Rarity: discovering rare cell populations from single-cell imaging data

• DOI: 10.1093/bioinformatics/btad750
• Publication Date: 2023-12-14T00:38:50Z
• Authors: Kaspar Märtens; Michele Bortolome...; Lucia Montorsi; Jo Spencer; Francesca Ciccarelli; Christopher Yau
• Keywords: Original Paper; Rare Cells; Bayesian; Clustering

Cell type identification plays an important role in the analysis and interpretation of single-cell data can be carried out via supervised or unsupervised clustering approaches. Supervised methods are best suited where we list all cell types their respective marker genes a priori, while algorithms look for groups cells with similar expression properties. This property permits both known unknown populations, making suitable discovery. Success is dependent on relative strength signature each group as well number cells. Rare therefore present particular challenge that magnified when they defined by differentially expressing small genes.