Scholarlink

Single-cell data analysis (scRNA-seq) is massive in size, because of which dimensionality reduction becomes unavoidable. Over the years, there have been many methods for dimensionality reduction proposed for scRNA-seq data. Dimensionality reduction directly impacts downstream analysis, including normalisation and clustering, among others. To date, there are very few comparison studies on the effectiveness of different dimensionality reduction methods on scRNA-seq data. Therefore, this paper serves as an empirical study on the different dimensionality reduction methods and their effectiveness in terms of measurements of accuracy, robustness and lineage construction. We compare 13 different dimensionality reduction methods on five datasets carefully chosen across a broad range of species and sample sizes. Among the methods, SC3 and scScope were found to perform best in effectiveness.

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

A Review of Dimensionality Reduction Methods on Single-Cell RNA-Seq (scRNA-Seq) Data

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Reema Joshi,
  • Rosy Sarmah

摘要

Single-cell data analysis (scRNA-seq) is massive in size, because of which dimensionality reduction becomes unavoidable. Over the years, there have been many methods for dimensionality reduction proposed for scRNA-seq data. Dimensionality reduction directly impacts downstream analysis, including normalisation and clustering, among others. To date, there are very few comparison studies on the effectiveness of different dimensionality reduction methods on scRNA-seq data. Therefore, this paper serves as an empirical study on the different dimensionality reduction methods and their effectiveness in terms of measurements of accuracy, robustness and lineage construction. We compare 13 different dimensionality reduction methods on five datasets carefully chosen across a broad range of species and sample sizes. Among the methods, SC3 and scScope were found to perform best in effectiveness.