MarkerCount: A stable, count-based cell type identifier for single-cell RNA-seq experiments.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Accurate cell type identification is crucial for single-cell RNA sequencing (scRNA-seq) data analysis.
• Existing methods may struggle to identify novel or rare cell populations.
• There is a need for robust cell identification tools that can handle diverse datasets.

Purpose of the Study:

• To introduce MarkerCount, a novel computational tool for cell type identification in scRNA-seq data.
• To evaluate MarkerCount's performance against existing marker- and reference-based identification methods.
• To demonstrate MarkerCount's capability in identifying both known and unknown cell populations.

Main Methods:

• MarkerCount utilizes the number of expressed genes (marker count) for cell identification, irrespective of expression level.
• It operates in both reference-based and marker-based modes, allowing flexibility with existing data or marker lists.
• The method refines cell type assignments by re-evaluating uncertain cells and cluster-based classifications.

Main Results:

• MarkerCount demonstrated superior performance in identifying known cell populations compared to existing methods.
• The tool effectively identified unknown cell clusters, suggesting potential for discovering novel cell subtypes.
• Comparative analyses across multiple datasets confirmed MarkerCount's enhanced accuracy and robustness.

Conclusions:

• MarkerCount offers a powerful and versatile approach for cell type identification in scRNA-seq experiments.
• Its ability to identify both known and unknown cell populations makes it valuable for comprehensive biological discovery.
• This method advances the analysis of scRNA-seq data, facilitating deeper insights into cellular heterogeneity.