With the rapid development of single-cell RNA sequencing technology, it is possible to dissect cell-type composition at high resolution. A number of methods have been developed with the purpose to identify rare cell types. However, existing methods are still not scalable to large datasets, limiting their utility. To overcome this limitation, researchers at the Dana-Farber Cancer Institute have developed a new software package, called GiniClust3, which is an extension of GiniClust2 and significantly faster and memory-efficient than previous versions.
Using GiniClust3, it only takes about 7 h to identify both common and rare cell clusters from a dataset that contains more than one million cells. Cell type mapping and perturbation analyses show that GiniClust3 could robustly identify cell clusters.
Analysis of mouse brain dataset with more than one million cells
a An overview of the GiniClust3 pipeline. Input single-cell expression matrix is clustered based on features selected by Gini index (GiniIndexClust) and by Fano factor (FanoFactorClust), respectively. The results are then integrated using a cluster-aware, weighted consensus clustering algorithm (ConsensusClust). b UMAP visualization of the gene expression patterns based on Fano-factor (top) and Gini index (bottom) selected features, respectively. Consensus clustering results are indicated by different colors. c The proportion of rare cell cluster in entire population. d Heatmap of cell type mapping of common and rare clusters from scMCA analysis. Bar plot in the top indicates the cell number for each cluster
Taken together, these results suggest that GiniClust3 is a powerful tool to identify both common and rare cell population and can handle large dataset.
Availability – GiniCluster3 is implemented in the open-source python package and available at https://github.com/rdong08/GiniClust3.