Welcome to CellMirror’s documentation !

Graphical abstract

Highlights

• CellMirror enbales learning a biologically meaningful common feature space

• CellMirror contributes to learning interpretable features

• CellMirror facilitates detecting finer domains from ST data missed by competing methods

• CellMirror is robust to decipher cell populations from ST data by unpaired scRNA-seq data

Install CellMirror

• Installation
  • Installation from Wheels

Get Started

• Introduction

Tutorials

• Alignment of Tumors and Cell Lines
  • Step0: Loading packages (Python)
  • Step1: Reading and preprocessing data (Python)
  • Step2: Training cLDVAE model (Python)
  • Step3: Saving cLDVAE outputs (Python)
  • Step4: Implementing MNN on the processed data (Python)
  • Step5: Visualization (R)
• Biological Interpretability
  • Kernal density plot of features and related genes (Python)
  • Gene set enrichment analysis for extracted features (R)
  • Scatter plot between salient features and tumor purity (R)
• Integrating ST and scRNA-seq Data
  • Step0: Loading packages (Python)
  • Step1: Reading and preprocessing data (Python)
  • Step2: Training cLDVAE model (Python)
  • Step3: Saving cLDVAE outputs (Python)
  • Step4: Implementing MNN on the processed data (Python)
  • Step5: Saving results for further visualization (Python)
  • Step6: Visualization (R)

Credits

• Credits

Citation

• Xia J, Cui J, Huang Z, Zhang S, Yao F, Zhang Y, Zuo C. CellMirror: Deciphering Cell Populations from Spatial Transcriptomics Data by Interpretable Contrastive Learning. 2023 IEEE International Conference on Medical Artificial Intelligence (MedAI), Beijing, China, 2023, pp. 165-176, doi: 10.1109/MedAI59581.2023.00029.