Sparse representation learning methods for the study of spatial transcriptomics
The main learning method (dimension reduction) used: Sparse Filtering
This repository is for the study "Improved Unsupervised Representation Learning of Spatial Transcriptomic Data with Sparse Filtering"
The paper link: https://www.biorxiv.org/content/10.1101/2021.10.11.464002v2.abstract
Morphontogeny
This package has functions that are used in the analysis of the data.
https://github.com/BioProteanLabs/Morphontogeny
To install this repo using pip: pip install git+https://github.com/BioProteanLabs/Morphontogeny.git
Sparse Filtering
This is the python package for the Sparse Filtering method.
Link to the paper: https://papers.nips.cc/paper/2011/file/192fc044e74dffea144f9ac5dc9f3395-Paper.pdf
Link to the repo: https://github.com/jmetzen/sparse-filtering
To install this repo using pip: pip install sparse_fitering
Numpy, Pandas, SciKit-Learn
The notebooks for the analysis of the data are in the Notebooks folder.
The files that could be used for the analyses are in the files folder.