Decoding epilepsy's molecular blueprint: Machine learning unravels transcriptomic subtypes and regulatory networks
Epilepsie in der molekularen Blaupause entschlüsselt: Machine Learning zeigt transcriptomische Subtypen und regulatorische Netzwerke
Epilepsia · 2026-06 · 67 (6) · 3211-3225
DOI: 10.1002/epi.70161 ↗Auf einen Blick
Bei 95 Patient:innen mit therapierefraktärer Epilepsie ergab RNA-Sequenzierung zwei transcriptomische Subtypen. Ein Random-Forest-Modell klassifizierte sie mit 96 % Genauigkeit und AUC .95.
Fragestellung
Lässt sich drug-resistant epilepsy mit einem transcriptombasierten, machine-learning-gestützten Ansatz molekular klassifizieren?
Methode
RNA-Sequenzierung von 153 resezierten Proben aus 95 Patient:innen mit DRE; unsupervised clustering, Netzwerkansatz und Machine-Learning-Klassifikation.
Ergebnisse
Zwei molekulare Subtypen unterschieden sich von traditionellen pathologischen Klassifikationen. Das Modell basierte auf vier Signalwegen: neuroactive ligand–receptor interaction, cAMP signaling, GABAergic synapse und calcium signaling.
Schlussfolgerung der Autor:innen
Die Autor:innen sehen die Subtypen und Signalwege als molekulare Merkmale der Epilepsie und als möglichen Ansatz für eine molekular basierte Klassifikation.
Kernaussagen
- 2 transcriptomische Subtypen wurden identifiziert
- Random Forest: 96 % Genauigkeit, AUC .95
- 4 Signalwege bildeten die Klassifikation
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift Epilepsia, online erschienen am 13.03.2026. Verfasst von Yanping Weng und 11 weiteren Autor:innen. Beteiligte Einrichtung: Human Phenome Institute, Zhangjiang Fudan International Innovation Center, MOE Key Laboratory of Contemporary Anthropology, Fudan University, Shangha…. Hauptthemen laut MEDLINE: Machine Learning, Transcriptome, Drug Resistant Epilepsy, Gene Regulatory Networks. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Municipal Science and TechnologyMajor Project, National Key Research and Development Program of China, Ningbo Science and Technology project.
Abstract (Originalsprache)
Objective
Drug‐resistant epilepsy (DRE) affects approximately one‐third of patients with epilepsy. The molecular heterogeneity underlying DRE remains poorly defined, largely due to limited access to resected brain tissue and substantial genetic diversity. Current classifications rely primarily on clinical symptoms and histopathological features rather than molecular mechanisms, constraining mechanistic insight and the development of targeted therapies. This study aimed to develop a transcriptome‐based, machine learning–guided framework for molecular classification of DRE.
Methods
We performed comprehensive RNA sequencing on 153 surgically resected samples from 95 patients with DRE. Two transcriptomic subtypes were identified through unsupervised clustering. We also leveraged a weighted correlation network‐based framework and systematic transcriptional signature comparison and developed a classification model using machine learning algorithms.
Results
Unsupervised clustering revealed two molecular subtypes that diverged from traditional pathological classifications, indicating an alternative transcriptomic basis for epilepsy pathogenesis. A classification model was constructed based on four key differentially regulated pathways: (1) neuroactive ligand–receptor interaction, (2) cAMP signaling, (3) γ‐aminobutyric acid (GABA)ergic synapse, and (4) calcium signaling. Among the tested algorithms, the random forest model demonstrated superior performance, achieving 96% classification accuracy with an area under the curve (AUC) of .95. Significance These molecular subtypes and their pathways could serve as key molecular hallmarks of epilepsy, offering valuable insights for developing targeted therapies. Moreover, our findings introduce a novel framework for classifying epilepsy based on its molecular nature, potentially connecting the clinical symptoms with the underlying causes more effectively.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Human Phenome Institute, Zhangjiang Fudan International Innovation Center, MOE Key Laboratory of Contemporary Anthropology, Fudan University, Shanghai, China.
- Department of Neurology, Children's Hospital of Fudan University, Shanghai, China.
- Shanghai Key Laboratory of Gene Editing and Cell Therapy for Rare Diseases, Fudan University, Shanghai, China.
- Ningbo Key Laboratory of Genomic Medicine and Birth Defects Prevention, The Affiliated Women and Children's Hospital of Ningbo University, Ningbo, China.
- Department of Neurology, Children's Hospital of Fudan University, National Children's Medical Center, Shanghai, China.
- Department of Neurosurgery, Shanghai Children's Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China.
ORCID-Profile: Yanping Weng, Rui Zhao, Yi Wang, Yu An
Förderung
- Municipal Science and TechnologyMajor Project
- National Key Research and Development Program of China
- Ningbo Science and Technology project
Zitieren
APA
Weng, Y., Ma, Y., Hou, W., Li, H., Zhou, Y., Zhao, R., Li, H., Chen, L., Ma, Y., Jin, L., Wang, Y., & An, Y. (2026). Decoding epilepsy's molecular blueprint: Machine learning unravels transcriptomic subtypes and regulatory networks. Epilepsia, 67(6), 3211-3225. https://doi.org/10.1002/epi.70161
Vancouver
Weng Y, Ma Y, Hou W, Li H, Zhou Y, Zhao R, et al. Decoding epilepsy's molecular blueprint: Machine learning unravels transcriptomic subtypes and regulatory networks. Epilepsia. 2026;67(6):3211-3225. doi:10.1002/epi.70161
BibTeX
@article{Weng2026,
author = {Weng, Yanping and Ma, Yu and Hou, Wanwan and Li, Haibo and Zhou, Yuanfeng and Zhao, Rui and Li, Hao and Chen, Lian and Ma, Yangyang and Jin, Li and Wang, Yi and An, Yu},
title = {Decoding epilepsy's molecular blueprint: Machine learning unravels transcriptomic subtypes and regulatory networks},
journal = {Epilepsia},
year = {2026},
volume = {67},
number = {6},
pages = {3211--3225},
doi = {10.1002/epi.70161}
}
EUVIMED