OCAT - One Cell At A Time

OCAT provides a fast and memory-efficient framework for analyzing and integrating large-scale scRNA-seq data.

➕ Method

OCAT constructs sparse representation of cell features through ghost cells in the datasets. These ghost cells serve as bridges to inform on cell-cell similarity between the original cells. With the sparse features extracted, OCAT provides an efficient framework for cell type clustering and dataset integration that achieves state-of-the-art performance.

📐 Requirements and Installation

• Linux/Unix
• Python 3.7

Install OCAT package from PyPI. Pre-installation of Numpy and Cython required.

$ pip install numpy
$ pip install OCAT

➕ Tutorials

• Clustering and Differential Gene Analysis of Mouse Brain scRNA-seq Data (Zeisel et al. 2015)
• Integration of 5 Human Pancreatic scRNA-seq Datasets
• Clustering of Spatial scRNA-seq Data
• Trajectory and pseudotime inference using HSMM dataset
• Cell Inference of new incoming data based on reference dataset