论文部分内容阅读
目的检测p16、RASSF1A和脆性组氨酸三联体基因(FHIT)甲基化水平及外周血DNA端粒长度,建立并探讨判别分析与决策树2种分类模型在肺癌辅助诊断中的意义。方法采用甲基化特异性PCR、实时荧光定量PCR法测定200名正常对照、200例肺癌患者外周血p16、RASSF1A和FHIT基因甲基化水平和DNA端粒长度,建立决策树、判别分析2种肺癌判别诊断模型。结果肺癌组和对照组中p16、RASSF1A和FHIT基因启动子甲基化水平(%)分别为0.59(0.16~4.50)与0.36(0.06~4.00)(P=0.008)、27.62(9.09~52.86)与17.17(3.86~50.87)(P=0.038)、3.33(1.86~6.40)与2.85(1.39~5.44)(P=0.002);端粒长度分别为(0.93±0.32)和(1.16±0.57)(P<0.001),4项生物标志在2组间差异均有统计学意义;判别分析、决策树对预测集的预测准确度分别为64%、83%;ROC曲线下面积分别为0.640、0.830,差异有统计学意义(P<0.05)。结论数据挖掘工具建立的决策树模型判别诊断肺癌的效果优于判别分析。
Objective To detect the methylation level of p16, RASSF1A, FHIT and telomere length of peripheral blood DNA, and to establish and discuss the significance of discriminant analysis and decision tree classification in the diagnosis of lung cancer. Methods Methylation-specific PCR and real-time fluorescence quantitative PCR were used to detect the methylation level of p16, RASSF1A and FHIT gene and DNA telomere length in 200 normal controls and 200 lung cancer patients. Diagnostic model of lung cancer. Results The promoter methylation levels of p16, RASSF1A and FHIT gene in lung cancer group and control group were 0.59 (0.16-4.50) and 0.36 (0.06-4.00) respectively (P = 0.008), 27.62 (9.09-52.86) The telomere length were 17.17 (3.86-50.87), 3.33 (1.86-6.40) and 2.85 (1.39-5.44) respectively (P = 0.002). The length of telomere was (0.93 ± 0.32) and (1.16 ± 0.57) 0.001). The differences of the four biomarkers between the two groups were statistically significant. The accuracy of the discriminant analysis and the prediction tree were 64% and 83% respectively. The areas under the ROC curve were 0.640 and 0.830, respectively Statistical significance (P <0.05). Conclusion The decision tree model established by data mining tools is superior to discriminant analysis in diagnosing lung cancer.