B-cell epitope prediction.

BackgroundBrucellosis is a significant zoonotic disease that impacts people globally, and its diagnosis has long posed challenges. This study aimed to explore the application value of multi-epitope fusion protein in the diagnosis of human brucellosis.MethodsEight important Brucella outer membrane proteins (OMPs) were selected: BP26, omp10, omp16, omp25, omp2a, omp2b, and omp31. Bioinformatics techniques were used to predict the immune epitopes of these proteins, and a multi-epitope fusion protein was designed. This fusion protein was used as the antigen for indirect enzyme-linked immunosorbent assay (iELISA) testing on 100 positive and 96 negative serum samples. The performance of the fusion protein in diagnosing brucellosis was evaluated using receiver operating characteristic (ROC) curves.ResultsA total of 31 epitopes were predicted from the eight proteins, and a multi-epitope fusion protein was successfully obtained. For the detection of human serum samples, the area under the ROC curve (AUC) of the fusion protein was 0.9594, with a positive diagnostic accuracy of 91.26% and a negative diagnostic accuracy of 93.55%. The area under the ROC curve (AUC) for lipopolysaccharides (LPS) was 0.9999, with a positive diagnostic accuracy of 100% and a negative diagnostic accuracy of 98.97%.ConclusionsThe fusion protein constructed using bioinformatics techniques, as the diagnostic antigen, showed significantly reduced cross-reactivity and enhanced specificity, improving diagnostic accuracy. This not only saves time but also avoids the preparation of LPS antigens, making the diagnostic process safer and more convenient.

研究背景：布鲁氏菌病（Brucellosis）是一种影响全球人类的重要人畜共患病，其诊断长期以来均存在挑战。本研究旨在探讨多表位融合蛋白在人布鲁氏菌病诊断中的应用价值。研究方法：研究选取了8种重要的布鲁氏菌外膜蛋白（Brucella outer membrane proteins, OMPs）：BP26、omp10、omp16、omp25、omp2a、omp2b及omp31。采用生物信息学技术预测上述蛋白的免疫表位，进而设计得到多表位融合蛋白。将该融合蛋白作为抗原，用于间接酶联免疫吸附试验（indirect enzyme-linked immunosorbent assay, iELISA）检测100份阳性血清样本与96份阴性血清样本。通过受试者工作特征（receiver operating characteristic, ROC）曲线评估该融合蛋白用于布鲁氏菌病诊断的性能。研究结果：从8种蛋白中共预测得到31个表位，成功获得多表位融合蛋白。在人血清样本检测中，该融合蛋白的ROC曲线下面积（area under the ROC curve, AUC）为0.9594，诊断阳性准确率为91.26%，阴性准确率为93.55%。脂多糖（lipopolysaccharides, LPS）的ROC曲线下面积为0.9999，诊断阳性准确率为100%，阴性准确率为98.97%。研究结论：以生物信息学技术构建的融合蛋白作为诊断抗原，其交叉反应性显著降低、特异性得以提升，诊断准确性得到改善。该方案不仅节省了检测时间，还无需制备脂多糖抗原，使诊断过程更为安全便捷。