Deep learning analysis of UPLC-MS/MS-based metabolomics data to predict Alzheimer's disease

Authors

Kesheng Wang, School of Nursing, Health Sciences Center, West Virginia University, Morgantown, WV 26506, USA. Electronic address: kesheng.wang@hsc.wvu.edu.
Laurie A. Theeke, School of Nursing, The George Washington University, Ashburn, VA 20147, USA.
Christopher Liao, Department of Electrical and Computer Engineering, Boston University, MA 02215, USA.
Nianyang Wang, Department of Health Policy and Management, School of Public Health, University of Maryland, College Park, MD 20742, USA.
Yongke Lu, Department of Biomedical Sciences, Joan C. Edwards School of Medicine, Marshall University, Huntington, WV 25755, USA.
Danqing Xiao, Department of STEM, School of Arts and Sciences, Regis College, Weston, MA 02493, USA.
Chun Xu, Department of Health and Biomedical Sciences, College of Health Professions, University of Texas Rio Grande Valley, Brownsville, TX 78520, USA. Electronic address: chun.xu@utrgv.edu.

Document Type

Journal Article

Publication Date

10-15-2023

Journal

Journal of the neurological sciences

Volume

453

DOI

10.1016/j.jns.2023.120812

Keywords

APOE-ε4; Alzheimer's disease; Deep learning; Glycohyodeoxycholic acid (GHDCA); H2O package; LASSO; Metabolomics

Abstract

OBJECTIVE: Metabolic biomarkers can potentially inform disease progression in Alzheimer's disease (AD). The purpose of this study is to identify and describe a new set of diagnostic biomarkers for developing deep learning (DL) tools to predict AD using Ultra Performance Liquid Chromatography Mass Spectrometry (UPLC-MS/MS)-based metabolomics data. METHODS: A total of 177 individuals, including 78 with AD and 99 with cognitive normal (CN), were selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort along with 150 metabolomic biomarkers. We performed feature selection using the Least Absolute Shrinkage and Selection Operator (LASSO). The H2O DL function was used to build multilayer feedforward neural networks to predict AD. RESULTS: The LASSO selected 21 metabolic biomarkers. To develop DL models, the 21 biomarkers identified by LASSO were imported into the H2O package. The data was split into 70% for training and 30% for validation. The best DL model with two layers and 18 neurons achieved an accuracy of 0.881, F1-score of 0.892, and AUC of 0.873. Several metabolomic biomarkers involved in glucose and lipid metabolism, in particular bile acid metabolites, were associated with APOE-ε4 allele and clinical biomarkers (Aβ42, tTau, pTau), cognitive assessments [the Alzheimer's Disease Assessment Scale-cognitive subscale 13 (ADAS13), the Mini-Mental State Examination (MMSE)], and hippocampus volume. CONCLUSIONS: This study identified a new set of diagnostic metabolomic biomarkers for developing DL tools to predict AD. These biomarkers may help with early diagnosis, prognostic risk stratification, and/or early treatment interventions for patients at risk for AD.

APA Citation

Wang, Kesheng; Theeke, Laurie A.; Liao, Christopher; Wang, Nianyang; Lu, Yongke; Xiao, Danqing; and Xu, Chun, "Deep learning analysis of UPLC-MS/MS-based metabolomics data to predict Alzheimer's disease" (2023). GW Authored Works. Paper 3601.
https://hsrc.himmelfarb.gwu.edu/gwhpubs/3601

Department

Nursing Faculty Publications