No project description provided
Project description
SpaTranslator
- SpaTranslator is a graph-based generative model for cross-modality prediction in spatial omics data. It enables the in silico generation of missing omics modalities—such as spatial ATAC-seq, RNA-seq, and histone modification profiles—across paired or unpaired tissue sections, by leveraging spatial neighborhood information and adversarial learning.
🧠 Highlights
- 🔄 Cross-modality generation between spatial transcriptomics and epigenomics/proteomics
- 📈 Superior performance on 12 benchmark datasets over state-of-the-art methods (multiDGD, JAMIE, scButterfly, scPair)
- 🧬 Supports multiple data types: RNA, ATAC, histone marks (H3K27me3, H3K4me3), protein
- 📍 Graph Neural Network integration for spatial structure modeling
- 🧪 Flexible in intra- and inter-dataset prediction tasks
📂 Datasets
SpaTranslator is benchmarked on:
- Spatial-ATAC-RNA-seq (mouse brain)
- MISAR-seq (mouse brain at stages E11, E13, E15, E18)
- 10x Visium RNA-Protein (human tonsil&lymph node)
Detailed documentation is written in the jupyter notebook
🚀 Installation Guide
Install with pip
conda create -n SpaTranslator python==3.9
pip install spatranslator
Install PyTorch Geometric and Related Dependencies It's crucial to install versions of these packages that match your PyTorch and CUDA versions: https://pytorch-geometric.readthedocs.io/en/latest/install/installation.html#installation-from-source
The following command installs pyg_lib, torch_scatter, torch_sparse, torch_cluster, and torch_spline_conv. Note: The URL https://data.pyg.org/whl/torch-2.6.0+cu124.html specifies wheels for PyTorch 2.6.0 and CUDA 12.4. Adjust ${TORCH} and ${CUDA} in the URL below to match your installed PyTorch version and CUDA version. You can find the correct URL on the PyTorch Geometric installation page.
# Example for PyTorch 2.6.0 and CUDA 12.4
# YOU MUST ADJUST THE URL BASED ON YOUR PYTORCH AND CUDA VERSION
pip install torch_geometric
pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.6.0+cu124.html
📄 License
SpaTranslator is released under the MIT License.
You are free to use, modify, and distribute this software for academic and non-commercial purposes with proper attribution.
📬 Contact
If you have questions, feature requests, or would like to contribute:
- 💡 Issues: Please submit via the GitHub Issues Page
- ✉️ Email:
maosheng@westlake.edu.cn/donghongyu@westlake.edu.cn
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file spatranslator-1.0.8.tar.gz.
File metadata
- Download URL: spatranslator-1.0.8.tar.gz
- Upload date:
- Size: 28.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.9.21
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
b21b50686f17fe677643d9230f18d8f5f38f431645ef172d915aa52ae96a5c42
|
|
| MD5 |
ecf5dba7e52a6cb5a4d98706af6dacee
|
|
| BLAKE2b-256 |
35da3e1367af8d8f8369e31ea5122f6a59f08dc45e391bcaad32b3ac7c2c8a70
|
File details
Details for the file spatranslator-1.0.8-py3-none-any.whl.
File metadata
- Download URL: spatranslator-1.0.8-py3-none-any.whl
- Upload date:
- Size: 28.5 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.9.21
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
32540fe841bd8c14987688487cc6ef83717ed04b34cac1f40058e584fe78a839
|
|
| MD5 |
9165467ce164f8d3393d5d26f42e22f6
|
|
| BLAKE2b-256 |
c1546a2cddfad4847c282ac5f5299bd8e759b5295b0e50e06c081d2c96b3686d
|