Liu Chao, Liu Lanchun, Gao Jialiang, Wang Jie, Liu Yongmei
Guang'anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China.
Graduate School, Beijing University of Chinese Medicine, Beijing, China.
Front Genet. 2021 Nov 18;12:780431. doi: 10.3389/fgene.2021.780431. eCollection 2021.
Coronary heart disease (CHD) is a global health concern with high morbidity and mortality rates. This study aimed to identify the possible long non-coding RNA (lncRNA) biomarkers of CHD. The lncRNA- and mRNA-related data of patients with CHD were downloaded from the Gene Expression Omnibus database (GSE113079). The limma package was used to identify differentially expressed lncRNAs and mRNAs (DElncRNAs and DEmRNAs, respectively). Then, miRcode, TargetScan, miRDB, and miRTarBase databases were used to form the competing endogenous RNA (ceRNA) network. Furthermore, SPSS Modeler 18.0 was used to construct a logistic stepwise regression prediction model for CHD diagnosis based on DElncRNAs. Of the microarray data, 70% was used as a training set and 30% as a test set. Moreover, a validation cohort including 30 patients with CHD and 30 healthy controls was used to verify the hub lncRNA expression through real-time reverse transcription-quantitative PCR (RT-qPCR). A total of 185 DElncRNAs (114 upregulated and 71 downregulated) and 382 DEmRNAs (162 upregulated and 220 downregulated) between CHD and healthy controls were identified from the microarray data. Furthermore, through bioinformatics prediction, a 38 lncRNA-21miRNA-40 mRNA ceRNA network was constructed. Next, by constructing a logistic stepwise regression prediction model for 38 DElncRNAs, we screened two hub lncRNAs AC010082.1 and AC011443.1 ( < 0.05). The sensitivity, specificity, and area under the curve were 98.41%, 100%, and 0.995, respectively, for the training set and 93.33%, 91.67%, and 0.983, respectively, for the test set. We further verified the significant upregulation of AC010082.1 ( < 0.01) and AC011443.1 ( < 0.05) in patients with CHD using RT-qPCR in the validation cohort. Our results suggest that lncRNA AC010082.1 and AC011443.1 are potential biomarkers of CHD. Their pathological mechanism in CHD requires further validation.
冠心病（CHD）是一个全球性的健康问题，发病率和死亡率都很高。本研究旨在确定冠心病可能的长链非编码RNA（lncRNA）生物标志物。从基因表达综合数据库（GSE113079）下载冠心病患者的lncRNA和mRNA相关数据。使用limma软件包来识别差异表达的lncRNAs和mRNAs（分别为DElncRNAs和DEmRNAs）。然后，使用miRcode、TargetScan、miRDB和miRTarBase数据库构建竞争性内源RNA（ceRNA）网络。此外，使用SPSS Modeler 18.0基于DElncRNAs构建用于冠心病诊断的逻辑逐步回归预测模型。在微阵列数据中，70%用作训练集，30%用作测试集。此外，一个包含30例冠心病患者和30例健康对照的验证队列用于通过实时逆转录定量PCR（RT-qPCR）验证关键lncRNA的表达。从微阵列数据中鉴定出冠心病患者与健康对照之间共有185个DElncRNAs（114个上调和71个下调）和382个DEmRNAs（162个上调和220个下调）。此外，通过生物信息学预测，构建了一个由38个lncRNA-21个miRNA-40个mRNA组成的ceRNA网络。接下来，通过为38个DElncRNAs构建逻辑逐步回归预测模型，筛选出两个关键lncRNAs AC010082.1和AC011443.1（<0.05）。训练集的灵敏度、特异性和曲线下面积分别为98.41%、100%和0.995，测试集分别为93.33%、91.67%和0.983。我们在验证队列中使用RT-qPCR进一步验证了冠心病患者中AC010082.1（<0.01）和AC011443.1（<0.05）的显著上调。我们的结果表明，lncRNA AC010082.1和AC011443.1是冠心病的潜在生物标志物。它们在冠心病中的病理机制需要进一步验证。
