📄 中文摘要
DNA甲基化作为表观遗传修饰，调控神经谱系的承诺，影响转录程序。单细胞亚硫酸盐测序（scBS-Seq）的最新进展揭示了在前体细胞向皮层神经元分化过程中，甲基组的异质性不断演变。然而，细胞中数十万个CpG位点的高维性以及甲基化变化的动态非线性特性，对传统统计和机器学习模型提出了重大挑战。现有方法的主要局限性包括：空间稀疏性导致每个位点的甲基化估计不可靠，以及时间动态性未能有效捕捉甲基化的演变过程。
图神经网络深度学习在神经分化中DNA甲基化动态的应用
📄 English Summary
**Graph Temporal Deep Learning for DNA Methylation Dynamics in Neuronal Differentiation**
Epigenetic modifications, particularly DNA methylation, play a crucial role in neuronal lineage commitment by modulating transcriptional programs. Recent advancements in single-cell bisulfite sequencing (scBS-Seq) have uncovered heterogeneous methylomes that evolve as progenitor cells differentiate into cortical neurons. However, the high dimensionality—hundreds of thousands of CpG sites per cell—and the dynamic, non-linear nature of methylation changes present significant challenges for conventional statistical and machine learning models. The main limitations of existing approaches include spatial sparsity, which leads to unreliable per-site methylation estimates, and temporal dynamics, which fail to effectively capture the evolution of methylation.