scDNM-VAE 通过带符号的树突门控实现对单细胞 RNA 测序数据的可直接检查的深度聚类
scDNM-VAE enables directly inspectable deep clustering of single-cell RNA-seq data through signed dendritic gating
浏览论文内容
中文总结 AI 辅助
本研究提出 scDNM-VAE 框架，其结合变分自编码器与树突神经元启发头部实现可直接检查的单细胞 RNA-seq 深度聚类，在多数据集上表现良好，且决策相关信息分布于潜在空间。
中文摘要 AI 辅助
针对单细胞 RNA 测序的深度聚类模型通常通过潜在机制或基于质心的机制分配细胞，这些机制难以检查。我们引入 scDNM-VAE（单细胞树突神经元模型变分自编码器），这是一种将变分自编码器与受树突神经元启发的头部相结合的深度聚类框架。聚类分配由可学习的带符号突触权重和阈值控制：权重符号决定门控对潜在坐标的响应方向，其大小控制陡峭程度，权重-阈值对决定转换位置。因此，训练后的聚类函数可直接检查，无需拟合事后解释模型。我们在涵盖免疫细胞、皮质细胞、心肌细胞和造血细胞的四个数据集上，将 scDNM-VAE 与 scVI 后接 KMeans 以及 MLP-DEC 消融模型进行基准测试。scDNM-VAE 在 PBMC3k 上的表现优于 scVI，在 Human Heart Cell Atlas 和 Paul15 上表现相当，在 Zeisel 上表现稍差，但产生了生物学上一致的标记基因特征。消融每个聚类的三个最高幅度突触维度，在所有数据集上导致的细胞重分配数量大于随机维度消融的情况，但幅度较小，且在 Zeisel 上可忽略不计。这些结果表明，带符号的树突门控支持具有参数可检查决策函数的竞争性聚类，同时表明与决策相关的信息分布在潜在空间中。
英文摘要
Deep clustering models for single-cell RNA sequencing often assign cells through latent or centroid-based mechanisms that are difficult to inspect. We introduce scDNM-VAE (single-cell Dendritic Neuron Model Variational Autoencoder), a deep clustering framework that combines a variational autoencoder with a dendritic neuron-inspired head. Cluster assignments are governed by learnable signed synaptic weights and thresholds: the weight sign determines the direction of a gate's response to a latent coordinate, its magnitude controls steepness, and the weight-threshold pair determines the transition location. The trained clustering function can therefore be inspected directly without fitting a post-hoc explanation model. We benchmark scDNM-VAE on four datasets spanning immune, cortical, cardiac, and hematopoietic cells against scVI followed by KMeans and an MLP-DEC ablation. scDNM-VAE performs better than scVI on PBMC3k, comparably on the Human Heart Cell Atlas and Paul15, and worse on Zeisel, while producing biologically coherent marker-gene signatures. Ablating each cluster's three highest-magnitude synaptic dimensions causes numerically greater reassignment than random-dimension ablation across all datasets, but the margins are modest and negligible on Zeisel. These results show that signed dendritic gating supports competitive clustering with a parameter-inspectable decision function, while indicating that decision-relevant information is distributed across the latent space.
