GT-Mamba: A Topology-Aware Graph-State Space Model for Robust and Interpretable Epigenetic Age Prediction.

Han Wang, Hui Wang, Yanting Tong, Yuanyuan Liu, Qu Jing, Guan Ning Lin, Li Zhang

School of Information Science and Technology, Institute of Computational Biology, Northeast Normal University, 130117, Changchun, China.School of Information Science and Technology, Institute of Computational Biology, Northeast Normal University, 130117, Changchun, China. · pubmed

Paper

Abstract

Current epigenetic clocks face a trade-off between predictive accuracy and biological interpretability, often relying on dataset-specific correction to generalize across cohorts. We propose GT-Mamba, a novel architecture that integrates a Structure-Aware Graph Transformer with the Mamba state space model. This design captures CpG topological correlations and genome-wide long-range dependencies.

The following analysis is generated by AI and may contain errors. It is separate from the authors’ abstract and is not peer review.

Research summary

The paper proposes a novel deep learning architecture (GT-Mamba) to improve the accuracy and interpretability of existing epigenetic clocks by capturing topological CpG correlations. This is an incremental methodological advance in computational biology rather than a breakthrough in understanding or intervening in the root causes of aging.

Author of this paper? Send a correction.