Identifying Lipid Metabolism-Related Therapeutic Targets and Diagnostic Markers for Lung Adenocarcinoma by Mendelian Randomization and Machine Learning Analysis.

IF 2.3 3区医学 Q3 ONCOLOGY

Thoracic Cancer Pub Date : 2025-03-01 DOI:10.1111/1759-7714.70020

Su Wei, Zhou Guangyao, Tian Xiangdong, Guo Feng, Zhang Lianmin, Zhang Zhenfa

{"title":"Identifying Lipid Metabolism-Related Therapeutic Targets and Diagnostic Markers for Lung Adenocarcinoma by Mendelian Randomization and Machine Learning Analysis.","authors":"Su Wei, Zhou Guangyao, Tian Xiangdong, Guo Feng, Zhang Lianmin, Zhang Zhenfa","doi":"10.1111/1759-7714.70020","DOIUrl":null,"url":null,"abstract":"Background: Lipid metabolic disorders are emerging as a recognized influencing factors of lung adenocarcinoma (LUAD). This study aims to investigate the influence of lipid metabolism-related genes (LMRGs) on the diagnosis and treatment of LUAD and to identify significant biomarkers.Methods: DESeq2 and robust rank aggregation (RRA) analyses were employed to determine the differential expression of LMRGs from TCGA-LUAD and five GEO datasets. Mendelian randomization (MR) was conducted utilizing protein quantitative trait loci (pQTLs) in the deCODE, prot-a, and UKB-PPP Study to estimate causal relationships between plasma proteins and LUAD within the ieu-a-984, ieu-a-965, and FinnGen R10 cohorts as potential drug targets of LUAD. Subsequently, an optimal machine learning model for diagnosing LUAD was established by comparing four models: support vector machine, random forest (RF), glmBoost, and eXtreme Gradient Boosting. Finally, the diagnostic performance of five plasma proteins was validated through nomogram analysis, calibration curve assessment, decision curve analysis (DCA), independent internal and external datasets.Result: A total of five biomarkers were identified from 1034 LMRGs via MR and differential expression analysis. TNFRSF21 exhibited a positive association with LUAD risk; conversely, BCHE, FABP4, LPL, and PLBD1 demonstrated negative correlations with this risk. The RF machine learning model was determined to be the optimal model for diagnosing LUAD using these five plasma proteins. Ultimately, nomogram construction, calibration curve analysis, DCA, as well as independent internal and external dataset validation confirmed that these biomarkers exhibit excellent diagnostic performance.Conclusions: BCHE, FABP4, LPL, PLBD1, and TNFRSF21 represent potential novel reliable diagnostic markers as well as therapeutic targets for LUAD.","PeriodicalId":23338,"journal":{"name":"Thoracic Cancer","volume":"16 6","pages":"e70020"},"PeriodicalIF":2.3000,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11922676/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Thoracic Cancer","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1111/1759-7714.70020","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ONCOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Lipid metabolic disorders are emerging as a recognized influencing factors of lung adenocarcinoma (LUAD). This study aims to investigate the influence of lipid metabolism-related genes (LMRGs) on the diagnosis and treatment of LUAD and to identify significant biomarkers.

Methods: DESeq2 and robust rank aggregation (RRA) analyses were employed to determine the differential expression of LMRGs from TCGA-LUAD and five GEO datasets. Mendelian randomization (MR) was conducted utilizing protein quantitative trait loci (pQTLs) in the deCODE, prot-a, and UKB-PPP Study to estimate causal relationships between plasma proteins and LUAD within the ieu-a-984, ieu-a-965, and FinnGen R10 cohorts as potential drug targets of LUAD. Subsequently, an optimal machine learning model for diagnosing LUAD was established by comparing four models: support vector machine, random forest (RF), glmBoost, and eXtreme Gradient Boosting. Finally, the diagnostic performance of five plasma proteins was validated through nomogram analysis, calibration curve assessment, decision curve analysis (DCA), independent internal and external datasets.

Result: A total of five biomarkers were identified from 1034 LMRGs via MR and differential expression analysis. TNFRSF21 exhibited a positive association with LUAD risk; conversely, BCHE, FABP4, LPL, and PLBD1 demonstrated negative correlations with this risk. The RF machine learning model was determined to be the optimal model for diagnosing LUAD using these five plasma proteins. Ultimately, nomogram construction, calibration curve analysis, DCA, as well as independent internal and external dataset validation confirmed that these biomarkers exhibit excellent diagnostic performance.

Conclusions: BCHE, FABP4, LPL, PLBD1, and TNFRSF21 represent potential novel reliable diagnostic markers as well as therapeutic targets for LUAD.

查看原文本刊更多论文

通过孟德尔随机化和机器学习分析确定肺腺癌脂质代谢相关的治疗靶点和诊断标志物。

背景：脂质代谢紊乱正在成为公认的肺腺癌（LUAD）的影响因素。本研究旨在探讨脂质代谢相关基因（LMRGs）对LUAD诊断和治疗的影响，并鉴定有意义的生物标志物。方法：采用DESeq2和稳健秩聚集（robust rank aggregation， RRA）分析来确定来自TCGA-LUAD和五个GEO数据集的LMRGs的差异表达。利用deCODE、prot-a和UKB-PPP研究中的蛋白质数量性状位点（pqtl）进行孟德尔随机化（MR），以估计作为LUAD潜在药物靶点的ieu-a-984、ieu-a-965和FinnGen R10队列中血浆蛋白与LUAD之间的因果关系。随后，通过比较支持向量机、随机森林（RF）、glmBoost和eXtreme Gradient Boosting四种模型，建立了诊断LUAD的最优机器学习模型。最后，通过模态图分析、校准曲线评估、决策曲线分析（DCA）、独立的内部和外部数据集验证5种血浆蛋白的诊断性能。结果：通过MR和差异表达分析，从1034个LMRGs中共鉴定出5个生物标志物。TNFRSF21与LUAD风险呈正相关；相反，BCHE、FABP4、LPL和PLBD1与该风险呈负相关。RF机器学习模型被确定为使用这五种血浆蛋白诊断LUAD的最佳模型。最终，nomogram construction, calibration curve analysis， DCA，以及独立的内部和外部数据集验证证实了这些生物标志物具有优异的诊断性能。结论：BCHE、FABP4、LPL、PLBD1和TNFRSF21是潜在的新型可靠的LUAD诊断标志物和治疗靶点。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Thoracic Cancer ONCOLOGY-RESPIRATORY SYSTEM

CiteScore

5.20

自引率

3.40%

发文量

439

审稿时长

2 months

期刊介绍： Thoracic Cancer aims to facilitate international collaboration and exchange of comprehensive and cutting-edge information on basic, translational, and applied clinical research in lung cancer, esophageal cancer, mediastinal cancer, breast cancer and other thoracic malignancies. Prevention, treatment and research relevant to Asia-Pacific is a focus area, but submissions from all regions are welcomed. The editors encourage contributions relevant to prevention, general thoracic surgery, medical oncology, radiology, radiation medicine, pathology, basic cancer research, as well as epidemiological and translational studies in thoracic cancer. Thoracic Cancer is the official publication of the Chinese Society of Lung Cancer, International Chinese Society of Thoracic Surgery and is endorsed by the Korean Association for the Study of Lung Cancer and the Hong Kong Cancer Therapy Society. The Journal publishes a range of article types including: Editorials, Invited Reviews, Mini Reviews, Original Articles, Clinical Guidelines, Technological Notes, Imaging in thoracic cancer, Meeting Reports, Case Reports, Letters to the Editor, Commentaries, and Brief Reports.