scMMT (single-cell multi-modal data and multi-task learning tool) is a powerful deep learning computational tool designed for the analysis of CITE-seq and scRNA-seq data. It offers various functionalities such as cell annotation, protein expression prediction, and low-dimensional embedding. With scMMT, researchers can efficiently explore and interpret complex single-cell datasets, enabling deeper insights into cellular heterogeneity and intercellular interactions.
conda create -n scMMT python=3.10
conda activate scMMTpip install scMMTAlternatively, you can also install the package directly from GitHub via
pip install git+https://github.com/SongqiZhou/scMMT.git(1) Seurat 4 human peripheral blood mononuclear cells (GEO: GSE164378).
(2) H1N1 influenza PBMC dataset (https://doi.org/10.35092/yhjc.c.4753772).
(3) COVID dataset(https://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-10026/).
(4) Simulation dataset (https://github.com/SongqiZhou/scMMT/releases/tag/scMMT).
The University of Pennsylvania has put these data sets together for the convenience of downloading. Download Here. The reference github link is: https://github.com/jlakkis/sciPENN_codes
- Python >= 3.10
- torch >= 2.0.0
- scanpy >= 1.9.3
- scikit-learn >= 1.2.2
- scikit-learn-intelex >= 2023.1.1
- pandas >= 2.0.1
- numpy >= 1.24.3
- scipy >= 1.10.1
- tqdm >= 4.65.0
- anndata >= 0.9.1