应用蛋白质芯片与人工神经网络建立胃癌诊断模型

Classification and diagnostic prediction of gastric cancer using protein profiling of the serum and bioinformatics

目的应用血清蛋白质芯片质谱技术结合生物信息学方法筛选胃癌相关标记,建立诊断胃癌的蛋白质标记模型。方法应用表面增强激光解吸电离飞行时间质谱技术检测了74例胃癌、70例健康对照样本的血清蛋白质质谱,结合人工神经网络建立胃癌诊断模型。结果从胃癌组与健康对照组中筛选出了4个蛋白质荷比峰,建立胃癌诊断模型,该诊断模型的特异性为100%(95%的置信区间为94.6%~100.0%),敏感性为92.0%(88.7%~95.2%),准确率为95.8%(86.4%~97.5%)。结论成功建立了胃癌诊断模型,该模型在胃癌的诊断中较传统方法具有更高的敏感性和特异性,可进一步研究与应用。

Objective To develop a bioinformatics tool and to identify aproteomic pattern in the serum to distinguish gastric cancer from healthy individuals.Methods 144 serum samples including 74 gastric cancer patients and 70 healthy individuals were determined by surface-enhanced laser desorption/ionization(SELDI) mass spectrometry.Four peaks were selected to distinguish gastric cancer from healthy individuals.Results The diagnostic pattern separated gastric cancer from healthy samples with a specificity of 100% and...