A self-contained demo of Cyclum, in Jupyter notebook, is available in mESC.ipynb. It works with the mESC dataset (Computational analysis of cell-to-cell heterogeneity in single-cell RNA-sequencing data reveals hidden subpopulations of cells). It shows shows how to start with an expression matrix, then decide the optimal dimensionality, and finally calculate the circular pseudotime. On a computer equipped with an Cuda-ready GPU, it will take about 20 minutes to run. If only CPU is available, it may take around 1 hour.
The analysis of the result was done in R, and a demo is available in mesc_analysis.Rmd. Running it will take around five seconds. The expected result is available in "mesc_analysis.nb.html". It relies on cyclum-pseudotime.h5, the result of the Jupyter notebook.
The code is on Debian GNU/Linux 10 (buster) with both CPU and GPU. The code should run on most mainstream systems (Linux, Mac, Windows) supporting Tensorflow.
| Software | Version |
|---|---|
| python | 3.7.4 |
| keras | 2.2.4 |
| tensorflow | 1.14.0 |
| numpy | 1.16.5 |
| pandas | 0.25.2 |
| scikit-learn | 0.21.3 |
| h5py | 2.9.0 |
| hdf5 | 1.10.4 |
| jupyter | 1.0.0 |
We recommend Miniconda to manage the packages. The code should work on packages of newer versions, but in case it fails, you can return to the specific version by, for example, conda install python=3.7.4.
No installation is needed.
More instructions is availabe through our github repository: https://github.com/KChen-lab/Cyclum.