基于支持向量机对肾母细胞瘤患者血清蛋白质标记物的检测分析  

Detecting biomarkers from serum in nephroblastoma patients with support vector machine

作　　者：王家祥[1] 张蛟[1] 刘秋亮[1] 王利[1] 范应中[1] 余捷凯[2] 郑树[2] 

机构地区：[1]郑州大学第一附属医院外科医学部,450052 [2]浙江大学第二医院肿瘤研究所

出　　处：《中华医学杂志》2006年第42期2982-2985,共4页National Medical Journal of China

基　　金：国家自然科学基金资助项目(30571930)

摘　　要：目的检测肾母细胞瘤患儿血清蛋白质,筛选特异的蛋白质标记物,构建用于肾母细胞瘤早期诊断的血清蛋白质指纹图谱模型。方法应用表面增强激光解析电离飞行时间质谱(SELDI-TOF-MS)技术检测75例血清标本(肾母细胞瘤30例,其他小儿腹腔实体肿瘤25例,正常小儿20例)的蛋白质质谱,并结合生物信息学方法(支持向量机)分析数据。结果筛选出4个质荷比(m/z)位于6984.5、6455.5、6914.0、3256.7的蛋白质标记物,构建肾母细胞瘤早期诊断模型。经留一法交叉验证,区分肾母细胞瘤和正常小儿的血清蛋白质指纹图谱模型特异性为100%,敏感性为100%;区分肾母细胞瘤与其他腹腔实体肿瘤的血清蛋白质指纹图谱模型特异性为100%,敏感性为93.3%。结论表面增强激光解析电离飞行时间质谱技术结合支持向量机建立肾母细胞瘤血清蛋白质指纹图谱模型是早期诊断肾母细胞瘤的一种特异性强、敏感性高的新方法,可用于肾母细胞瘤早期诊断与肿瘤标志物筛选研究。Objective To find new biomarkers and to establish serum protein fingerprint models for early detection and diagnosis of nephroblastoma by SELDI-TOF-MS and bioinformatics tools. Methods Seventy five serum samples from 35 nephroblastoma patients,30 children＇s abdominal solid tumor patients, and 20 healthy children were bound to WCX2 protein chip and tested by surface enhanced laser desorption/ ionization time of flight-mass spectrometry （SELDI-TOF-MS）. The data of spectra were analyzed by support vector machine（SVM） , Results Four peaks with m/z of 6984. 5,6455.5,6914. 0,3256. 7 were selected as potential biomarkers, The detective model combined with 2 biomarkers could separate nephroblastoma from the healthy group with a sensitivity of 100%, and a specificity of 100%, The diagnostic model combined with 2 biomarkers could separate nephroblastoma from other child＇s abdominal solid tumors with a sensitivity of 93, 3% ,and a specificity of 100%. Conclusion High sensitivity and specificity achived by this method show great potential for early diagnosis of nephroblastoma, and screening for new tumor biomarkers.

关 键 词：肾母细胞瘤 诊断 支持向量机 蛋白质指纹图谱 SELDI-TOF-MS 

分 类 号：R737.11[医药卫生—肿瘤]