🤖 AI Summary
This study addresses the significant heterogeneity of DNA methylation biomarkers for Alzheimer’s disease (AD) across tissues, which hampers reproducibility and clinical translation. To overcome this challenge, the authors propose a novel Transformer-based framework integrating convolutional and self-attention mechanisms, which jointly models both shared and tissue-specific methylation signals through CpG linear projection, tissue embedding, and covariate integration. The method consistently outperforms conventional machine learning models across six GEO datasets and an independent ADNI cohort. Furthermore, by leveraging SHAP and Grad-CAM++ for multi-level interpretability, the approach identifies methylation patterns linked to AD-relevant biological pathways—including immune receptor signaling, glycosylation, and lipid metabolism—thereby substantially enhancing both cross-tissue diagnostic performance and mechanistic insight into AD pathogenesis.
📝 Abstract
Alzheimer’s disease (AD) is a multifactorial neurodegenerative disorder characterized by progressive cognitive decline and widespread epigenetic dysregulation in the brain. DNA methylation, as a stable yet dynamic epigenetic modification, holds promise as a noninvasive biomarker for early AD detection. However, methylation signatures vary substantially across tissues and studies, limiting reproducibility and translational utility. To address these challenges, we develop MethConvTransformer, a transformer-based deep learning framework that integrates DNA methylation profiles from both brain and peripheral tissues to enable biomarker discovery. The model couples a CpG-wise linear projection with convolutional and self-attention layers to capture local and long-range dependencies among CpG sites, while incorporating subject-level covariates and tissue embeddings to disentangle shared and region-specific methylation effects. In experiments across six GEO datasets and an independent ADNI validation cohort, our model consistently outperforms conventional machine-learning baselines, achieving superior discrimination and generalization. Moreover, interpretability analyses using linear projection, SHAP, and Grad-CAM++ reveal biologically meaningful methylation patterns aligned with AD-associated pathways, including immune receptor signaling, glycosylation, lipid metabolism, and endomembrane (ER/Golgi) organization. Together, these results indicate that MethConvTransformer delivers robust, cross-tissue epigenetic biomarkers for AD while providing multi-resolution interpretability, thereby advancing reproducible methylation-based diagnostics and offering testable hypotheses on disease mechanisms.