Integrating single-cell RNA-Seq and machine learning to dissect a novel Palmitoylation-related prognostic signature of glioblastoma - Research paper

Sep 30, 2026

Glioblastoma (GBM) is an aggressive, heterogeneous brain tumor with poor prognosis, and palmitoylation is implicated in its development, yet its molecular mechanisms and prognostic value remain unclear. This study uses single-cell scoring methods and multiple machine-learning strategies to derive an AUCell representative palmitoylation score and construct a risk model. Immune infiltration, immunotherapy assessment, drug screening, and molecular docking further evaluate therapeutic relevance, identifying macrophage-associated palmitoylation features and ZDHHC2 as a key target.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Abstract

Background

Methods

Results

Conclusion

Background

Clinical and biological context of GBM

Role of single-cell sequencing and machine learning

Palmitoylation as a regulatory modification

Zhang et al. BMC Neurology (2026) 26:20 [https://doi.org/10.1186/s12883-025-04551-4](https://doi.org/10.1186/s12883-025-04551-4) BMC Neurology RESEARCH Open Access Integrating single-cell RNA-Seq and machine learning to dissect a novel Palmitoylationrelated prognostic signature of glioblastoma Zhu Zhang 1†, Haojie Zheng2†, Chunli Yang3 and Zhiying Lin3* Abstract Background Glioblastoma (GBM) represents a profoundly aggressive and heterogeneous brain neoplasm linked to a bleak prognosis. Palmitoylation plays a key role in the development and progression of GBM, but its molecular mechanism and prognostic significance in GBM are still not fully understood. This study aims to explore the prognostic biomarkers of GBM based on palmitoylation-related genes. Methods Eight scoring methods, including AUCell, UCell, singscore, ssGSEA, JASMINE, VAM, scSE, and viper, were used to score each sample. In addition, the palmitoylation score calculated by the AUCell algorithm is selected as the representative. In order to screen the genes related to GBM survival and build a risk prognosis model, 101 algorithms constructed by 10 kinds of machine learning are arranged and combined for variable screening and model building, and then immune infiltration and immunotherapy evaluation, drug screening, and molecular docking are carried out. Results We observed that macrophages in GBM cell types have the highest palmitoylation score. The secondary dimensionality reduction clustering of macrophages showed that the palmitoylation of PLCG2 + macrophages was significantly higher than that of other subtypes, and three core prognostic genes (ZDHHC2, ZDHHC4, ZDHHC20) were screened out by machine learning. A higher risk score is significantly related to worse clinical status and most immune labels. Among them, ZDHHC2 was significantly up-regulated in GBM in several verification groups. Molecular docking found that quercetin was the best targeted drug for ZDHHC2 . Conclusion This study revealed for the first time the heterogeneity of palmitoylation at the GBM single-cell level. The identification of ZDHHC2, ZDHHC4, and ZDHHC20 as key regulators of palmitoylation in GBM emphasized their potential as biomarkers and therapeutic targets. Keywords Glioblastoma, Machine learning algorithms, Single-cell sequencing data, Palmitoylation, Molecular docking †Zhu Zhang and Haojie Zheng contributed equally to this work. *Correspondence: Zhiying Lin lzy1 [191528410@126.com](191528410@126.com) 1Department of Neurosurgery, Zhongshan City People’s Hospital, Zhongshan, China 2The Eighth Affiliated Hospital, Southern Medical University (The First People’s Hospital of Shunde, Foshan),, Shunde, China 3Jiangxi Provincial People’s Hospital, The First Affiliated Hospital of Nanchang Medical College, No.92 Aiguo Road, Nanchang, Jiangxi Province 330006, China © The Author(s) 2025. 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. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit [http://creati](http://creati)[vecommons.org/l](vecommons.org/l)icenses/by-nc-nd/4.0/. Zhang et al. BMC Neurology (2026)

FAQs

What question does the study address about glioblastoma?

The study investigates palmitoylation-related genes and how they can generate prognostic biomarkers for glioblastoma, given that the underlying mechanisms and prognostic significance are not fully understood.

How is palmitoylation quantified and used for prognostic modeling?

Samples are scored using eight methods, and the palmitoylation score from the AUCell algorithm is selected as the representative. Machine-learning algorithms are then used to screen genes and build a risk prognosis model.

Which key prognostic genes and immune-cell findings are reported?

Macrophage cell types show the highest palmitoylation score, and PLCG2+ macrophages exhibit significantly higher palmitoylation than other subtypes. Machine learning identifies ZDHHC2, ZDHHC4, and ZDHHC20 as core prognostic genes.

What therapeutic relevance is suggested by molecular docking?

Molecular docking indicates that quercetin is the best targeted drug for ZDHHC2, supporting its potential as a therapeutic target alongside biomarker value.

Sep 30, 2026

Upload by Rainbow Cat