal clear cell carcinoma, subtype recognition

摘要： 为分析肾透明细胞癌（KIRC）的转录组学数据，利用自注意力机制构建改良分类模型。构建了一种新的自注意力亚型识别神经网络（SSRNN），其包含了编码器和分类器部分，以自注意力机制为主要改良方式。在筛出358个与生存相关的蛋白编码基因后，利用聚类分析确定了三种亚型最为适宜。对于C1、C2和C3三组癌症亚型进行了临床信息比较和生存分析比较，揭示了各组在生存结局上的差异。SSRNN取得了最优异的分类性能，取得了93.44%的曲线下面积。基因表达热图提示三种亚型的基因表达存在差异，推测基因的低表达指示较好的生存预后。对三种亚型两两间进行差异分析并绘制火山图，共可获取266个差异基因。GO和KEGG富集分析及节点图的绘制则有利于揭示癌症相关的功能和通路。因此，SSRNN具有较高的预测精度和稳健性，可有效地利用组学数据进行KIRC的生存预测，并筛选合理的生物学标志物，具有较高的方法学意义和应用价值。

关键词: 自注意力, 深度学习, 组学数据, 肾透明细胞癌, 亚型识别

LI Yang, CHEN Xicheng, WU Yazhou. Self-Attention Subtype Recognition Neural Network for Classification of Kidney Renal Clear Cell Carcinoma Data[J]. Computer Engineering and Applications, 2023, 59(23): 104-113.

李阳, 陈锡程, 伍亚舟. KIRC组学数据分类的自注意亚型识别神经网络[J]. 计算机工程与应用, 2023, 59(23): 104-113.

References

[1] BRAY F，FERLAY J，SOERJOMATARAM I，et al.Global cancer statistics 2018：GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J].CA：A Cancer Journal for Clinicians，2018，68（6）：394-424.
[2] SIEGEL R L，MILLER K D，JEMAL A.Cancer statistics，2020[J].CA：A Cancer Journal for Clinicians，2020，70（1）：7-30.
[3] ZHAO W，LUO J，JIAO S.Comprehensive characterization of cancer subtype associated long non-coding RNAs and their clinical implications[J].Scientific Reports，2014，4（1）：6591.
[4] BAHADORAM S，DAVOODI M，HASSANZADEH S，et al.Renal cell carcinoma：an overview of the epidemiology，diagnosis，and treatment[J].Giornale Italiano Di Nefrologia：Organo Ufficiale Della Societa Italiana Di Nefrologia，2022，39（3）：1-17.
[5] LOBO J，OHASHI R，AMIN M B，et al.WHO 2022 landscape of papillary and chromophobe renal cell carcinoma[J].Histopathology，2022，81（4）：426-438.
[6] HUANG S H，LO Y S，LUO Y C，et al.CoMI：consensus mutual information for tissue-specific gene signatures[J].BMC Bioinformatics，2022，22（S10）：624.
[7] LINEHAN W M，RICKETTS C J.The cancer genome Atlas of renal cell carcinoma：findings and clinical implications[J].Nature Reviews Urology，2019，16（9）：539-552.
[8] SUN Y，JIANG Y，LI Y，et al.Identification of cancer omics commonality and difference via community fusion[J].Statistics in Medicine，2019，38（7）：1200-1212.
[9] YIN L，DUAN J J，BIAN X W，et al.Triple-negative breast cancer molecular subtyping and treatment progress[J].Breast Cancer Research，2020，22（1）：61.
[10] WANG X，YU G，WANG J，et al.Lung cancer subtype diagnosis using weakly-paired multi-omics data[J].Bioinformatics，2022，38（22）：5092-5099.
[11] RUMGAY H，FERLAY J，DE MARTEL C，et al.Global，regional?and national burden of primary liver cancer by subtype[J].European Journal of Cancer，2022，161：108-118.
[12] MEHBOOB F，RAUF A，JIANG R，et al.Towards robust diagnosis of COVID-19 using vision self-attention transformer[J].Scientific Reports，2022，12（1）：8922.
[13] ALFASLY S，CHUI C K，JIANG Q，et al.An effective video transformer with synchronized spatiotemporal and spatial self-attention for action recognition[J].IEEE Transactions on Neural Networks and Learning Systems，2022：1-14.
[14] LIN S，WANG Y，ZHANG L，et al.MDF-SA-DDI：predicting drug-drug interaction events based on multi-source drug fusion，multi-source feature fusion and transformer self-attention mechanism[J].Briefings in Bioinformatics，2022，23（1）：bbab421.
[15] AL-HEJRI A M，AL-TAM R M，FAZEA M，et al.ETECADx：ensemble self-attention transformer encoder for breast cancer diagnosis using full-field digital X-ray breast images[J].Diagnostics，2022，13（1）：1-30.
[16] YUAN F，LU L，ZHANG Y，et al.Data mining of the cancer-related lncRNAs GO terms and KEGG pathways by using mRMR method[J].Mathematical Biosciences，2018，304：1-8.
[17] CHONG X，PENG R，SUN Y，et al.Identification of key genes in gastric cancer by bioinformatics analysis[J].BioMed Research International，2020，2020：1-12.
[18] LI L，LU Y，LIU Y，et al.Network pharmacology analysis of Huangqi Jianzhong Tang targets in gastric cancer[J].Frontiers in Pharmacology，2022，13：882147.
[19] WU F，CAI J，WEN C，et al.Co-sparse non-negative matrix factorization[J].Frontiers in Neuroscience，2021，15：804554.
[20] GAO Y，CHURCH G.Improving molecular cancer class discovery through sparse non-negative matrix factorization[J].Bioinformatics，2005，21：3970-3975.
[21] HAMAMOTO R，TAKASAWA K，MACHINO H，et al.Application of non-negative matrix factorization in oncology：one approach for establishing precision medicine[J].Briefings in Bioinformatics，2022，23（4）：bbac246.
[22] AONISHI T，MARUYAMA R，ITO T，et al.Imaging data analysis using non-negative matrix factorization[J].Neuroscience Research，2022，179：51-56.
[23] PARKER C R，WENDEL G D.The effects of syphilis on endocrine function of the fetoplacental unit[J].American Journal of Obstetrics and Gynecology，1988，159（6）：1327-1331.
[24] MIAO S，SU H，CHEN S.Matrix-weighted consensus of second-order discrete-time multiagent systems[J].IEEE Transactions on Neural Networks and Learning Systems，2022：1-10.
[25] YU H，HE J，WANG X，et al.A comparison of functional features of Chinese and US mobile apps for pregnancy and postnatal care：a systematic app store search and content analysis[J].Frontiers in Public Health，2022，10：826896.
[26] TANG J，WANG Y，LUO Y，et al.Computational advances of tumor marker selection and sample classification in cancer proteomics[J].Computational and Structural Biotechnology Journal，2020，18：2012-2025.
[27] ALJABRI M，ALHAIDARI F，MOHAMMAD R M A，et al.An assessment of lexical，network，and content-based features for detecting malicious URLs using machine learning and deep learning models[J].Computational Intelligence and Neuroscience，2022，2022：3241216.
[28] JIN X，XU A，BIE R，et al.Machine learning techniques and chi-square feature selection for cancer classification using SAGE gene expression profiles[M]//LI J，YANG Q，TAN A H.Data mining for biomedical applications：3916.Berlin，Heidelberg：Springer，2006：106-115.
[29] SONG X，LIU X，LIU F，et al.Comparison of machine learning and logistic regression models in predicting acute kidney injury：a systematic review and meta-analysis[J].International Journal of Medical Informatics，2021，151：104484.
[30] AL FRYAN L H，SHOMO M I，ALAZZAM M B，et al.Processing decision tree data using Internet of things （IoT） and artificial intelligence technologies with special reference to medical application[J].BioMed Research International，2022，2022：8626234.
[31] HANKO M，GRENDáR M，SNOPKO P，et al.Random forest-based prediction of outcome and mortality in patients with traumatic brain injury undergoing primary decompressive craniectomy[J].World Neurosurgery，2021，148：450-458.
[32] MENG C，JU Y，SHI H.TMPpred：a support vector machine-based thermophilic protein identifier[J].Analytical Biochemistry，2022，645：114625.
[33] ZHOU Y，ZHOU B，PACHE L，et al.Metascape provides a biologist-oriented resource for the analysis of systems-level datasets[J].Nature Communications，2019，10（1）：1523.
[34] XU J，XIANG L，LIU Q，et al.Stacked sparse autoencoder（SSAE） for nuclei detection on breast cancer histopathology images[J].IEEE Transactions on Medical Imaging，2016，35（1）：119-130.
[35] SONG H，RUAN C，XU Y，et al.Survival stratification for colorectal cancer via multi-omics integration using an autoencoder-based model[J].Experimental Biology and Medicine，2022，247（11）：898-909.
[36] TONG L，MITCHEL J，CHATLIN K，et al.Deep learning based feature-level integration of multi-omics data for breast cancer patients survival analysis[J].BMC Medical Informatics and Decision Making，2020，20（1）：225.
[37] YANG H，CHEN R，LI D，et al.Subtype-GAN：a deep learning approach for integrative cancer subtyping of multi-omics data[J].Bioinformatics，2021，37（16）：2231-2237.
[38] SUN Q，CHENG L，MENG A，et al.SADLN：self-attention based deep learning network of integrating multi-omics data for cancer subtype recognition[J].Frontiers in Genetics，2023，13：1032768.
[39] LIU C，ZHANG L，NIU J，et al.Intelligent prognostics of machining tools based on adaptive variational mode decomposition and deep learning method with attention mechanism[J].Neurocomputing，2020，417：239-254.
[40] LI M，WANG Y，WANG Z，et al.A deep learning method based on an attention mechanism for wireless network traffic prediction[J].Ad Hoc Networks，2020，107：102258.