Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

https://github.com/ZJUFanLab/bulk2space

a spatial deconvolution method based on deep learning frameworks, which converts bulk transcriptomes into spatially resolved single-cell expression profiles
https://github.com/ZJUFanLab/bulk2space

bulk-sequencing deep-learning scrna-seq spatial-transcriptomics

Last synced: 17 days ago
JSON representation

a spatial deconvolution method based on deep learning frameworks, which converts bulk transcriptomes into spatially resolved single-cell expression profiles

Host: GitHub
URL: https://github.com/ZJUFanLab/bulk2space
Owner: ZJUFanLab
License: gpl-3.0
Created: 2021-12-31T03:58:01.000Z (over 2 years ago)
Default Branch: main
Last Pushed: 2023-03-22T05:51:52.000Z (over 1 year ago)
Last Synced: 2024-02-28T21:31:39.681Z (4 months ago)
Topics: bulk-sequencing, deep-learning, scrna-seq, spatial-transcriptomics
Language: Jupyter Notebook
Homepage:
Size: 12.1 MB
Stars: 94
Watchers: 4
Forks: 20
Open Issues: 8
Metadata Files:
- Readme: README.md
- License: LICENSE

Lists

awesome-deconvolution - Bulk2Space - cell expression profiles ([Liao et al 2022](https://www.biorxiv.org/content/10.1101/2022.01.15.476472v1)). (Methods)

README

        # Bulk2Space v1.0.0

## De novo analysis of bulk RNA-seq data at spatially resolved single-cell resolution 

### Jie Liao^†,  Jingyang Qian^†, Yin Fang^†, Zhuo Chen^†, Xiang Zhuang^†, ..., Huajun Chen\*, Xiaohui Fan*

[![python 3.8](https://img.shields.io/badge/python-3.8-brightgreen)](https://www.python.org/) [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.7134575.svg)](https://doi.org/10.5281/zenodo.7134575) [![DOI](https://img.shields.io/badge/DOI-10.1038%2Fs41467--022--34271--z-yellowgreen)](https://www.nature.com/articles/s41467-022-34271-z)

Bulk2Space is a two-step spatial deconvolution method based on deep learning frameworks, which converts bulk transcriptomes into spatially resolved single-cell expression profiles.

![Image text](images/overview.jpeg)

## Requirements and Installation

[![deep-forest 0.1.5](https://img.shields.io/badge/deep--forest-0.1.5-success)](https://pypi.org/project/deep-forest/) [![numpy 1.19.2](https://img.shields.io/badge/numpy-1.19.2-green)](https://github.com/numpy/numpy) [![pandas 1.1.3](https://img.shields.io/badge/pandas-1.1.3-yellowgreen)](https://github.com/pandas-dev/pandas) [![scikit-learn 1.0.1](https://img.shields.io/badge/scikit--learn-1.0.1-yellow)](https://github.com/scikit-learn/scikit-learn) [![scipy 1.5.2](https://img.shields.io/badge/scipy-1.5.2-orange)](https://github.com/scipy/scipy) [![scanpy 1.8.1](https://img.shields.io/badge/scanpy-1.8.1-ff69b4)](https://pypi.org/project/scanpy/) [![easydict 1.9](https://img.shields.io/badge/easydict-1.9-informational)](https://pypi.org/project/easydict/) [![tqdm 4.50.2](https://img.shields.io/badge/tqdm-4.50.2-9cf)](https://pypi.org/project/tqdm/) [![Unidecode 1.3.0](https://img.shields.io/badge/Unidecode-1.3.0-inactive)](https://pypi.org/project/Unidecode/) 

### Create and activate Python environment

For Bulk2Space, the python version need is over 3.8. If you have installed Python3.6 or Python3.7, consider installing Anaconda, and then you can create a new environment.

```

conda create -n bulk2space python=3.8

conda activate bulk2space

```

### Install pytorch

The version of pytorch should be suitable to the CUDA version of your machine. You can find the appropriate version on the [PyTorch website](https://pytorch.org/get-started/locally/).

Here is an example with CUDA11.6:

```

pip install torch --extra-index-url https://download.pytorch.org/whl/cu116

```

### Install other requirements

```

cd bulk2space-main

pip install -r requirements.txt

```

### Install Bulk2Space

```

python setup.py build

python setup.py install

```

## Quick Start

To use Bulk2Space we require five formatted `.csv` files as input (i.e. read in by pandas). We have included two test datasets 

in the [tutorial/data/example_data folder](tutorial/data/example_data) of this repository as examples to show how to use Bulk2Space. 

If you choose the spot-based data (10x Genomics, ST, or Slide-seq, etc) as spatial reference, please refer to:

* [Demonstration of Bulk2Space on demo1 dataset](tutorial/demo1.ipynb)

If you choose the image-based data (MERFISH, SeqFISH, or STARmap, etc) as spatial reference, please refer to:

* [Demonstration of Bulk2Space on demo2 dataset](tutorial/demo2.ipynb)

For more details about the format of input and the description of parameters, see the [Tutorial Handbook](tutorial/handbook.md).

 

## Tutorials

Additional step-by-step tutorials now available! Preprocessed datasets used can be downloaded from [Google Drive (PDAC)](https://drive.google.com/file/d/1xB-Gk_KLxQA320-tycJp4CFHA66zF3LE/view?usp=sharing) and [Google Drive (hypothalamus)](https://drive.google.com/file/d/1ZGstNzVX-YxofrPP8ZVmr0Zu4nd_O_bZ/view?usp=sharing).

* [Integrating spatial gene expression and histomorphology in pancreatic ductal adenocarcinoma (PDAC)](tutorial/pdac.ipynb)

* [Spatially resolved analysis of mouse hypothalamus by Bulk2Space](tutorial/hypothalamus.ipynb)

## About

Should you have any questions, please feel free to contact the co-first authors of the manuscript, Dr. Jie Liao ([email protected]), Mr. Jingyang Qian ([email protected]), Miss Yin Fang ([email protected]), Mr. Zhuo Chen ([email protected]), or Mr. Xiang Zhuang ([email protected]).

## References

Liao, J., Qian, J., Fang, Y. et al. De novo analysis of bulk RNA-seq data at spatially resolved single-cell resolution. Nat Commun 13, 6498 (2022). [https://doi.org/10.1038/s41467-022-34271-z](https://doi.org/10.1038/s41467-022-34271-z)