论文部分内容阅读
目的探索基于Au蛋白芯片的尿蛋白标志物模型早期快速诊断糖尿病肾病(DN)的应用价值。方法应用表面增强激光解析电离飞行时间质谱(SELDI-TOF-MS)技术及Au芯片检测86例DN患者、40例糖尿病(DM)患者及55名健康人(对照组)尿蛋白质谱。获得的谱图用Biomaker Wizard3.1软件筛选差异蛋白,通过BPS软件建立决策树辨别分析模型,评价其临床诊断价值。部分筛选的差异蛋白通过比对标准蛋白质谱数据进行鉴定。结果DN患者与对照组尿差异蛋白质峰有37个(P<0.05),BPS筛选66 916质荷比(m/z)蛋白建立的模型诊断DN敏感度100%,特异度97.78%。对DM和DN患者尿蛋白质谱图分析后得到24个差异蛋白质峰(P<0.05),BPS筛选4 008、11 619、66 916 m/z蛋白建立的模型区分两者的灵敏度和特异度均为100%。比对标准蛋白质谱数据,DN患者尿中11 619、23529、66 916、79 378 m/z蛋白,可能分别为β_2-微球蛋白、α_1-微球蛋白、白蛋白、转铁蛋白。结论利用SELDI及Au芯片检测尿标志蛋白在鉴别蛋白尿来源、DN快速诊断及肾损害评估中具有重要应用价值。
Objective To explore the value of early detection of diabetic nephropathy (DN) by urinary protein marker model based on Au protein chip. Methods Urine protein profiles of 86 DN patients, 40 DM patients and 55 healthy controls (control group) were detected by SELDI-TOF-MS and Au chip. The obtained spectra were screened by Biomaker Wizard 3.1 software for differential proteins, and the BPS software was used to establish the discriminant analysis model of the decision tree to evaluate its clinical diagnostic value. Partially screened differential proteins were identified by aligning standard protein profiling data. Results There were 37 urine protein peaks (P <0.05) in patients with DN and those in control group. The sensitivity and specificity of the model established by BPS screening of 66 916 m / z protein were 100% and 97.78%, respectively. 24 differential protein peaks (P <0.05) were obtained from urine protein profiles of patients with DM and DN, and the sensitivity and specificity of the models established by BPS screening of 4 008, 11 619 and 66 916 m / z proteins were 100%. Compared with the standard protein data, 11 619,23529,66 916,79 378 m / z in urine of DN patients may be β_2-microglobulin, α_1-microglobulin, albumin and transferrin respectively. Conclusion The detection of urinary marker protein using SELDI and Au chips is of great value in distinguishing the origin of proteinuria, rapid diagnosis of DN and assessment of renal damage.