CLEAR: coverage-based limiting-cell experiment analysis for RNA-seq.

Area of Science:

• Transcriptomics
• Bioinformatics
• Genomics

Background:

• Single-cell RNA sequencing (scRNA-seq) enables transcriptome profiling of limited clinical samples and rare cell populations without RNA extraction.
• Limiting-cell RNA-seq (lcRNA-seq) utilizes single-cell chemistries for low cell numbers.
• A need exists for algorithms to select robust transcripts from lcRNA-seq data for comparative analyses.

Purpose of the Study:

• To present CLEAR, a workflow for identifying reliably quantifiable transcripts in lcRNA-seq data.
• To enable robust differentially expressed gene (DEG) analysis from low-input samples.
• To improve the analysis of rare cell populations and clinical samples.

Main Methods:

• Developed the CLEAR algorithm using total RNA from chronic lymphocytic leukemia (CLL) CD5+ and CD5- cells.
• Evaluated CLEAR performance with FACS-sorted cells from mouse Dentate Gyrus (DG).
• Applied CLEAR to two public lcRNA-seq datasets.

Main Results:

• CLEAR analysis of CLL samples showed increased shared transcripts and improved principal component analysis (PCA) separation of cell types.
• In mouse DG samples, CLEAR identified and removed noisy transcripts, enhancing PCA separation of cell populations.
• CLEAR demonstrated utility in publicly available datasets and can be applied to large clinical trials with imputation.

Conclusions:

• CLEAR is valuable for transcriptome profiling and DEG analysis of lcRNA-seq data, particularly for immune cells.
• The workflow addresses a critical need in pre-processing lcRNA-seq data.
• CLEAR facilitates the analysis of rare cell populations in clinical and murine neural samples without pooling.