非靶向代谢组学和机器学习揭示了淋巴结结核的代谢失调

IF 5.6 1区化学 Q1 CHEMISTRY, ANALYTICAL

Talanta Pub Date : 2025-07-17 DOI:10.1016/j.talanta.2025.128583

Peijun Chen , Yuehui Yu , Ying Zhang , Gaoyi Yang

{"title":"非靶向代谢组学和机器学习揭示了淋巴结结核的代谢失调","authors":"Peijun Chen , Yuehui Yu , Ying Zhang , Gaoyi Yang","doi":"10.1016/j.talanta.2025.128583","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Lymph node tuberculosis (LNTB) is the most prevalent form of extrapulmonary tuberculosis; however, differentiating it from non-LNTB remains challenging due to overlapping clinical features and suboptimal diagnostic methods. Current diagnostic methods for LNTB lack both sensitivity and specificity. This study aimed to characterize the metabolic differences between LNTB and non-LNTB patients, elucidate the pathological mechanisms underlying LNTB, and identify diagnostic biomarkers using machine learning models.</div></div><div><h3>Methods</h3><div>Serum samples from 40 LNTB patients and 30 non-LNTB patients were analyzed using ultra-high-performance liquid chromatography-mass spectrometry. Differential metabolites were identified based on a variable importance in projection >1, false discovery rate-adjusted p-value <0.05. Pathway enrichment analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG). Machine learning, including support vector machines and random forest, were employed to screen for diagnostic biomarkers, which were validated by receiver operating characteristic curves.</div></div><div><h3>Results</h3><div>Among the 1294 detected metabolites, 89 exhibited significant differences between the two groups. By integrating KEGG enrichment with topological analysis, phenylalanine, tyrosine, and tryptophan biosynthesis possessed the highest impact, followed by phenylalanine metabolism, and aminoacyl-tRNA biosynthesis. Machine learning identified four biomarkers: Leu-Ala [area under the curve (AUC) = 0.8292], evodiamine (AUC = 0.7558), fenazaquin (AUC = 0.7175), and acetol (AUC = 0.7117). Leu-Ala demonstrated the highest diagnostic accuracy, with a sensitivity of 73.5 % and specificity 86.7 % at a cutoff value of 0.62.</div></div><div><h3>Conclusions</h3><div>Untargeted metabolomics revealed dysregulation in the biosynthesis of phenylalanine, tyrosine, and tryptophan, phenylalanine metabolism, as well as in aminoacyl-tRNA biosynthesis in LNTB. Additional, Leu-Ala was identified as a novel diagnostic biomarker. The integrating of metabolomics with machine learning presents a promising approach for LNTB detection, though larger validation studies are necessary.</div></div>","PeriodicalId":435,"journal":{"name":"Talanta","volume":"297 ","pages":"Article 128583"},"PeriodicalIF":5.6000,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Non-targeted metabolomics and machine learning reveal metabolic dysregulation in lymph node tuberculosis\",\"authors\":\"Peijun Chen , Yuehui Yu , Ying Zhang , Gaoyi Yang\",\"doi\":\"10.1016/j.talanta.2025.128583\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Background</h3><div>Lymph node tuberculosis (LNTB) is the most prevalent form of extrapulmonary tuberculosis; however, differentiating it from non-LNTB remains challenging due to overlapping clinical features and suboptimal diagnostic methods. Current diagnostic methods for LNTB lack both sensitivity and specificity. This study aimed to characterize the metabolic differences between LNTB and non-LNTB patients, elucidate the pathological mechanisms underlying LNTB, and identify diagnostic biomarkers using machine learning models.</div></div><div><h3>Methods</h3><div>Serum samples from 40 LNTB patients and 30 non-LNTB patients were analyzed using ultra-high-performance liquid chromatography-mass spectrometry. Differential metabolites were identified based on a variable importance in projection >1, false discovery rate-adjusted p-value <0.05. Pathway enrichment analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG). Machine learning, including support vector machines and random forest, were employed to screen for diagnostic biomarkers, which were validated by receiver operating characteristic curves.</div></div><div><h3>Results</h3><div>Among the 1294 detected metabolites, 89 exhibited significant differences between the two groups. By integrating KEGG enrichment with topological analysis, phenylalanine, tyrosine, and tryptophan biosynthesis possessed the highest impact, followed by phenylalanine metabolism, and aminoacyl-tRNA biosynthesis. Machine learning identified four biomarkers: Leu-Ala [area under the curve (AUC) = 0.8292], evodiamine (AUC = 0.7558), fenazaquin (AUC = 0.7175), and acetol (AUC = 0.7117). Leu-Ala demonstrated the highest diagnostic accuracy, with a sensitivity of 73.5 % and specificity 86.7 % at a cutoff value of 0.62.</div></div><div><h3>Conclusions</h3><div>Untargeted metabolomics revealed dysregulation in the biosynthesis of phenylalanine, tyrosine, and tryptophan, phenylalanine metabolism, as well as in aminoacyl-tRNA biosynthesis in LNTB. Additional, Leu-Ala was identified as a novel diagnostic biomarker. The integrating of metabolomics with machine learning presents a promising approach for LNTB detection, though larger validation studies are necessary.</div></div>\",\"PeriodicalId\":435,\"journal\":{\"name\":\"Talanta\",\"volume\":\"297 \",\"pages\":\"Article 128583\"},\"PeriodicalIF\":5.6000,\"publicationDate\":\"2025-07-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Talanta\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0039914025010732\",\"RegionNum\":1,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CHEMISTRY, ANALYTICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Talanta","FirstCategoryId":"92","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0039914025010732","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, ANALYTICAL","Score":null,"Total":0}

引用次数: 0

摘要

背景：淋巴结结核（LNTB）是肺外结核最常见的形式；然而，由于重叠的临床特征和不理想的诊断方法，将其与非lntb区分开来仍然具有挑战性。目前LNTB的诊断方法缺乏敏感性和特异性。本研究旨在表征LNTB和非LNTB患者之间的代谢差异，阐明LNTB的病理机制，并利用机器学习模型识别诊断生物标志物。方法采用超高效液相色谱-质谱法对40例LNTB患者和30例非LNTB患者的血清样本进行分析。差异代谢物是根据预测的不同重要性来确定的>；1，假发现率调整p值<；0.05。通路富集分析使用京都基因与基因组百科全书（KEGG）进行。采用机器学习（包括支持向量机和随机森林）筛选诊断性生物标志物，并通过受试者工作特征曲线进行验证。结果在1294种代谢物中，89种代谢物在两组间存在显著差异。将KEGG富集与拓扑分析相结合，发现苯丙氨酸、酪氨酸和色氨酸的生物合成影响最大，其次是苯丙氨酸代谢，然后是氨基酰基trna的生物合成。机器学习鉴定出四种生物标志物：Leu-Ala[曲线下面积(AUC) = 0.8292]、evodiamine （AUC = 0.7558）、fenazaquin （AUC = 0.7175）和acetol （AUC = 0.7117）。Leu-Ala的诊断准确率最高，敏感性为73.5%，特异性为86.7%，临界值为0.62。结论靶向代谢组学揭示了LNTB中苯丙氨酸、酪氨酸和色氨酸的生物合成、苯丙氨酸的代谢以及氨基酰基trna的生物合成异常。此外，Leu-Ala被确定为一种新的诊断性生物标志物。代谢组学与机器学习的结合为LNTB检测提供了一种很有前途的方法，尽管还需要更大规模的验证研究。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

Non-targeted metabolomics and machine learning reveal metabolic dysregulation in lymph node tuberculosis

查看原文本刊更多论文

Non-targeted metabolomics and machine learning reveal metabolic dysregulation in lymph node tuberculosis

Background

Lymph node tuberculosis (LNTB) is the most prevalent form of extrapulmonary tuberculosis; however, differentiating it from non-LNTB remains challenging due to overlapping clinical features and suboptimal diagnostic methods. Current diagnostic methods for LNTB lack both sensitivity and specificity. This study aimed to characterize the metabolic differences between LNTB and non-LNTB patients, elucidate the pathological mechanisms underlying LNTB, and identify diagnostic biomarkers using machine learning models.

Methods

Serum samples from 40 LNTB patients and 30 non-LNTB patients were analyzed using ultra-high-performance liquid chromatography-mass spectrometry. Differential metabolites were identified based on a variable importance in projection >1, false discovery rate-adjusted p-value <0.05. Pathway enrichment analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG). Machine learning, including support vector machines and random forest, were employed to screen for diagnostic biomarkers, which were validated by receiver operating characteristic curves.

Results

Among the 1294 detected metabolites, 89 exhibited significant differences between the two groups. By integrating KEGG enrichment with topological analysis, phenylalanine, tyrosine, and tryptophan biosynthesis possessed the highest impact, followed by phenylalanine metabolism, and aminoacyl-tRNA biosynthesis. Machine learning identified four biomarkers: Leu-Ala [area under the curve (AUC) = 0.8292], evodiamine (AUC = 0.7558), fenazaquin (AUC = 0.7175), and acetol (AUC = 0.7117). Leu-Ala demonstrated the highest diagnostic accuracy, with a sensitivity of 73.5 % and specificity 86.7 % at a cutoff value of 0.62.

Conclusions

Untargeted metabolomics revealed dysregulation in the biosynthesis of phenylalanine, tyrosine, and tryptophan, phenylalanine metabolism, as well as in aminoacyl-tRNA biosynthesis in LNTB. Additional, Leu-Ala was identified as a novel diagnostic biomarker. The integrating of metabolomics with machine learning presents a promising approach for LNTB detection, though larger validation studies are necessary.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Talanta 化学-分析化学

CiteScore

12.30

自引率

4.90%

发文量

861

审稿时长

29 days

期刊介绍： Talanta provides a forum for the publication of original research papers, short communications, and critical reviews in all branches of pure and applied analytical chemistry. Papers are evaluated based on established guidelines, including the fundamental nature of the study, scientific novelty, substantial improvement or advantage over existing technology or methods, and demonstrated analytical applicability. Original research papers on fundamental studies, and on novel sensor and instrumentation developments, are encouraged. Novel or improved applications in areas such as clinical and biological chemistry, environmental analysis, geochemistry, materials science and engineering, and analytical platforms for omics development are welcome. Analytical performance of methods should be determined, including interference and matrix effects, and methods should be validated by comparison with a standard method, or analysis of a certified reference material. Simple spiking recoveries may not be sufficient. The developed method should especially comprise information on selectivity, sensitivity, detection limits, accuracy, and reliability. However, applying official validation or robustness studies to a routine method or technique does not necessarily constitute novelty. Proper statistical treatment of the data should be provided. Relevant literature should be cited, including related publications by the authors, and authors should discuss how their proposed methodology compares with previously reported methods.