From groundbreaking discoveries to cutting-edge research, our researchers are empowering the next generation of female science, technology, engineering and mathematics (STEM) leaders. Get inspired by our #WomeninSTEM
DUBStepR is an algorithm for feature selection based on gene–gene correlations. A key feature of DUBStepR is the use of a stepwise approach to identify an initial core set of genes that most strongly represent coherent expression variation in the dataset. Uniquely, DUBStepR defines a novel graph-based measure of cell aggregation in the feature space (termed density index (DI)), and uses this measure to optimize the number of features.
DUBStepR requires R version >= 3.5.0.
DUBStepR is available as an R package on CRAN (https://CRAN.R-project.org/package=DUBStepR), and is well documented for easy integration into the Seurat pipeline. The source code is also freely available on GitHub (https://github.com/prabhakarlab/DUBStepR).
Ranjan, B., Sun, W., Park, J. et al. DUBStepR is a scalable correlation-based feature selection method for accurately clustering single-cell data. Nat Commun 12, 5849 (2021).https://doi.org/10.1038/s41467-021-26085-2 ;