摘要
目的运用加权基因共表达网络分析(WGCNA)和最小绝对收缩和选择算子(LASSO)方法筛选出与骨肉瘤转移和预后相关的基因,为骨肉瘤患者是否发生转移及其预后提供参考依据。方法利用UCSC Xena网站下载癌症基因组图谱-产生有效治疗方法研究项目(TCGA-TARGET)骨肉瘤患者基因表达谱和临床相关资料(检索时间:建库至2021年5月31日),数据库内含有84例骨肉瘤患者,其中男性37例,女性47例;平均年龄14.99岁(标准差7.80岁)。采用WGCNA和LASSO对84例骨肉瘤患者基因表达谱进行分析并筛选与预后和骨肉瘤转移相关基因;利用ONCOMINE数据库(检索时间:建库至2021年5月31日)进行外部验证。采用基因本体论(GO)和京都基因和基因组百科全书(KEGG)分析筛选模块涉及功能和通路,基因组富集分析(GSEA)分析核心基因所涉及通路。结果通过WGCNA筛选一共获得67个模块,其中深蓝模块为核心模块,对深蓝模块中所包含的214基因进行LASSO分析,结合蛋白质-蛋白质相互作用(PPI)分析和复杂分子关联分析(MOCDE)得分,筛选出线粒体核糖体蛋白48基因(MRPL48)、核糖体RNA加工8基因(RRP8)、核糖体RNA加工9基因(RRP9)作为核心靶基因。利用受试者工作特性(ROC)曲线分析核心基因预测骨肉瘤转移概率,结果提示RRP8,曲线下面积(AUC)=0.504,P=0.955;RRP9,AUC=0.509,P=0.913;MRPL48,AUC=0.663,P=0.025。可以明确MRPL48在骨肉瘤患者中的表达高低与其发生、预后明确相关。结论基于WGCNA并结合LASSO、PPI、MOCDE方法共同筛选获得的MRPL48与骨肉瘤发生和患者生存相关,可为骨肉瘤发生、转移和治疗的研究提供参考依据。
Objective To screen metastasis and prognosis of osteosarcoma(OS) related genes by weighted gene co-expression network analysis( WGCNA) and LASSO methods, and provide reference for metastasis and prognosis of OS patients.Methods The UCSC Xena website was used to download The Cancer Genome Atlas-Therapeutically Applicable Research To Generate Effective Treatments(TCGA-TARGET) gene expression profiles and clinically relevant data of OS patients(search time: database establishment to May 31, 2021), which enrolled 84 OS patients, including 37 males and 47 females, with mean age of 14.99 years old(standard deviation 7.80 years old). The gene ontology(GO) and Kyoto Encyclopedia of Genes and Genomes(KEGG) were used to analyze functions and pathways of screening modules, and gene set enrichment analysis(GSEA)was used to analyze pathways involved in core genes. Results The 67 modules were obtained by WGCNA screening, of which midnight blue module was core module. The least absolute contraction and selection operator(LASSO) analysis were performed for 214 gene of midnight blue module, combined with protein-protein interaction(PPI) analysis and complex molecular associa tion analysis(MOCDE) score, and mitochondrial ribosomal protein L48 gene( MRPL48), ribosomal RNA processing 8 gene(RRP8), ribosomal RNA processing 9 gene(RRP9) were screened as core target genes. The receiver operating characteristic(ROC) curves were used to analyze core genes for predicting OS metastasis probability, and analysis results showed that RRP8area under curve(AUC) = 0.504, P = 0.955;RRP9 AUC = 0.509, P = 0.913;MRPL48 AUC = 0.663, P = 0.025, which confirmed that the expression of MRPL48 in OS patient was clearly related to its occurrence and prognosis. Conclusion It is demonstrated that MRPL48 screened by WGCNA combined with LASSO, PPI and MOCDE methods is related to the occurrence of osteosarcoma and survival of OS patient, which can provide reference for the study of occurrence, metastasis and treatment of OS.
作者
张毅
姚士军
刘朝旭
ZHANG Yi;YAO Shi-jun;LIU Chao-xu(Taikang Tongji(Wuhan)Hospital,Wuhan 430000,Hubei,China;Tongji Hospital,Tongji Medical College of HUSTf Wuhan 430050,Hubei,China)
出处
《生物医学工程与临床》
CAS
2022年第3期351-358,共8页
Biomedical Engineering and Clinical Medicine
基金
国家自然科学基金面上项目(51877097)。
