CAJ | 학술논문

目的筛选淋巴瘤患者血清蛋白标志物建立预测模型并评价其诊断价值.方法收集81例淋巴瘤病患者和86例正常人血清标本,利用表面增强激光解吸电离飞行时间质谱(SELDI-TOF-MS)及CM10(弱阳离子表面)蛋白芯片检测并筛选淋巴瘤血清标志蛋白,分别结合Biomarker Patterns Software(BPS)软件和人工神经网络技术建立智能预测模型,评价模型诊断价值.每条芯片随机选择一个点检测标准蛋白,评价实验重复性.结果淋巴瘤与对照间共检测到65个蛋白质峰,其中39个差异有意义(P＜0.05).BPS筛选质荷比4359,6673,8978,15 190蛋白质建立决策树模型预测淋巴瘤的灵敏度为92.7%,特异度为89.1%,同时建立人工神经网络模型预测淋巴瘤灵敏度和特异度分别为87.8%和87.0%.标准蛋白表达丰度68.3±8.2,变异系数(CV)12.0%.结论利用标准蛋白作为控制物在一定程度上保证了SELDI实验重复性,筛选的标志蛋白建立决策树模型在分子水平早期预测淋巴瘤中具有潜在应用价值.
목적 사선림파류환자혈청단백표지물건립예측모형병평개기진단개치.방법 수집81례림파류병환자화86례정상인혈청표본,이용표면증강격광해흡전리비행시간질보(SELDI-TOF-MS)급CM10(약양리자표면)단백심편검측병사선림파류혈청표지단백,분별결합Biomarker Patterns Software(BPS)연건화인공신경망락기술건립지능예측모형,평개모형진단개치.매조심편수궤선택일개점검측표준단백,평개실험중복성.결과 림파류여대조간공검측도65개단백질봉,기중39개차이유의의(P＜0.05).BPS사선질하비4359,6673,8978,15 190단백질건립결책수모형예측림파류적령민도위92.7%,특이도위89.1%,동시건립인공신경망락모형예측림파류령민도화특이도분별위87.8%화87.0%.표준단백표체봉도68.3±8.2,변이계수(CV)12.0%.결론 이용표준단백작위공제물재일정정도상보증료SELDI실험중복성,사선적표지단백건립결책수모형재분자수평조기예측림파류중구유잠재응용개치.
Objective To evaluate the diagnostic significance of serum protein markers in patient with lymphoma.Methods The sera of 81 patients with lymphoma and 86 healthy controls were analyzed by SELDI-TOF-MS and CM10 protein chips.The profiling was used for identifying the statistically significant peaks as well as for developing predictive model based on biomarker patterns software (BIAS) and artificial neural network (ANN).Results A total of 65 protein peaks were obtained and 39 differential peaks were selected (P＜0.05).A diagnostic model constructed by BPS with proteins of m/z 4 359,6 673,8 978 and 15 190 could successfully diagnose lymphoma with sensitivity of 92.7% and specificity of 89.1%.The ANN model based on the four markers distinguishing lymphoma between healthy controls had a sensitivity of 87.8% and a specificity of 87.0%.Conclusions Using standard protein as a control material can improve the experimental reproducibility.This strategy is a potential method for rapid diagnosis of lymphoma.