Single-cell, spatial, and machine learning analyses identify ANKIB1 as an oligodendrocyte-anchored palmitoylation-associated candidate biomarker of Alzheimer’s disease

Alzheimer’s disease (AD) is characterized by progressive neurodegeneration and cognitive decline, yet most candidate therapies targeting amyloid-β have failed to robustly modify disease progression, underscoring an incomplete mechanistic understanding. Oligodendrocytes (ODCs) have recently emerged as active contributors to AD pathology, but the cell-type-specific role of protein S-palmitoylation — a reversible lipid post-translational modification regulating membrane targeting and trafficking — in AD remains uncharacterized. Here we performed an integrated multi-omics analysis combining single-cell RNA sequencing (GSE233208; 48,714 cells), five bulk transcriptomic cohorts (769 samples), and 10× Visium spatial transcriptomic data from 16 brain sections.

Palmitoylation activity was quantified at single-cell resolution using a curated 46-gene set; machine learning frameworks (LASSO, random forest, and XGBoost) were used to distil a parsimonious biomarker panel; and four complementary metrics (Jaccard index, overlap coefficient, colocalization percentage, and odds ratio) were applied to evaluate spatial gene–cell colocalization. Oligodendrocytes emerged as the cell population most strongly enriched for palmitoylation activity in the AD brain, supported by per-cell-type scoring, odds ratio enrichment, and an ODC-anchored 658-gene palmitoylation module. Intersection with AD-associated differentially expressed genes yielded 89 candidate genes, from which machine learning identified a seven-gene biomarker panel (ANKIB1, SPAG9, HBP1, SLC25A29, FAM171A1, ATXN10, and NDUFV1).

ANKIB1 consistently topped every layer of analysis, with the highest AUC for AD discrimination (0.813, 95% CI 0.782–0.844) and persistent spatial colocalization with ODC-enriched regions across early AD and AD. Network pharmacology nominated AHR and TRIM11 as candidate upstream regulators and catalogued compounds with reported gene interactions. These findings identify the ODC–palmitoylation axis as a previously underappreciated dimension of AD biology and position ANKIB1 as a leading candidate biomarker of ODC dysfunction in AD.

The authors would like to thank Guangzhou Zhongxin Runkang Biotechnology Co., Ltd. for technical assistance. This work was supported by the Natural Science Foundation of Guangdong Province, China (2026A1515010026, 2025A1515012719), the Key Scientific Research Platform and Project for Universities and Colleges of Guangdong Province from the Department of Education of Guangdong Province, China (2023ZDZX2005), and the Scientific Research Foundation of the Administration of Traditional Chinese Medicine of Guangdong Province, China (20251251), and the Postdoctoral Research Startup Funding of Guangzhou City (02018358). These authors contributed equally: Jinshu Liang, Zongtang Xu, Ziting Zhu.

Department of Anesthesiology, The Third Affiliated Hospital of Southern Medical University, Guangzhou, Guangdong Province, China Jinshu Liang, Simin Tang, Yingdong Deng, Xiangsheng Zhang, Hao Chen & Jun Zhou Department of Neurology, the First Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong Province, China Zongtang Xu, Ziting Zhu, Fengchu Liang & Yuwan Lin Department of Neurology, The Second Affiliated Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China Correspondence to Yuwan Lin or Jun Zhou. The authors declare no competing interests. This study used only de-identified publicly available datasets from the Gene Expression Omnibus and did not involve new human subjects or animal research; therefore, ethical approval was not required.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it.