Overview¶
Publications¶
Identification of hematological cell types from heterogeneous single cell RNA-seq data.
Polled Digital Cell Sorter (p-DCS): Automatic identification of hematological cell types from single cell RNA-sequencing clusters. Sergii Domanskyi, Anthony Szedlak, Nathaniel T Hawkins, Jiayin Wang, Giovanni Paternostro & Carlo Piermarocchi, BMC Bioinformatics volume 20, Article number: 369 (2019) https://doi.org/10.1186/s12859-019-2951-x
Description of the package functionality¶
The main class of DigitalCellSorter. The class includes tools for:
Pre-preprocessing of single cell RNA sequencing data
Quality control
Batch effects correction
Cells anomaly score evaluation
Dimensionality reduction
Clustering
Annotation of cell types
Vizualization
Post-processing
Versions change log¶
- 1.3.7
Added a function to import data from kallisto-bustools and cellranger
Updated documentation
- 1.3.6
Added quick-demo materials
- 1.3.5
Miscellaneous code improvements and bug fixes
- 1.3.4.0-1.3.4.11
Integrated plotly offline figure saving (when orca is unavailable)
Added Quality Control pre-cut
- 1.3.2
Added Hopfield landscape visuzlization capability
Added network of underlying biological gene-gene interaction to the Hopfield annotaiton scheme
- 1.3.1
Minor API modifications
- 1.3.0
Modified pDCS algorithm for cell type identification to account for markers that should not be expressed in a given cell type (negative markers)
Modified pDCS celltype/marker matrix normalization
Modified pDCS algorithm account for low quality scores
Added Hopfield classifier for cell type annotation
Added ratio method for cell type annotation
Added options for consensus cell type annotation
Added cell markers pie summary function and plot
Added t-test for individual gene plot
Added several new user functions, for efficient and flexible extraction of cells, genes, clusters, etc.
Added anomaly score calculation and visualization
Refactored function for extraction of new markers based on cell type annotations to separate it from function process() of class DigitalCellSorter
Optimized implementation (for higher performance) of various function of this package
Detailed visualization functions API
Incorporated different clustering methods in addition to the widely-utilized hierarchical clustering
Incorporated several types of high-dimensional data projection methods, such as efficient t-SNE, UMAP and simple PCA components.
Extended options for input data format
Included a set of functions to load data from Human Cell Atlas (HCA) and prepare it for processing
- 1.2.3
API updates, documentation updates
- 1.2.1
Minor updates, reshaped DigitalCellSorter into a stand-alone package
- 1.2.0
More features, better runtime efficiency
- 1.1
Updated method, signature matrices
- 1.0
First Release