Deep Embedded Single-cell RNA-seq Clustering implementation with pytorch, you can find implementation of tensorflow version of DESC in https://github.com/eleozzr/desc. We will try more complicated network structure and loss function base on DESCtorch in the future!!! I will reproduce all result of this paper in my free time.
conda create -n DESCtorch python=3.6.10
conda activate DESCtorch
pip install DESCtorch
then you can run the tutorial of DESCtorch for several datasets(in run_dataset folder).
I use draw.io to plot the workflow, you can find the source file of draw.io in Figure1_reproduce folder
For compared methods, I will not compare BERMUDA and Seurat2, becuase It's difficult to install and run. I will compare DEST torch with Harmony，BBKNN， fastMNN，scVI，Seurat3，Scanorama to reproduce the result.
(Comparison of the robustness of different methods for batch definition based on the macaque retina scRNA-seq data) will be updated in my free time
(The Comparison between DESC and scVI when batch information was not provided in the analysis of macaque retina data) will be updated in my free time
(Clustering results for the pancreatic islet data generated from different scRNA-seq protocols) will be updated in my free time
(The results of PBMC data generated by Kang et al) will be updated in my free time
(The results of mouse bone marrow data generated by Paul et al) will be updated in my free time
(The estimated pseudotime plots for the human monocyte data) will be updated in my free time
Li, X., Wang, K., Lyu, Y., Pan, H., Zhang, J., Stambolian, D., ... & Li, M. (2020). Deep learning enables accurate clustering with batch effect removal in single-cell RNA-seq analysis. Nature communications, 11(1), 1-14.