# Proteome Profiling of Cerebrospinal Fluid and Machine Learning Reveal Protein Classifiers of Two Forms of Alzheimer’s Disease Characterized by Increased or Not Altered Levels of Tau

## TL;DR

This study uses advanced proteomics and machine learning to identify a 15-protein signature in cerebrospinal fluid that can distinguish two subtypes of Alzheimer’s disease and non-AD controls.

## Contribution

A novel 15-protein biomarker panel was identified using DIA proteomics and machine learning to classify Alzheimer’s subtypes and controls.

## Key findings

- Distinct protein signatures were found for Aβ+/tau+ and Aβ+/tau− Alzheimer’s subtypes.
- A 15-protein panel accurately distinguished AD subtypes and non-AD controls across datasets.
- Several proteins in the panel were elevated in preclinical stages, suggesting early diagnostic potential.

## Abstract

Alzheimer’s disease (AD) is a multifactorial neurodegenerative disorder that presents with heterogeneous clinical and pathological features, necessitating improved biomarkers for accurate diagnosis and patient stratification. In this study, we applied a data-independent acquisition-based proteomics workflow to cerebrospinal fluid (CSF) samples from 138 individuals, including AD patients with high (Aβ+/tau+) or normal (Aβ+/tau−) CSF tau levels, and non-AD controls. Analysis using an Astral mass spectrometer enabled unprecedented proteome depth, identifying 2661 proteins with high data completeness. Comparative proteomic profiling revealed distinct protein signatures for Aβ+/tau+ and Aβ+/tau− subtypes. These findings were validated using an independent internal cohort and further corroborated with publicly available datasets from larger external AD cohorts, demonstrating the robustness and reproducibility of our results. Using machine learning, we identified a panel of 15 protein classifiers that accurately distinguished the two AD subtypes and controls across datasets. Notably, several of these proteins were elevated in the preclinical stage, underscoring their potential utility for early diagnosis and stratification. Together, our results demonstrate the power of data-independent acquisition proteomics on the Astral platform, combined with machine learning, to uncover subtype-specific biomarkers of AD and support the development of personalized diagnostic strategies.

•Astral MS and DIA enabled deep and comprehensive CSF proteome profiling.•Distinct CSF proteomes identified in Aβ+/tau+ and Aβ+/tau− AD subtypes.•Machine learning revealed a 15-protein panel distinguishing AD subtypes.•Classifier differentiates Aβ+/tau+ and Aβ+/tau− from non-AD controls.•The 15-protein signature holds promise as a biomarker panel for AD.

Astral MS and DIA enabled deep and comprehensive CSF proteome profiling.

Distinct CSF proteomes identified in Aβ+/tau+ and Aβ+/tau− AD subtypes.

Machine learning revealed a 15-protein panel distinguishing AD subtypes.

Classifier differentiates Aβ+/tau+ and Aβ+/tau− from non-AD controls.

The 15-protein signature holds promise as a biomarker panel for AD.

Deep DIA-based proteomics using high-resolution Astral MS enabled comprehensive profiling of cerebrospinal fluid from Alzheimer’s patients. Machine learning identified a 15-protein signature that distinguishes Aβ+/tau+ and Aβ+/tau− subtypes from each other and from non-AD controls. These findings highlight the potential of CSF proteomics to refine Alzheimer’s disease stratification and support precision medicine approaches.