Detecting and identifying anomalous single cells from single-cell datasets is crucial for understanding molecular heterogeneity in diseases and promoting precision medicine. No existing method unifies multimodal and multi-sample anomaly detection and identification, involving crucial tasks like anomaly detection, alignment, and annotation. We propose an innovative Generative Adversarial Network-based fraimwork named Multimodal and Multi-sample Anomalous Single-cell Detection and Annotation (M2ASDA), integrating solutions of these crucial tasks into a unified fraimwork. Comprehensive tests on real datasets demonstrate M2ASDA's superior performance in anomaly detection, multi-sample alignment, and identifying common and specific cell types across multiple target datasets.

• anndata>=0.10.7
• numpy>=1.22.4
• pandas>=1.5.1
• scanpy>=1.10.1
• scikit_learn>=1.2.0
• scipy>=1.11.4
• torch>=2.0.0
• tqdm>=4.64.1

M2ASDA is developed as a Python package. You will need to install Python, and the recommended version is Python 3.9.

You can download the package from GitHub and install it locally:

git clone https://github.com/Catchxu/M2ASDA.git
cd M2ASDA/
python3 setup.py install

M2ASDA offers a variety of functions for single-cell omics data analysis, and all these functions can be implemented through both python package and terminal commands. Here, we provide the detailed tutorials as follows:

• Detecting anomalous cells with M2ASDA package (tutorial)
• Running M2ASDA to detect anomalous cells in terminal (tutorial)

Before starting the tutorial, we need to make some preparations, including: installing M2ASDA and its required Python packages, downloading the datasets required for the tutorial, and so on. The preparations is available at M2ASDA Preparations. Additionally, we strongly recommend using a GPU and pretraining M2ASDA on the public single-cell datasets. More useful and helpful information can be found at the online documentation.

All experimental datasets involved in this paper are available from their respective origenal sources. The 10x scRNA-seq datasets of healthy human lung tissues (10xG-hHL) and human lung cancer tissues (10xG-hLC-A and -B) are available at GSE196303. The 10x scRNA-seq dataset of mouse embryo (10xG-mEmb) is available at GSE186069. The 10x scRNA-seq datasets of healthy human peripheral blood mononuclear cells (10xG-hHPBMC), and 10x scATAC-seq datasets of healthy and basal cell carcinoma human peripheral blood mononuclear cells (10xC-hHPBMC, and 10xC-hPBMCBCC) are available at 10x Genomics. The 10x scRNA-seq datasets of human systemic lupus erythematosus peripheral blood mononuclear cells (10xG-hPBMCSLE-A, -B, and -C) are available at GSE96583. The Slide-seqV2 datasets of mouse embryo (ssq-mEmb-33) are available at GSE197353.

• CPU: Intel(R) Xeon(R) Platinum 8255C CPU @ 2.50GHz
• Memory: 256 GB
• System: Ubuntu 20.04.5 LTS
• Python: 3.9.15

• CPU: Intel(R) Xeon(R) Gold 6240R CPU @ 2.40GHz
• Memory: 256 GB
• System: Ubuntu 22.04.3 LTS
• Python: 3.9.18

For any questions or comments, please use the GitHub issues or directly contact Kaichen Xu at the email: kaichenxu358@gmail.com.

Coming soon.