cellity

cellity identifies and filters low-quality cells from single-cell RNA sequencing (scRNA-seq) datasets to improve the reliability of downstream analyses.

Key Features:

• Support Vector Machine (SVM) approach: Employs a support vector machine to classify and filter low-quality cells in scRNA-seq data.
• Curated feature set: Uses a curated set of over 20 biological and technical features to assess cell quality.
• Improved classification accuracy: Demonstrates over 30% improvement in classification accuracy on datasets of more than 5,000 cells, including CD4+ T cells, bone marrow dendritic cells, and mouse embryonic stem cells.

Scientific Applications:

• Immunology: Enables accurate profiling of immune cell populations such as CD4+ T cells and dendritic cells by removing low-quality cells.
• Stem cell research: Supports detailed analysis of stem cell differentiation and development by improving single-cell data quality for mouse embryonic stem cells and related samples.
• Cancer biology: Facilitates identification of tumor heterogeneity and rare cancer cell subpopulations through refined scRNA-seq quality control.

Methodology:

Leverages a support vector machine applied to the curated set of biological and technical features to distinguish high-quality from low-quality cells.