摘要
应用SELDI技术和生物信息学方法从血清中筛选能反映乳腺癌术前分期的蛋白质峰并构建检测模型。采用CM10芯片,对34例Ⅰ～Ⅱ期乳腺癌患者和31例Ⅲ～Ⅳ期乳腺癌的血清进行了检测,发现11个蛋白质峰在两组患者之间表达量有显著性差异(P<0.05),M/Z为M2042.87、M2459.83、M3881.37、M4804.47、M6683.24和M6706.06的6个蛋白质峰被选为分类变量构成决策树分类模型,该模型的交叉验证(测试组)总准确率为80.0%,Ⅰ～Ⅱ期乳腺癌检出率为82.4%,Ⅲ～Ⅳ期检出率为77.4%。SELDI技术在乳腺癌患者术前分级的判断方面具有一定的应用价值。
To screen serum proteins peaks which are fit for estimation of preoperative grading and found predictive models of breast cancer by SELDI-TOF-MS and bioinformatics tools. Serum samples from breast cancer patients including grade Ⅰ/Ⅱ （ 34 cases） and grade Ⅲ/Ⅳ （31 cases） were analyzed. 11 discrepant proteins peaks were screened between two groups （ P 〈 0.05 ）. The best predictive model composed with the M/Z of M2042.87, M2459.83, M3881.37, M4804.47, M6683.24 and M6706.06. The accuracy of cross verification of the model was 80.0% , the accuracy of predicting grade Ⅰ～Ⅱ and grade Ⅲ～Ⅳ breast cancer was 82.4% and 77.4%, respectively.
出处
《中国生物工程杂志》
CAS
CSCD
北大核心
2009年第9期56-60,共5页
China Biotechnology
