UF Health Cancer Center Bioinformatics Hub
Experimental Design
scRNA-seq

scRNA-seq

Experimental Design Considerations

Note: This section covers standard droplet-based single-cell RNA-seq protocols. For spatial transcriptomics approaches, see Spatial RNA-seq and 10X Visium.

Replication Strategy

Plan for at least 2-3 biological replicates per condition when possible. Balance sample size with cell number - fewer samples with more cells per sample may be preferable for rare cell type detection, while more samples with fewer cells provide better population-level inference.

Sample Preparation

  • Cell viability: Maintain >80% viability throughout processing. Dead cells can create ambient RNA contamination
  • Single-cell suspension: Ensure complete dissociation without excessive stress. Optimize dissociation protocols for your tissue type
  • Processing time: Minimize time from dissociation to capture (<30 minutes) to reduce stress-induced gene expression changes
  • Cell concentration: Target 700-1,200 cells/μL for optimal capture efficiency

Platform Considerations

  • 10X Genomics: Most common platform, good for large cell numbers (500-10,000 cells)
  • Smart-seq protocols: Better gene coverage per cell, suitable for smaller cell numbers with deeper profiling
  • Cell capture targets: Plan for 3,000-10,000 cells per sample, accounting for ~65% capture efficiency

Quality Control Expectations

  • Genes per cell: Expect 1,000-4,000 detected genes per cell for mammalian samples
  • UMI counts: Target 1,000-10,000 UMIs per cell depending on cell type and platform
  • Mitochondrial gene expression: <20% for healthy cells (higher may indicate cell stress/death)
  • Doublet rates: Expect 0.4-1.6% per 1,000 cells captured

Batch Effects and Experimental Design

  • Multiplexing: Consider cell hashing or genetic multiplexing to reduce batch effects
  • Processing batches: Balance conditions across processing days when multiple batches are necessary
  • Ambient RNA: Account for background contamination, especially important for tissue samples

ENCODE Guidelines

While ENCODE standards for bulk RNA-seq provide general guidance, single-cell specific considerations include the Human Cell Atlas guidelines for standardized sample processing and metadata collection.

Data Processing Workflow

[Link to your analysis pipeline documentation]