Integrating machine learning and WGCNA for nomogram diagnostic model unveiling previously unrecognized phase separation-related molecular markers and immuno-cytotoxic pathways in Williams-Beuren syndrome / Mingyi WANG in Research in Autism, 128 (October 2025)  

Public ISBD [article]  inResearch in Autism > 128 (October 2025) . - p.202691 Titre : Integrating machine learning and WGCNA for nomogram diagnostic model unveiling previously unrecognized phase separation-related molecular markers and immuno-cytotoxic pathways in Williams-Beuren syndrome Type de document : texte imprimé Auteurs : Mingyi WANG, Auteur ; Hong WANG, Auteur ; Xiao ZHU, Auteur ; Yongmei HUANG, Auteur Article en page(s) : p.202691 Langues : Anglais (eng) Mots-clés : Williams-Beuren syndrome  Biomarker  Phase separation  Machine learning  Nomogram diagnostic model  WGCNA Index. décimale : PER Périodiques Résumé : Background Understanding the role of phase separation-related genes in Williams-Beuren Syndrome (WBS), a neurodevelopmental disorder, may provide valuable insights into its pathophysiology. This study aimed to identify potential biomarkers and functional pathways associated with WBS through integrated analyses. Methods Differentially expressed genes (DEGs) in WBS patients were identified using GEO datasets. Weighted Gene Co-Expression Network Analysis (WGCNA) was employed to explore co-expression modules, and machine learning techniques were applied to select potential biomarkers. Gene Set Enrichment Analysis (GSEA) was used to investigate functional pathways, while single-sample GSEA (ssGSEA) assessed marker gene activity. Results We identified 3519 differentially expressed genes (DEGs) in WBS samples, including 19 core phase separation-related genes. WGCNA revealed six co-expression modules, with the yellow module exhibiting the strongest correlation. Functional analysis indicated enrichment in glycolipid binding and cytoskeletal structural components. Disease ontology analysis implicated developmental and ocular disorders. ssGSEA highlighted associations with immune-related pathways. MAG and ZNF385A emerged as potential diagnostic biomarkers. Conclusion Our integrated approach, combining machine learning and WGCNA, highlights the potential of phase separation-related biomarkers and immune/cytotoxic pathways in the diagnosis of WBS. This study provides valuable insights into the development of diagnostic models for WBS and underscores the importance of investigating protein phase separation in neurodevelopmental disorders. En ligne : https://doi.org/10.1016/j.reia.2025.202691 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=570 [article] Integrating machine learning and WGCNA for nomogram diagnostic model unveiling previously unrecognized phase separation-related molecular markers and immuno-cytotoxic pathways in Williams-Beuren syndrome [texte imprimé] / Mingyi WANG, Auteur ; Hong WANG, Auteur ; Xiao ZHU, Auteur ; Yongmei HUANG, Auteur . - p.202691. Langues : Anglais (eng) in Research in Autism > 128 (October 2025) . - p.202691 Mots-clés : Williams-Beuren syndrome  Biomarker  Phase separation  Machine learning  Nomogram diagnostic model  WGCNA Index. décimale : PER Périodiques Résumé : Background Understanding the role of phase separation-related genes in Williams-Beuren Syndrome (WBS), a neurodevelopmental disorder, may provide valuable insights into its pathophysiology. This study aimed to identify potential biomarkers and functional pathways associated with WBS through integrated analyses. Methods Differentially expressed genes (DEGs) in WBS patients were identified using GEO datasets. Weighted Gene Co-Expression Network Analysis (WGCNA) was employed to explore co-expression modules, and machine learning techniques were applied to select potential biomarkers. Gene Set Enrichment Analysis (GSEA) was used to investigate functional pathways, while single-sample GSEA (ssGSEA) assessed marker gene activity. Results We identified 3519 differentially expressed genes (DEGs) in WBS samples, including 19 core phase separation-related genes. WGCNA revealed six co-expression modules, with the yellow module exhibiting the strongest correlation. Functional analysis indicated enrichment in glycolipid binding and cytoskeletal structural components. Disease ontology analysis implicated developmental and ocular disorders. ssGSEA highlighted associations with immune-related pathways. MAG and ZNF385A emerged as potential diagnostic biomarkers. Conclusion Our integrated approach, combining machine learning and WGCNA, highlights the potential of phase separation-related biomarkers and immune/cytotoxic pathways in the diagnosis of WBS. This study provides valuable insights into the development of diagnostic models for WBS and underscores the importance of investigating protein phase separation in neurodevelopmental disorders. En ligne : https://doi.org/10.1016/j.reia.2025.202691 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=570

---

1 (1 - 1 / 1) Par page : 25 50 100 200