论文部分内容阅读
目的:筛选并建立新疆维吾尔族食管癌血清蛋白指纹图谱诊断模型,为食管癌的诊断与临床筛查提供新的途径。方法:采用弱阳离子交换蛋白质芯片(CM10蛋白芯片)及表面增强激光解析离子化飞行时间质谱(SELDI-TOF-MS)技术对23例新疆维吾尔族食管癌和33例新疆维吾尔族正常对照者血清指纹图谱进行检测,所得结果用ZUCI-蛋白芯片数据分析系统(ZUCI-Protein Chip Data Analyze System)软件包进行分析,通过支持向量机运算建立区分新疆维吾尔族食管癌蛋白指纹图谱诊断模型,并用留一法交叉验证作用评估模型,判别效果。结果:通过软件包运算,用2个质荷比峰(3269.4621、6056.8714m/z)建立了新疆维吾尔族食管癌蛋白指纹图谱诊断模型,准确度为92.9%,灵敏度为91.3%,特异度为93.9%,阳性预测值为91.3%。结论:SELDI-TOF-MS技术结合支持向量机建立新疆维吾尔族食管癌血清蛋白质指纹图谱模型为早期筛查及诊断新疆维吾尔族食管癌提供了一种特异性强、灵敏度高的新方法,值得进一步的研究和应用。
Objective: To screen and establish a diagnostic model of serum protein fingerprinting of esophageal cancer in Uighur Xinjiang, and to provide a new approach for the diagnosis and clinical screening of esophageal cancer. Methods: Serum fingerprints of 23 Uygur esophageal cancer patients and 33 Uygur normal controls were detected by using the weak cation-exchange protein chip (CM10) and SELDI-TOF-MS (surface enhanced laser desorption ionization time of flight mass spectrometry) The results were analyzed by ZUCI-Protein Chip Data Analyze System software package. The diagnosis model of distinguishing the Uygur esophageal cancer protein fingerprinting was established by support vector machine (SVM) Cross-validation role model to determine the effect. Results: The diagnostic fingerprint of Uygur esophageal cancer protein fingerprinting was established by software package calculation with two peaks of mass-to-charge ratio (3269.4621, 6056.8714m / z) with the accuracy of 92.9%, the sensitivity of 91.3% and the specificity of 93.9 %, Positive predictive value of 91.3%. Conclusion: The SELDI-TOF-MS combined with support vector machine to establish Xinjiang Uygur esophageal cancer serum protein fingerprinting model provides a new method with high specificity and sensitivity for early screening and diagnosis of Uygur esophageal cancer, which is worth further Research and application.