Scholarlink

<p>Dimensionality reduction is routinely applied to single-cell transcriptomic data to improve interpretability, remove noise and redundancy, and enable visualization. Most existing methods aim at preserving the most prominent data properties, which can lead to omission of rare but important signals. Here we propose a novel framework, SAKURA, that uses knowledge-derived genes of interest to guide dimensionality reduction, which can help cluster rare cells and separate highly similar cell subpopulations. We demonstrate the utility of our framework in identifying endocrine cell subtypes in the pancreatic islet, highly similar hematopoietic subpopulations, and rare senescent cells.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

SAKURA: a knowledge-guided approach to recovering important, rare signals from single-cell data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月4日

查看全文
  • Zhenghao Zhang,
  • Jiamin Chen,
  • Haoran Wu,
  • Kelly Yichen Li,
  • Peter D. Adams,
  • Pamela Itkin-Ansari,
  • Kevin Y. Yip

摘要

Dimensionality reduction is routinely applied to single-cell transcriptomic data to improve interpretability, remove noise and redundancy, and enable visualization. Most existing methods aim at preserving the most prominent data properties, which can lead to omission of rare but important signals. Here we propose a novel framework, SAKURA, that uses knowledge-derived genes of interest to guide dimensionality reduction, which can help cluster rare cells and separate highly similar cell subpopulations. We demonstrate the utility of our framework in identifying endocrine cell subtypes in the pancreatic islet, highly similar hematopoietic subpopulations, and rare senescent cells.