Welcome to scProca’s documentation!

scProca is a package designed to integrate and generate single-cell proteomics from transcriptomics, implemented in PyTorch.

News

2025-9-29: Now published in IEEE Journal of Biomedical and Health Informatics. https://doi.org/10.1109/JBHI.2025.3615771 .

Introduction

Overview of scProca. (A) Schematic representation of scProca within the framework of deep generative models. scProca is capable of inferring batch-corrected, integrated latent variables from scRNA-seq and CITE-seq data, and generating the expression profiles of ADT for scRNA-seq cells. (B) The variational auto-encoder with cross-attention introduced in scProca. Cross-attention is used to incorporate CITE-seq cells as references, completing representation of scRNA-seq cells in the ADT embedding space.

UMAP visualization of the low-dimensional representations obtained by scProca on the SLN dataset, colored by batch annotation and cell types. Color by batch annotation: (A) Only SLN111-D1 serves as CITE-seq data, while the others serve as scRNA-seq data. (B) Both SLN111-D1 and SLN111-D2 serve as CITE-seq data, while both SLN208-D1 and SLN208-D2 serve as scRNA-seq data. Color by cell types: (C) Same setup as (A). (D) Same setup as (B).

Installation and Reproducibility

• Installation
• Reproducibility

Jupyter notebooks for examples

• Integrate and generate CITE-seq PBMC Datasets from scRNA-seq PBMC Datasets
• Integrate and generate CITE-seq SLN Datasets from scRNA-seq SLN Datasets
• Integrate and generate CITE-seq SLN Datasets from scRNA-seq SLN Datasets using experimental batches

Main functions and parameters

• scproca.model

Citation

@article{xiong2025scproca,
  author  = {Xiong, Jiankang and Zheng, Shuqiao and Gong, Fuzhou and Ma, Liang and Wan, Lin},
  journal = {IEEE Journal of Biomedical and Health Informatics},
  title   = {{scProca}: A Cross-Attention-Enhanced Deep Generative Model for Single-Cell Transcriptomics and Proteomics Integration and Imputation},
  year    = {2025},
  pages   = {1-11},
  keywords = {Proteomics;Proteins;Transcriptomics;RNA;Mathematical models;Biomedical measurement;Sequential analysis;Imputation;Data models;Training;Multi-omics integration;Single-cell imputation;Deep generative model;Attention mechanisms;Transcriptomics and proteomics},
  doi     = {10.1109/JBHI.2025.3615771}
}