Simulation-based benchmarking of isoform quantification in single-cell RNA-seq.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) offers potential for precise isoform quantification by removing cellular heterogeneity.
• Established best practices for isoform quantification using scRNA-seq data are lacking.
• Isoform expression patterns in individual cells remain largely uncharacterized.

Purpose of the Study:

• To benchmark the performance of five popular isoform quantification tools using scRNA-seq data.
• To evaluate the accuracy of these tools on simulated and real scRNA-seq datasets.
• To gain biological insights into isoform expression variability at the single-cell level.

Main Methods:

• Simulation of scRNA-seq data based on SMARTer and SMART-seq2 protocols.
• Benchmarking of five widely used isoform quantification tools.
• Comparative analysis of quantification performance between simulated and real scRNA-seq data.
• Investigation of isoform expression patterns in bulk versus single-cell RNA sequencing.

Main Results:

• Most tested isoform quantification tools performed well on simulated scRNA-seq data.
• Performance degradation compared to bulk RNA sequencing was minimal.
• Analysis of real data revealed that genes expressing two isoforms in bulk RNA-seq typically express only one or no isoforms in individual cells.

Conclusions:

• Current isoform quantification tools are largely suitable for scRNA-seq applications.
• scRNA-seq reveals significant differences in isoform usage compared to bulk RNA-seq, highlighting cell-to-cell variability.
• Further research is needed to understand the biological implications of single-cell isoform expression patterns.