新方法DMT-Dens能精准保留生物数据的密度分布，让稀有细胞状态更易识别。
DMT-Dens: Density-preserving manifold visualization for biological data
- 基于潜在令牌Transformer架构，结合排名对齐与硬对偶聚合
- 在生物数据集上密度保持能力优于多数现有方法，相关性提升显著
- 适合研究稀有或连续细胞状态的生物学家，尤其关注数据密度的分析
低维嵌入广泛用于探索单细胞等高维生物数据中的细胞状态异质性。尽管许多方法能保留局部邻域结构，但可能扭曲观测样本的采样密度，改变密集区与稀疏区的视觉对比，影响稀有、过渡或连续细胞状态的解读。我们提出DMT-Dens，一种基于潜在令牌Transformer编码器的参数化流形可视化方法。该模型融合基于排名的流形对齐与硬对偶聚合机制，并通过优化处理输入与二维嵌入空间中k近邻对数半径估计间的皮尔逊相关系数来实现密度保持。基准测试表明，该方法在生物数据集上表现出优异的密度保持性能，同时保持了良好的标签可分性。
原文摘要 · Abstract (English)
Motivation: Low-dimensional embeddings are widely used to explore cell-state heterogeneity in single-cell and other high-dimensional biological data. Although many methods preserve local neighborhoods, they may distort the apparent sampling density of processed observations, altering the visual contrast between dense and sparse regions and complicating the interpretation of rare, transitional, or continuous cell-state populations. Results: We present DMT-Dens, a parametric manifold-visualization method built on a latent-token Transformer encoder. The model integrates rank-based manifold alignment with hard-pair aggregation. To preserve density, it optimizes a loss based on the Pearson correlation between k-nearest-neighbor log-radius estimates in the processed input and two-dimensional embedding spaces. Benchmark evaluations demonstrate strong density preservation, particularly on biological datasets, while retaining competitive label separability. Availability: Source code, data-processing scripts, and resolved experiment configurations are available at https://github.com/Ruizhe-wang/DMT-Dens.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。