DMT-Dens: Density-preserving manifold visualization for biological data

2026-08-18

AI Summary

为解决低维嵌入方法在生物数据中导致密度失真的问题，DMT-Dens通过优化基于Pearson相关性的损失函数来保持样本密度，同时保持良好的标签可分离性。

Abstract

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.

Problem

• low-dimensional embeddings
• cell-state heterogeneity
• density preservation
• single-cell data
• biological data

Innovation

• density preservation
• latent-token Transformer encoder
• rank-based manifold alignment
• hard-pair aggregation