Wu Qiaofeng, Zou Meng, Yang Mingxiao, Zhou Siyuan, Yan Xianzhong, Sun Bo, Wang Yong, Chang Shyang, Tang Yong, Liang Fanrong, Yu Shuguang
Acupuncture and Tuina College, Chengdu University of Traditional Chinese Medicine, Chengdu, Sichuan, 610075, China.
National Center for Mathematics and Interdisciplinary Sciences, Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, 100080, China.
Sci Rep. 2016 Jan 8;6:18852. doi: 10.1038/srep18852.
Metabonomics methods have gradually become important auxiliary tools for screening disease biomarkers. However, recognition of metabolites or potential biomarkers closely related to either particular clinical symptoms or prognosis has been difficult. The current study aims to identify potential biomarkers of functional dyspepsia (FD) by a new strategy that combined hydrogen nuclear magnetic resonance ((1)H NMR)-based metabonomics techniques and an integrative multi-objective optimization (LPIMO) method. First, clinical symptoms of FD were evaluated using the Nepean Dyspepsia Index (NDI), and plasma metabolic profiles were measured by (1)H NMR. Correlations between the key metabolites and the NDI scores were calculated. Then, LPIMO was developed to identify a multi-biomarker panel by maximizing diagnostic ability and correlation with the NDI score. Finally, a KEGG database search elicited the metabolic pathways in which the potential biomarkers are involved. The results showed that glutamine, alanine, proline, HDL, β-glucose, α-glucose and LDL/VLDL levels were significantly altered in FD patients. Among them, phosphatidycholine (PtdCho) and leucine/isoleucine (Leu/Ile) were positively and negatively correlated with the NDI Symptom Index (NDSI) respectively. Our procedure not only significantly improved the credibility of the biomarkers, but also demonstrated the potential of further explorations and applications to diagnosis and treatment of complex disease.
代谢组学方法已逐渐成为筛选疾病生物标志物的重要辅助工具。然而，识别与特定临床症状或预后密切相关的代谢物或潜在生物标志物一直很困难。本研究旨在通过一种新策略来识别功能性消化不良（FD）的潜在生物标志物，该策略结合了基于氢核磁共振（（1）H NMR）的代谢组学技术和综合多目标优化（LPIMO）方法。首先，使用Nepean消化不良指数（NDI）评估FD的临床症状，并通过（1）H NMR测量血浆代谢谱。计算关键代谢物与NDI评分之间的相关性。然后，开发LPIMO以通过最大化诊断能力和与NDI评分的相关性来识别多生物标志物组合。最后，通过KEGG数据库搜索得出潜在生物标志物所涉及的代谢途径。结果表明，FD患者的谷氨酰胺、丙氨酸、脯氨酸、高密度脂蛋白（HDL）、β-葡萄糖、α-葡萄糖和低密度脂蛋白/极低密度脂蛋白（LDL/VLDL）水平发生了显著变化。其中，磷脂酰胆碱（PtdCho）和亮氨酸/异亮氨酸（Leu/Ile）分别与NDI症状指数（NDSI）呈正相关和负相关。我们的方法不仅显著提高了生物标志物的可信度，还展示了在复杂疾病诊断和治疗中进一步探索和应用的潜力。
