Scholarlink

<p>Single-cell RNA sequencing (scRNA-seq) is revolutionizing neuroimmune research, yet a critical bottleneck persists: the dissociation process introduces pervasive stress-altered transcription (SAT) that obscures genuine in vivo biology. Here, we present a comprehensive two-pronged strategy to eliminate these artifacts in key neuroimmune niches. First, we establish a benchmark by applying an optimized experimental protocol that minimizes dissociation stress. Comparative analysis against conventional methods allows us to map the landscape of SAT artifacts across cell types in the brain and skull bone marrow (SBM), defining distinct, tissue-specific signatures. We then employ random forest-based approach to refine these signatures into powerful SAT gene panels, creating a computational tool for the sensitive detection and removal of artifactual signals. Our experimental approach achieves a ~ 95% reduction in these artifacts, while computational approach alone also has a 60-64% reduction that enables the retrospective correction of published data. This work provides a robust framework to ensure single-cell transcriptomics faithfully captures biological reality, empowering more precise discoveries in neuroimmunology.</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

A two-pronged strategy eliminates dissociation artifacts for high-fidelity neuroimmune single-cell transcriptomics

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月6日

查看全文
  • Yuan Yan,
  • Bo Tang,
  • Pu Liu,
  • Dongmei You,
  • Yi Lv,
  • Jinmeng Yi,
  • Yuzhang Wu,
  • Yiguo Qiu

摘要

Single-cell RNA sequencing (scRNA-seq) is revolutionizing neuroimmune research, yet a critical bottleneck persists: the dissociation process introduces pervasive stress-altered transcription (SAT) that obscures genuine in vivo biology. Here, we present a comprehensive two-pronged strategy to eliminate these artifacts in key neuroimmune niches. First, we establish a benchmark by applying an optimized experimental protocol that minimizes dissociation stress. Comparative analysis against conventional methods allows us to map the landscape of SAT artifacts across cell types in the brain and skull bone marrow (SBM), defining distinct, tissue-specific signatures. We then employ random forest-based approach to refine these signatures into powerful SAT gene panels, creating a computational tool for the sensitive detection and removal of artifactual signals. Our experimental approach achieves a ~ 95% reduction in these artifacts, while computational approach alone also has a 60-64% reduction that enables the retrospective correction of published data. This work provides a robust framework to ensure single-cell transcriptomics faithfully captures biological reality, empowering more precise discoveries in neuroimmunology.