Effects of Sample Size on Plant Single-Cell RNA Profiling

Area of Science:

• Plant biology
• Genomics
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) enables detailed molecular analysis of individual plant cells.
• Understanding cell developmental processes in plants requires robust scRNA-seq data.
• The impact of sample size on scRNA-seq outcomes in plants is not well-defined.

Purpose of the Study:

• To evaluate the effect of varying cell numbers on single-cell transcriptome analysis in plants.
• To determine optimal sample sizes for key scRNA-seq metrics in *Arabidopsis thaliana* root cells.
• To provide guidance for sample size selection in plant scRNA-seq studies.

Main Methods:

• Integration of ~57,000 *Arabidopsis thaliana* root cells from five published scRNA-seq studies.
• Subsampling of varying cell numbers from the integrated dataset for analysis.
• Assessment of principal component significance, cell clustering reliability, differentially expressed gene identification, and pseudotime estimation.

Main Results:

• Optimal principal component analysis was achieved with 20,000-30,000 cells.
• Reliable cell clustering was obtained with approximately 20,000 cells, with minimal gains from larger sample sizes.
• 96% of differentially expressed genes were identified using up to 20,000 cells.
• Stable pseudotime estimation was possible with as few as 5,000 cells.

Conclusions:

• Sample size significantly influences the outcomes of plant scRNA-seq analyses.
• Approximately 20,000 cells are recommended for robust cell clustering and gene identification in plant scRNA-seq.
• These findings offer a practical guide for optimizing sample size in plant single-cell transcriptomics research.