Machine Learning Selection of Most Predictive Brain Proteins Suggests Role of Sugar Metabolism in Alzheimer’s Disease

Abstract

Background: The complex and not yet fully understood etiology of Alzheimer’s disease (AD) shows important proteopathic signs which are unlikely to be linked to a single protein. However, protein subsets from deep proteomic datasets can be useful in stratifying patient risk, identifying stage dependent disease markers, and suggesting possible disease mechanisms. Objective: The objective was to identify protein subsets that best classify subjects into control, asymptomatic Alzheimer’s disease (AsymAD), and AD. Methods: Data comprised 6 cohorts; 620 subjects; 3,334 proteins. Brain tissue-derived predictive protein subsets for classifying AD, AsymAD, or control were identified and validated with label-free quantification and machine learning. Results: A 29-protein subset accurately classified AD (AUC = 0.94). However, an 88-protein subset best predicted AsymAD (AUC = 0.85) or Control (AUC = 0.89) from AD (AUC = 0.96). AD versus Control: APP, DHX15, NRXN1, PBXIP1, RABEP1, STOM, and VGF. AD versus AsymAD: ALDH1A1, BDH2, C4A, FABP7, GABBR2, GNAI3, PBXIP1, and PKAR1B. AsymAD versus Control: APP, C4A, DMXL1, EXOC2, PITPNB, REBEP1, and VGF. Additional predictors: DNAJA3, PTBP2, SLC30A9, VAT1L, CROCC, PNP, SNCB, PRKAR1B, ENPP6, HAPLN2, PSMD4, and CMAS. Conclusion: Biomarkers were dynamically separable across disease stages. Predictive proteins were significantly enriched to sugar metabolism.

What Threshold of Amyloid Reduction Is Necessary to Meaningfully Improve Cognitive Function in Transgenic Alzheimer’s Disease Mice?

“…Ideally, patients at high risk of acquiring dementia should be identified as early as possible. Early identification can be made possible with cerebrospinal fluid, genetic, and/or blood-based biomarkers combined with machine learning-based clinical recommendation systems, which use personalized data to calculate a specific dementia risk, dementia onset date, and confidence interval on the prediction(s) [53,54]. With such upfront quantitative forecasting available in the future, a physician could best inform at-risk asymptomatic patients of the pros and cons of prophylactic therapies aimed at preventing symptomatic AD.…”

Uncovering the Achilles heel of genetic heterogeneity: machine learning-based classification and immunological properties of necroptosis clusters in Alzheimer’s disease

“…Lately, machine learning-based algorithms, combing the clinical, digital pathology, multi-omics data, have been developed and applied in the diagnosis of diseases ( Gao et al, 2022 ), drug developments ( Rodriguez et al, 2021 ), and explorations of AD mechanisms ( Tandon et al, 2023 ). Such algorithms provide valuable tools for interpreting the heterogeneity and classifications of diseases.…”

“…Likewise, NK cells were found to contribute to neuroinflammation and AD-associated cognitive decline (Zhang et al, 2020), suggesting that targeting the innate immune cells might be novel avenues for the treatment of AD (Dubois et al, 2023). Such observations highlight a sophisticated crosstalk orchestrated by varies cellular identities in the microenvironment driven by innate and adaptive immunity, which Lately, machine learning-based algorithms, combing the clinical, digital pathology, multi-omics data, have been developed and applied in the diagnosis of diseases (Gao et al, 2022), drug developments (Rodriguez et al, 2021), and explorations of AD mechanisms (Tandon et al, 2023). Such algorithms provide valuable tools for interpreting the heterogeneity and classifications of diseases.…”

Impact of common ALDH2 inactivating mutation and alcohol consumption on Alzheimer’s disease

“…Also of interest may be the role of ALDH1A1 in AD, an enzyme involved in alcohol-induced facial flushing, alcohol sensitivity and dependence in Caucasians ( Yoshida et al, 1989 ; Spence et al, 2003 ), and in acetaldehyde detoxification ( Marchitti et al, 2008 ). Recently, the relationship between ALDH1A1 and AD has been reported ( Nikhil et al, 2019 ; Li X. et al, 2021 ; Tandon et al, 2023 ). Both epidemiological studies in AD patients, determining ALDH1A1 genotype and alcohol consumption, and mechanistic studies in mice with such mutations will allow a better assessment of the potential risk of alcohol consumptions for AD development in subjects with ALDH1A1 insufficiency.…”