Codes for paper: CVQVAE: A REPRESENTATION LEARNING METHOD FOR MULTI-OMICS SINGLE CELL DATA INTEGRATION. Accepted by MLCB 2022 （Spotlight presentation).
To install our model, please use:
git clone https://github.com/HelloWorldLTY/CVQVAE.git
We refer the codes of Vector Quantized Varitional Auto-encoder (VQVAE) model from VQVAE, the codes of Crossing model from CAE, and the codes for benchmark analysis from Benchmark.
Please check our main.py file as a simple tutorial.
The download link for benchmark datasets is here.
If you intend to cite our work, please use this link:
Liu, Tianyu, Grant Greenberg, and Ilan Shomorony. "CVQVAE: A representation learning based method for multi-omics single cell data integration." Machine Learning in Computational Biology. PMLR, 2022.