Alzheimer's disease (AD) is a universal neurodegenerative disease in older adults with incurable and progressive properties, urging for precise monitoring to perform timely treatment to delay its progression. Herein, we introduced a non-targeting magnetic metal-organic framework probe coupled with high-throughput mass spectrometry, creating a rapid screening strategy for highly specific peptides associated with AD. Notably, an elution-free extraction process was proposed, significantly reducing sample preprocessing time while simultaneously ensuring the efficient detection of captured peptides. Using this elution-free extraction process, high-quality peptide profiles were rapidly extracted from the hundreds of samples from both diseased and healthy individuals. By integrating machine learning algorithms, LC-MS/MS, and Uniprot database searching, we identified three specific serum endogenous peptides ( = 4215.41, 2884.77 and 2704.61) closely associated with AD. Remarkably, with the use of any single specific peptide, the AUC (Area Under the Curve) values can reach approximately 0.9 during monitoring AD. Moreover, integrating three specific biomarkers provides a robust basis for machine learning algorithms to build monitoring models, with AUC value up to 1.000. This work represents a substantial advancement in the development of peptide-specific precise monitoring approaches for complex diseases, serving as a catalyst for increased dedication to the molecular detection field.

中文翻译：

---

用于高度特异性肽介导的精确疾病监测的非靶向磁性金属有机框架探针

阿尔茨海默病（AD）是一种常见于老年人的神经退行性疾病，具有无法治愈和进行性的特点，迫切需要精确监测以及时治疗以延缓其进展。在此，我们引入了一种非靶向磁性金属有机框架探针与高通量质谱联用，为与 AD 相关的高度特异性肽创建了快速筛选策略。值得注意的是，提出了免洗脱提取工艺，显着减少了样品预处理时间，同时确保了捕获肽的有效检测。使用这种免洗脱提取工艺，可以从患病和健康个体的数百个样品中快速提取出高质量的肽谱。通过整合机器学习算法、LC-MS/MS 和 Uniprot 数据库搜索，我们鉴定了与 AD 密切相关的三种特定血清内源肽（= 4215.41、2884.77 和 2704.61）。值得注意的是，使用任何单一特定肽，在监测 AD 期间，AUC（曲线下面积）值可以达到约 0.9。此外，整合三种特定生物标志物为机器学习算法构建监测模型提供了坚实的基础，AUC值高达1.000。这项工作代表了针对复杂疾病的肽特异性精确监测方法的开发取得了重大进展，成为进一步致力于分子检测领域的催化剂。