Skip to content

Latest commit

 

History

History
67 lines (54 loc) · 4.06 KB

README.md

File metadata and controls

67 lines (54 loc) · 4.06 KB

Overview of SCDE

The scde package implements a set of statistical methods for analyzing single-cell RNA-seq data. scde fits individual error models for single-cell RNA-seq measurements. These models can then be used for assessment of differential expression between groups of cells, as well as other types of analysis. The scde package also contains the pagoda framework which applies pathway and gene set overdispersion analysis to identify aspects of transcriptional heterogeneity among single cells.

The overall approach to the differential expression analysis is detailed in the following publication:
"Bayesian approach to single-cell differential expression analysis" (Kharchenko PV, Silberstein L, Scadden DT, Nature Methods, doi:10.1038/nmeth.2967)

The overall approach to pathways and gene set overdispersion analysis is detailed in the following publication: "Characterizing transcriptional heterogeneity through pathway and gene set overdispersion analysis" (Fan J, Salathia N, Liu R, Kaeser G, Yung Y, Herman J, Kaper F, Fan JB, Zhang K, Chun J, and Kharchenko PV, Nature Methods, doi:10.1038/nmeth.3734)

For additional installation information, tutorials, and more, please visit the SCDE website ☞

Sample analyses and images

Single cell error modeling

scde fits individual error models for single cells using counts derived from single-cell RNA-seq data to estimate drop-out and amplification biases on gene expression magnitude.

Differential expression analysis

             lb   mle     ub    ce     Z    cZ
Dppa5a    8.075 9.965 11.541 8.075 7.160 5.968
Pou5f1    5.357 7.208  9.178 5.357 7.160 5.968
Gm13242   5.672 7.681  9.768 5.672 7.159 5.968
Tdh       5.829 8.075 10.281 5.829 7.159 5.968
Ift46     5.435 7.366  9.217 5.435 7.150 5.968

scde compares groups of single cells and tests for differential expression, taking into account variability in the single cell RNA-seq data due to drop-out and amplification biases in order to identify more robustly differentially expressed genes.

Pathway and gene set overdispersion analysis

scde contains pagoda routines that characterize aspects of transcriptional heterogeneity in populations of single cells using pre-defined gene sets as well as 'de novo' gene sets derived from the data. Significant aspects are used to cluster cells into subpopulations. A graphical user interface can be deployed to interactively explore results. See examples from the PAGODA publication here. See analysis of the PBMC data from 10x Genomics here.

scde is maintained by Jean Fan of the Kharchenko Lab at the Department of Biomedical Informatics at Harvard Medical School.


Contributing

We welcome any bug reports, enhancement requests, and other contributions. To submit a bug report or enhancement request, please use the scde GitHub issues tracker. For more substantial contributions, please fork this repo, push your changes to your fork, and submit a pull request with a good commit message. For more general discussions or troubleshooting, please consult the scde Google Group.