摘要
目的 筛选乳腺癌中差异表达的转运RNA衍生小RNA(tsRNA),构建与乳腺癌患者预后相关的预测模型.方法 收集癌症基因组图谱(TCGA)数据库中的乳腺癌患者的tsRNA表达谱以及临床相关数据,利用最小绝对值收敛和选择算子算法(LASSO)筛选出关键tsRNA,构建风险评分模型和列线图预测模型.结果 从乳腺癌差异表达的 106 条tsRNA中识别了 11条非零系数的tsRNA,据此构建了评估乳腺癌预后的风险评分模型和列线图预测模型,经测试集和校准曲线证实两模型均具有良好的预测能力.进一步预测上述11 条tsRNA的靶基因,发现其富集了8 条KEGG通路,包括多条与肿瘤的进展相关的通路;经相关性分析发现,NR2C2 和ZNFX1 与tRF-21-YBOZZ7ND的表达水平呈负相关.结论 成功构建了风险评分模型和列线图预测模型,有望为乳腺癌的预后评估提供新的方法.
Abstract
Objective To construct a predictive model related to the prognosis of breast cancer patients by screening the differential ex-pression of transferred RNA-derived small RNA(tsRNA)in breast cancer.Methods The tsRNA profiles in breast cancer and related clinical data were collected from The Cancer Genome Atlas(TCGA).Least Absolute Shrinkage and Selection Operator(LASSO)were used to screen the key tsRNA by which a risk score model and a nomogram prediction model were built.Results A total of 11 tsRNA with nonzero coefficients were identified from 106 differentially expressed tsRNA in breast cancer,by which a risk score model and a nomogram prediction model for predicting the prognosis of breast cancer were constructed.Two models were demonstrated to be capable of great prediction based on the test set and calibration curves.The further prediction for the target genes of the above 11 tsRNA re-vealed that they enriched in 8 KEGG pathways,including multiple pathways involved in tumor progression.The expression correlation analysis found that both NR2C2 and ZNFX1 negatively correlated with the expression levels of tRF-21-YBOZZZ7ND.Conclusion The risk score model and nomogram prediction model were successfully constructed,which are expected to provide new approaches and in-novative the concepts for evaluating the prognosis of breast cancer.
NETL
NSTL
维普
万方数据