cESFW is a feature correlation software based on the principles of Entropy Sorting.
The theory underpinning cESFW is outlined in the following publications:
Branching topology of the human embryo transcriptome revealed by entropy sort feature weighting
In Branching topology of the human embryo transcriptome revealed by entropy sort feature weighting, cESFW is used as part of a feature selection workflow to identify genes that are informative of cellular identity in single cell RNA sequeencing (scRNA-seq) data.
In Branching topology of the human embryo transcriptome revealed by entropy sort feature weighting we provide details for usage of cESFW.
The cESFW algorithm takes a 2D matrix as an input where the rows are samples and the columns are features. For scRNA-seq, the rows are cells and the columns are genes. Each feature in the 2D matrix should be scaled so that the maximum value is 1 and the minimum value is 0.
cESFW will then output an Entropy Sort Score (ESS) pairwise correlation matrix, and an Error Potential (EP) pairwise correlation significance matrix.
To find example workflows/vignettes for using cESFW as a feature selection algorithm, please go to our accompanying cESFW_Embryo_Topology_Paper repository.
- Retreive the ripository with:
git clone https://github.com/aradley/cESFW.git - Navigate to the directory where the clone was downloaded to, for example:
cd cESFW/ - Run the following on the command line:
python setup.py install