机器学习方法在代谢组学数据统计分析中的新近研究进展

doi:10.3969/j.issn.1006-5253.2024.04.015

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (4029 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS) 背景资料

文章导读

摘要代谢组学是研究生物体内所有小分子代谢产物的学科领域。其利用先进的技术平台如质谱和核磁共振等，可以全面、高通量地检测代谢物，为疾病的早期诊断、发病机制研究和个性化治疗提供新思路。由于代谢组学数据具有多样性、高维度、动态性、噪声和变异性等特点，为其数据分析方法研究提出了严峻挑战。本文拟通过归纳总结近年来主要的机器学习法尤其是其改进方法在代谢组学数据统计分析中的进展，为进行有效的数据分析，充分发挥代谢组学在医学研究中的应用提供重要依据。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	郝攀峰
	潘澍泽
	秦林原
	武晓岩

关键词 ：代谢组学, , 医学研究, , 机器学习方法

Abstract：Metabolomics is a disciplinary field that investigates all small-molecule metabolites within the biological organism. Utilizing advanced technological platforms such as mass spectrometry and nuclear magnetic resonance, comprehensive and high-throughput detection of metabolites can offer new avenues for early disease diagnosis, elucidation of pathogenic mechanisms, and personalized therapeutic approaches. Due to the diversity, high dimensionality, dynamics, noise, and variability inherent in metabolomics data, it poses formidable challenges for the research and development of data analysis methods. This paper aims to provide a comprehensive overview of recent advancements of machine learning methods for statistical analysis on metabolomics data, with particular focus on improved approaches. By synthesizing these methodologies, the intention is to offer a crucial foundation for effective data analysis and to fully harness the potential of metabolomics in medical research.

Key words： metabolomics medical research machine learning method

收稿日期: 2024-03-21

PACS:

基金资助:国家自然科学基金面上项目（82073536）；
广西自然科学基金面上项目（2022GXNSFAA035634）

通讯作者: 武晓岩，Email：wxywxy_79@126.com

引用本文:

郝攀峰，潘澍泽，秦林原，武晓岩 . 机器学习方法在代谢组学数据统计分析中的新近研究进展 [J]. 中国医院统计, 2024, 31(4): 313-320.
Hao Panfeng，Pan Shuze，Qin Linyuan，Wu Xiaoyan. Recent development of machine learning methods applied on statistical analysis for metabolomics data. journal1, 2024, 31(4): 313-320.

链接本文:

http://manu58.magtech.com.cn/Jwk_zgyytj/CN/10.3969/j.issn.1006-5253.2024.04.015 或 http://manu58.magtech.com.cn/Jwk_zgyytj/CN/Y2024/V31/I4/313

［1］NICHOLSON J K, LINDON J C, HOLMES E. 'Metabonomics': Understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data［J］. Xenobiotica, 1999,29(11):1181-1189. DOI: 10.1080/004982599238047.

［2］NAM M, JO S R, KIM Y C, et al. UPLC-QTOF-MS-based metabolomics and antioxidant capacity of Codonopsis lanceolata from different geographical origins［J］. Foods, 2023,12(2):267. DOI: 10.3390/foods12020267.

［3］PEARSON K. LIII. On lines and planes of closest fit to systems of points in space［J］. Lond Edinb Dublin Philos Mag J Sci, 1901,2(11):559-572. DOI: 10.1080/14786440109462720.

［4］MI J X, ZHANG Y N, LAI Z H, et al. Principal Component Analysis based on Nuclear norm Minimization［J］. Neural Netw, 2019,118:1-16. DOI: 10.1016/j.neunet.2019.05.020.

［5］TSUYUZAKI K, SATO H, SATO K, et al. Benchmarking principal component analysis for large-scale single-cell RNA-sequencing［J］. Genome Biol, 2020,21(1):9. DOI: 10.1186/s13059-019-1900-3.

［6］ZHENG H, ZHU Y J, SHAO X G, et al. Distinct metabolic signatures of hormone-sensitive and castration-resistant prostate cancer revealed by a 1H NMR-based metabolomics of biopsy tissue［J］. J Proteome Res, 2020,19(9):3741-3749. DOI: 10.1021/acs.jproteome.0c00282.

［7］YANG B, ZHANG C, CHENG S, et al. Novel metabolic signatures of prostate cancer revealed by 1H-NMR metabolomics of urine［J］. Diagnostics, 2021,11(2):149. DOI: 10.3390/diagnostics11020149.

［8］WARD J H. Hierarchical grouping to optimize an objective function［J］. J Am Stat Assoc, 1963,58(301):236-244. DOI: 10.2307/2282967.

［9］PETEGROSSO R, LI Z L, KUANG R. Machine learning and statistical methods for clustering single-cell RNA-sequencing data［J］. Brief Bioinform, 2020,21(4):1209-1223. DOI: 10.1093/bib/bbz063.

［10］KUDAL P, PATNAIK A, DAWAR S, et al. Segmentation of OECD countries on the basis of selected global environmental indicators using k-means non-hierarchical clustering［J］. Environ Sci Pollut Res Int, 2024,31(7):10334-10345. DOI: 10.1007/s11356-023-26679-x.

［11］ZOU Z H, HUA K, ZHANG X G. HGC: Fast hierarchical clustering for large-scale single-cell data［J］. Bioinformatics, 2021,37(21):3964-3965. DOI: 10.1093/bioinformatics/btab420.

［12］CHENG X Q, YAN C, JIANG H, et al. scHOIS: Determining cell heterogeneity through hierarchical clustering based on optimal imputation strategy［J］. IEEE/ACM Trans Comput Biol Bioinform, 2023,20(2):1431-1444. DOI: 10.1109/TCBB.2022.3203592.

［13］HAN Y Z, MA Z T, ZHOU M X, et al. Metabolomic profiling for drug-induced liver injury with autoantibodies［J］. Int Immunopharmacol, 2022,111:109084. DOI: 10.1016/j.intimp.2022.109084.

［14］MACQUEEN J. Some methods for classification and analysis of multivariate observations［J］. Berkeley Symp Math Stat Probab, 1967: 1(14): 281-297.

［15］ALAKWAA F M, SAVELIEFF M G. Bioinformatics analysis of metabolomics data unveils association of metabolic signatures with methylation in breast cancer［J］. J Proteome Res, 2020,19(7):2879-2889. DOI: 10.1021/acs.jproteome.9b00755.

［16］ROUMANI A M, MADKOUR A, OUZZANI M, et al. BioNetApp: An interactive visual data analysis platform for molecular expressions［J］. PLoS One, 2019,14(2):e0211277. DOI: 10.1371/journal.pone.0211277.

［17］URPI-SARDA M, ALMANZA-AGUILERA E, LLORACH R, et al. Non-targeted metabolomic biomarkers and metabotypes of type 2 diabetes: A cross-sectional study of PREDIMED trial participants［J］. Diabetes Metab, 2019,45(2):167-174. DOI: 10.1016/j.diabet.2018.02.006.

［18］ST-HLE L, WOLD S. Partial least squares analysis with cross-validation for the two-class problem: A Monte Carlo study［J］. J Chemom, 1987,1(3):185-196. DOI: 10.1002/cem.1180010306.

［19］AMINU M, AHMAD N A. Complex chemical data classification and discrimination using locality preserving partial least squares discriminant analysis［J］. ACS Omega, 2020,5(41):26601-26610. DOI: 10.1021/acsomega.0c03362.

［20］LEE L C, LIONG C Y, JEMAIN A A. Partial least squares-discriminant analysis (PLS-DA) for classification of high-dimensional (HD) data: A review of contemporary practice strategies and knowledge gaps［J］. Analyst, 2018,143(15):3526-3539. DOI: 10.1039/c8an00599k.

［21］KANG C D, ZHANG Y Y, ZHANG M Y, et al. Screening of specific quantitative peptides of beef by LCMS/MS coupled with OPLS-DA［J］. Food Chem, 2022,387:132932. DOI: 10.1016/j.foodchem.2022.132932.

［22］DU Y, FAN P Z, ZOU L H, et al. Serum metabolomics study of papillary thyroid carcinoma based on HPLC-Q-TOF-MS/MS［J］. Front Cell Dev Biol, 2021,9:593510. DOI: 10.3389/fcell.2021.593510.

［23］MCCULLOCH W S, PITTS W. A logical calculus of the ideas immanent in nervous activity［J］. Bull Math Biophys, 1943,5(4):115-133. DOI: 10.1007/BF02478259.
［24］LEE M Y, HU T. Computational methods for the discovery of metabolic markers of complex traits［J］. Metabolites, 2019,9(4):66. DOI: 10.3390/metabo9040066.

［25］ALSHAMMARI F, ALAM M B, NAZNIN M, et al. Optimization, metabolomic analysis, antioxidant potential andDepigmenting activity of polyphenolic compounds fromUnmature ajwa date seeds (Phoenix dactylifera L.) using ultrasonic-assisted extraction［J］. Antioxidants, 2024,13(2):238. DOI: 10.3390/antiox13020238.

［26］TSIGELNY I F. Artificial intelligence in drug combination therapy［J］. Brief Bioinform, 2019,20(4):1434-1448. DOI: 10.1093/bib/bby004.

［27］LIU J, LIU L X, GUO W, et al. A new methodology for sensory quality assessment of garlic based on metabolomics and an artificial neural network［J］. RSC Adv, 2019,9(31):17754-17765. DOI: 10.1039/c9ra01978b.

［28］TOUSSAINT P A, LEISER F, THIEBES S, et al. Explainable artificial intelligence for omics data: A systematic mapping study［J］. Brief Bioinform, 2023,25(1):bbad453. DOI: 10.1093/bib/bbad453.

［29］[KG*2]〖ZK(]CHEN N, WANG H B, WU B Q, et al. Using random forest to detect multiple inherited metabolic diseases simultaneously based on GCMS urinary metabolomics［J］. Talanta, 2021,235:122720. DOI: 10.1016/j.talanta.2021.122720.

［30］ABDELMOULA W M, STOPKA S A, RANDALL E C, et al. massNet: Integrated processing and classification of spatially resolved mass spectrometry data using deep learning for rapid tumor delineation［J］. Bioinformatics, 2022,38(7):2015-2021. DOI: 10.1093/bioinformatics/btac032.

［31］CORTES C, VAPNIK V. Support-vector networks［J］. Mach Learn, 1995,20(3):273-297. DOI: 10.1007/BF00994018.

［32］BARBERIS E, KHOSO S, SICA A, et al. Precision medicine approaches with metabolomics and artificial intelligence［J］. Int J Mol Sci, 2022,23(19):11269. DOI: 10.3390/ijms231911269.

［33］HOU X W, WANG G L, SU G Q, et al. Rapid identification of edible oil species using supervised support vector machine based on low-field nuclear magnetic resonance relaxation features［J］. Food Chem, 2019,280:139-145. DOI: 10.1016/j.foodchem.2018.12.031.

［34］SARKAR S, MALI K. Firefly-SVM predictive model for breast cancer subgroup classification with clinicopathological parameters［J］. Digit Health, 2023,9:20552076231207203. DOI: 10.1177/20552076231207203.

［35］ZHOU S L. Sparse SVM for sufficient data reduction［J］. IEEE Trans Pattern Anal Mach Intell, 2022,44(9):5560-5571. DOI: 10.1109/TPAMI.2021.3075339.

［36］TAKAHASHI Y, UEKI M, YAMADA M, et al. Improved metabolomic data-based prediction of depressive symptoms using nonlinear machine learning with feature selection［J］. Transl Psychiatry, 2020,10(1):157. DOI: 10.1038/s41398-020-0831-9.

［37］ZHENG H, ZHENG P, ZHAO L C, et al. Predictive diagnosis of major depression using NMR-based metabolomics and least-squares support vector machine［J］. Clin Chim Acta, 2017,464:223-227. DOI: 10.1016/j.cca.2016.11.039.

［38］BERGER J O. Statistical decision theory and Bayesian analysis［M］. 2nd ed.［S.l.］: Springer Science & Business Media,1985.

［39］FANG S N, LIN Z Z, ZHANG Z S, et al. Gas hydrate saturation estimates in the Muli permafrost area considering Bayesian discriminant functions［J］. J Petrol Sci Eng, 2020,195:107872. DOI: 10.1016/j.petrol.2020.107872.

［40］HUANG Z Y, KANG M M, LI G Y, et al. Predictive effect of Bayes discrimination in the level of serum protein factors and cognitive dysfunction in schizophrenia［J］. J Psychiatr Res, 2022,151:539-545. DOI: 10.1016/j.jpsychires.2022.05.004.

［41］BOBB J F, VALERI L, CLAUS HENN B, et al. Bayesian kernel machine regression for estimating the health effects of multi-pollutant mixtures［J］. Biostatistics, 2015,16(3):493-508. DOI: 10.1093/biostatistics/kxu058.

［42］MATTA K, LEFEBVRE T, VIGNEAU E, et al. Associations between persistent organic pollutants and endometriosis: A multiblock approach integrating metabolic and cytokine profiling［J］. Environ Int, 2022,158:106926. DOI: 10.1016/j.envint.2021.106926.

［43］HOLLAND J H. Genetic algorithms ［J］. Sci Am, 1992, 267(1): 66-73.DOI:10.1038/scientificamerican0792-66.

［44］LUO J, REN R, GUO K D. The deformation monitoring of foundation pit by back propagation neural network and genetic algorithm and its application in geotechnical engineering［J］. PLoS One, 2020,15(7):e0233398. DOI: 10.1371/journal.pone.0233398.

［45］NOUTAHI E, EL-MABROUK N. GATC: A genetic algorithm for gene tree construction under the Duplication-Transfer-Loss model of evolution［J］. BMC Genomics, 2018,19(Suppl 2):102. DOI: 10.1186/s12864-018-4455-x.

［46］LIU P, EL BASHA M D, LI Y, et al. Deep evolutionary networks with expedited genetic algorithms for medical image denoising［J］. Med Image Anal, 2019,54:306-315. DOI: 10.1016/j.media.2019.03.004.

［47］MOHAMMADPOUR T, BIDGOLI A M, ENAYATIFAR R, et al. Efficient clustering in collaborative filtering recommender system: Hybrid method based on genetic algorithm and gravitational emulation local search algorithm［J］. Genomics, 2019,111(6):1902-1912. DOI: 10.1016/j.ygeno.2019.01.001.

［48］JIN H, MOSELEY H N B. Moiety modeling framework for deriving moiety abundances from mass spectrometry measured isotopologues［J］. BMC Bioinformatics, 2019,20(1):524. DOI: 10.1186/s12859-019-3096-7.

［49］MARJIT S, BHATTACHARYYA T, CHATTERJEE B, et al. Simulated annealing aided genetic algorithm for gene selection from microarray data［J］. Comput Biol Med, 2023,158:106854. DOI: 10.1016/j.compbiomed.2023.106854.

［50］BREIMAN L. Random forests［J］. Mach Learn, 2001,45:5-32. DOI: 10.1023/A:1010933404324.

［51］RHODES J S, CUTLER A, MOON K R. Geometry and accuracy-preserving random forest proximities［J］. IEEE Trans Pattern Anal Mach Intell, 2023,45(9):10947-10959. DOI: 10.1109/TPAMI.2023.3263774.

［52］WALKER A M, CLIFF A, ROMERO J, et al. Evaluating the performance of random forest and iterative random forest based methods when applied to gene expression data［J］. Comput Struct Biotechnol J, 2022,20:3372-3386. DOI: 10.1016/j.csbj.2022.06.037.

［53］KONG Y C, YU T W. forgeNet: A graph deep neural network model using tree-based ensemble classifiers for feature graph construction［J］. Bioinformatics, 2020,36(11):3507-3515. DOI: 10.1093/bioinformatics/btaa164.

［54］MOTAMEDI F, P-REZ-S-NCHEZ H, MEHRIDEHNAVI A, et al. Accelerating big data analysis through LASSO-random forest algorithm in QSAR studies［J］. Bioinformatics, 2022,38(2):469-475. DOI: 10.1093/bioinformatics/btab659.

［55］HAUSCHILD A C, LEMANCZYK M, MATSCHINSKE J, et al. Federated Random Forests can improve local performance of predictive models for various healthcare applications［J］. Bioinformatics, 2022,38(8):2278-2286. DOI: 10.1093/bioinformatics/btac065.

［56］RAZAVI-TERMEH S V, SADEGHI-NIARAKI A, SEO M, et al. Application of genetic algorithm in optimization parallel ensemblebased machine learning algorithms to flood susceptibility mapping using radar satellite imagery［J］. Sci Total Environ, 2023,873:162285. DOI: 10.1016/j.scitotenv.2023.162285.

［57］FRIEDMAN J. Stochastic gradient boosting［J］. Comput Stat \\& Data Anal, 2002,38:367-378. DOI: 10.1016/S0167-9473(01)00065-2.

［58］SU X S, BAI M Y. Stochastic gradient boosting frequency-severity model of insurance claims［J］. PLoS One, 2020,15(8):e0238000. DOI: 10.1371/journal.pone.0238000.

［59］LI B, YU Q Z, PENG L. Ensemble of fast learning stochastic gradient boosting［J］. Commun Stat Simul Comput, 2022,51(1):40-52. DOI: 10.1080/03610918.2019.1645170.