ScQCenrich enables multi-metric quality control for single-cell RNA sequencing.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) is sensitive to experimental artifacts like dissociation stress.
• Current quality control (QC) methods often use fixed thresholds, leading to over-filtering or under-filtering of cells.
• There is a need for more robust and interpretable QC frameworks for scRNA-seq data.

Purpose of the Study:

• To introduce scQCenrich, a novel multi-metric QC framework for whole-cell scRNA-seq.
• To improve the accuracy and interpretability of QC in scRNA-seq analysis.
• To reduce over-filtering while preserving biologically relevant cell populations.

Main Methods:

• Developed scQCenrich, integrating canonical QC metrics with intronic fraction, MALAT1 enrichment, dissociation-stress features, and splice-aware information.
• Applied scQCenrich to diverse datasets including mouse brain, heart, and lung cancer.
• Compared scQCenrich performance against conventional and model-based QC methods.

Main Results:

• scQCenrich demonstrated reduced over-filtering compared to existing methods across multiple datasets.
• The framework successfully preserved coherent cell populations, including neuronal, erythroid, cardiomyocyte, and malignant cells.
• scQCenrich proved conservative on high-quality peripheral blood mononuclear cell data.
• Automated reports provided transparent links between QC calls, cluster metrics, marker genes, and functional enrichment.

Conclusions:

• scQCenrich offers a transparent, reproducible, and interpretable framework for scRNA-seq QC.
• The multi-metric approach enhances the reliability of cell quality assessment in scRNA-seq.
• This framework facilitates more accurate downstream analyses by improving cell data integrity.