Ribosomal proteins are blood biomarkers and associated with CD4+ T cell activation in Alzheimer's disease: a study based on machine learning strategies and scRNA-Seq data validation.

backgroundAlzheimer's disease (AD) is a widespread neurodegenerative disease that primarily affects the elderly. Unfortunately, the lack of convenient early diagnostic tools makes it difficult to intervene and treat the disease during its initial stages.

methodsWe obtained four bulk and single-cell RNA-sequencing peripheral blood samples related to AD from public databases. Using Boruta and LASSO machine learning algorithms, we screened the signature genes and constructed a diagnostic model using lightGBM. The model was further validated in a test cohort. Additionally, we extracted hub biomarkers using the protein-protein interactions method and validated them in a single-cell RNA-seq dataset.

resultsOur analysis revealed the identification of 37 AD-related peripheral blood signature genes, with their main enrichment in ribosome-related biological functions. Four core biomarkers, RPL24, RPL5, RPS27A, and RPS4X, were identified and exhibited good diagnostic power in the testing cohort. Immune infiltration analysis revealed a higher proportion of CD4+ T cells in AD patients' peripheral blood compared to healthy controls, with a negative correlation with the four ribosome-associated core genes. Validation in a single-cell RNA-seq dataset confirmed these findings.

conclusionsRibosomal family proteins have the potential to serve as biomarkers for the diagnosis and treatment of AD, and are associated with CD4+ T cell activation.