Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya Disease

Moyamoya disease (MMD) is a rare, chronic cerebrovascular disorder of uncertain etiology. Although abnormal glucose metabolism has been implicated, the contribution of glycosylation-related genes in MMD remains elusive. In this study, we analyzed 2 transcriptome data sets (GSE189993 and GSE131293) f...

Full description

Saved in:
Bibliographic Details
Main Authors: Cunxin Tan MD, PhD, Jing Wang MD, PhD, Yanru Wang MD, Shaoqi Xu BS, Zhenyu Zhou MD, Junze Zhang MD, PhD, Shihao He MD, PhD, Ran Duan MD, PhD
Format: Article
Language:English
Published: SAGE Publishing 2025-05-01
Series:Bioinformatics and Biology Insights
Online Access:https://doi.org/10.1177/11779322251342412
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850149339437989888
author Cunxin Tan MD, PhD
Jing Wang MD, PhD
Yanru Wang MD
Shaoqi Xu BS
Zhenyu Zhou MD
Junze Zhang MD, PhD
Shihao He MD, PhD
Ran Duan MD, PhD
author_facet Cunxin Tan MD, PhD
Jing Wang MD, PhD
Yanru Wang MD
Shaoqi Xu BS
Zhenyu Zhou MD
Junze Zhang MD, PhD
Shihao He MD, PhD
Ran Duan MD, PhD
author_sort Cunxin Tan MD, PhD
collection DOAJ
description Moyamoya disease (MMD) is a rare, chronic cerebrovascular disorder of uncertain etiology. Although abnormal glucose metabolism has been implicated, the contribution of glycosylation-related genes in MMD remains elusive. In this study, we analyzed 2 transcriptome data sets (GSE189993 and GSE131293) from the Gene Expression Omnibus (GEO) database to identify 723 differentially expressed genes (DEGs) between MMD patients and controls. Intersection genes with known glycosylation-related genes underwent Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. We utilized machine learning to select key hub genes, followed by immune cell infiltration and correlation analyses. In-depth immune cell analysis indicated that both CFP and MGAT5B were closely tied to various immune components, suggesting potential crosstalk between glycosylation pathways and immune regulation. Notably, CFP was positively associated with pDCs, HLA, and CCR, whereas MGAT5B correlated with B-cells, check-points, and T helper cells but showed a negative relationship with Tregs, hinting at an immunoregulatory mechanism influencing MMD progression. Motif-TF annotation highlighted csibp_M2095 as the motif with the highest normalized enrichment score (NES: 6.57). Reverse microRNA (miRNA)-gene prediction identified 75 miRNAs regulating these focus genes, along with 126 miRNA-miRNA interconnections. Connectivity Map (Cmap) analysis revealed that Chenodeoxycholic acid, MRS-1220, Phenytoin, and Piceid were strongly negatively correlated with MMD expression profiles, suggesting potential therapeutic candidates. Enzyme-linked immunosorbent assays confirmed elevated CFP and MGAT5B and reduced PTPN11 in MMD, aligning with our bioinformatic findings. Moreover, PTPN11 knockdown in human brain microvascular endothelial cells (HBMECs) significantly enhanced tube formation, indicating a role in vascular remodeling. Collectively, these results emphasize the importance of glycosylation-related genes and immune dysregulation in MMD pathogenesis. These findings broaden our understanding of MMD’s underlying mechanisms and underscore the necessity of continued research into glycosylation-driven pathways for improved disease management.
format Article
id doaj-art-d52cacbf936d4e14b24d8514d94df423
institution OA Journals
issn 1177-9322
language English
publishDate 2025-05-01
publisher SAGE Publishing
record_format Article
series Bioinformatics and Biology Insights
spelling doaj-art-d52cacbf936d4e14b24d8514d94df4232025-08-20T02:26:58ZengSAGE PublishingBioinformatics and Biology Insights1177-93222025-05-011910.1177/11779322251342412Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya DiseaseCunxin Tan MD, PhD0Jing Wang MD, PhD1Yanru Wang MD2Shaoqi Xu BS3Zhenyu Zhou MD4Junze Zhang MD, PhD5Shihao He MD, PhD6Ran Duan MD, PhD7Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, ChinaDepartment of Neurosurgery, Peking University International Hospital, Beijing, ChinaDepartment of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, ChinaSuzhou Vocational Health College, Suzhou, ChinaDepartment of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, ChinaDepartment of Pathology, Stanford University School of Medicine, Stanford, CA, USADepartment of Neurosurgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, ChinaDepartment of Neurosurgery, Peking University International Hospital, Beijing, ChinaMoyamoya disease (MMD) is a rare, chronic cerebrovascular disorder of uncertain etiology. Although abnormal glucose metabolism has been implicated, the contribution of glycosylation-related genes in MMD remains elusive. In this study, we analyzed 2 transcriptome data sets (GSE189993 and GSE131293) from the Gene Expression Omnibus (GEO) database to identify 723 differentially expressed genes (DEGs) between MMD patients and controls. Intersection genes with known glycosylation-related genes underwent Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. We utilized machine learning to select key hub genes, followed by immune cell infiltration and correlation analyses. In-depth immune cell analysis indicated that both CFP and MGAT5B were closely tied to various immune components, suggesting potential crosstalk between glycosylation pathways and immune regulation. Notably, CFP was positively associated with pDCs, HLA, and CCR, whereas MGAT5B correlated with B-cells, check-points, and T helper cells but showed a negative relationship with Tregs, hinting at an immunoregulatory mechanism influencing MMD progression. Motif-TF annotation highlighted csibp_M2095 as the motif with the highest normalized enrichment score (NES: 6.57). Reverse microRNA (miRNA)-gene prediction identified 75 miRNAs regulating these focus genes, along with 126 miRNA-miRNA interconnections. Connectivity Map (Cmap) analysis revealed that Chenodeoxycholic acid, MRS-1220, Phenytoin, and Piceid were strongly negatively correlated with MMD expression profiles, suggesting potential therapeutic candidates. Enzyme-linked immunosorbent assays confirmed elevated CFP and MGAT5B and reduced PTPN11 in MMD, aligning with our bioinformatic findings. Moreover, PTPN11 knockdown in human brain microvascular endothelial cells (HBMECs) significantly enhanced tube formation, indicating a role in vascular remodeling. Collectively, these results emphasize the importance of glycosylation-related genes and immune dysregulation in MMD pathogenesis. These findings broaden our understanding of MMD’s underlying mechanisms and underscore the necessity of continued research into glycosylation-driven pathways for improved disease management.https://doi.org/10.1177/11779322251342412
spellingShingle Cunxin Tan MD, PhD
Jing Wang MD, PhD
Yanru Wang MD
Shaoqi Xu BS
Zhenyu Zhou MD
Junze Zhang MD, PhD
Shihao He MD, PhD
Ran Duan MD, PhD
Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya Disease
Bioinformatics and Biology Insights
title Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya Disease
title_full Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya Disease
title_fullStr Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya Disease
title_full_unstemmed Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya Disease
title_short Integrative Machine Learning Approach to Explore Glycosylation Signatures and Immune Landscape in Moyamoya Disease
title_sort integrative machine learning approach to explore glycosylation signatures and immune landscape in moyamoya disease
url https://doi.org/10.1177/11779322251342412
work_keys_str_mv AT cunxintanmdphd integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease
AT jingwangmdphd integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease
AT yanruwangmd integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease
AT shaoqixubs integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease
AT zhenyuzhoumd integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease
AT junzezhangmdphd integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease
AT shihaohemdphd integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease
AT randuanmdphd integrativemachinelearningapproachtoexploreglycosylationsignaturesandimmunelandscapeinmoyamoyadisease