https://github.com/biomed-ai/sango
The official implementation for "SANGO".
https://github.com/biomed-ai/sango
bioinformatics cell-type-annotation cell-type-classification cell-type-identification sequence single-cell supervised-classification-methods
Last synced: 10 days ago
JSON representation
The official implementation for "SANGO".
- Host: GitHub
- URL: https://github.com/biomed-ai/sango
- Owner: biomed-AI
- Created: 2023-09-25T03:17:17.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-02-03T02:37:31.000Z (almost 2 years ago)
- Last Synced: 2024-05-06T00:02:48.486Z (over 1 year ago)
- Topics: bioinformatics, cell-type-annotation, cell-type-classification, cell-type-identification, sequence, single-cell, supervised-classification-methods
- Language: Jupyter Notebook
- Homepage:
- Size: 18.2 MB
- Stars: 6
- Watchers: 2
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: Readme.md
Awesome Lists containing this project
README
We propose a novel method, SANGO, for accurate single cell annotation by integrating genome sequences around the accessibility peaks within scATAC data.
# SANGO
The official implementation for "**SANGO**".
**Table of Contents**
* [Datasets](#Datasets)
* [Installation](#Installation)
* [Usage](#Usage)
* [Tutorial](#Tutorial)
* [Citation](#Citation)
## Datasets
We provide an easy access to the used datasets in the [synapse](https://www.synapse.org/#!Synapse:syn52559388/files/).
## Installation
To reproduce **SANGO**, we suggest first create a conda environment by:
~~~shell
conda create -n SANGO python=3.8
conda activate SANGO
~~~
and then run the following code to install the required package:
~~~shell
pip install -r requirements.txt
~~~
and then install [PyG](https://pytorch-geometric.readthedocs.io/en/latest/install/installation.html) according to the CUDA version, take torch-1.13.1+cu117 (Ubuntu 20.04.4 LTS) as an example:
~~~shell
pip install torch_geometric
pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-1.13.1+cu117.html
~~~
## Usage
### data preprocessing
In order to run **SANGO**, we need to first create anndata from the raw data.
The h5ad file should have cells as obs and peaks as var. There should be at least three columns in `var`: `chr`, `start`, `end` that indicate the genomic region of each peak. The h5ad file should also contain two columns in the `obs`: `Batch` and `CellType` （reference data）, where `Batch` is used to distinguish between reference and query data, and `CellType` indicates the true label of the cell.
Notice that we filter out peaks accessible in < 1% cells for optimal performance.
### Stage 1: embeddings extraction
The processed data are used as input to CACNN and a reference genome is provided to extract the embedding incorporating sequence information:
~~~shell
# Stage 1: embeddings extraction
cd SANGO/CACNN
python main.py -i ../../preprocessed_data/reference_query_example.h5ad \ # input data(after data preprocessing)
-g mm9 \ # reference genome
-o ../../output/reference_query_example # output path
~~~
Running the above command will generate three output files in the output path:
* `CACNN_train.log`: recording logs during training
* `CACNN_best_model.pt`: storing the model weights with the best AUC score during training
* `CACNN_output.h5ad`: an anndata file storing the embedding extracted by CACNN.
### Stage 2: cell type prediction
~~~shell
# Stage 2: cell type prediction
cd ../GraphTransformer
python main.py --data_dir ../../output/reference_query_example/CACNN_output.h5ad \ # input data
--train_name_list reference --test_name query \
--save_path ../../output \
--save_name reference_query_example
~~~
Running the above command will generate three output files in the output path:
* `model.pkl`: storing the model weights with the best valid loss during training.
* `embedding.h5ad`: an anndata file storing the embedding extracted by GraphTransformer. And `.obs['Pred']` saves the results of the prediction.
## Tutorial
### Tutorial 1: Cell annotations within samples (LargeIntestineB_LargeIntestineA)
1. Install the required environment according to [Installation](#Installation).
2. Create a `data` folder in the same directory as the 'SANGO' folder and download datasets from [LargeIntestineA_LargeIntestineB.h5ad](https://www.synapse.org/#!Synapse:syn52559388/files/).
3. Create a folder `genome` in the ./SANGO/CACNN/ directory and download [mm9.fa.h5](https://www.synapse.org/#!Synapse:syn52559388/files/).
4. For more detailed information, run the tutorial [LargeIntestineB_LargeIntestineA.ipynb](LargeIntestineB_LargeIntestineA.ipynb) for how to do data preprocessing and training.
### Tutorial 2: Cell annotations on datasets cross platforms (MosP1_Cerebellum)
1. Install the required environment according to [Installation](#Installation).
2. Create a `data` folder in the same directory as the 'SANGO' folder and download datasets from [MosP1_Cerebellum.h5ad](https://www.synapse.org/#!Synapse:syn52559388/files/).
3. Create a folder `genome` in the ./SANGO/CACNN/ directory and download [mm10.fa.h5](https://www.synapse.org/#!Synapse:syn52559388/files/).
4. For more detailed information, run the tutorial [MosP1_Cerebellum.ipynb](MosP1_Cerebellum.ipynb) for how to do data preprocessing and training.
### Tutorial 3: Cell annotations on datasets cross tissues (BoneMarrowB_Liver)
1. Install the required environment according to [Installation](#Installation).
2. Create a `data` folder in the same directory as the 'SANGO' folder and download datasets from [BoneMarrowB_Liver.h5ad](https://www.synapse.org/#!Synapse:syn52559388/files/).
3. Create a folder `genome` in the ./SANGO/CACNN/ directory and download [mm9.fa.h5](https://www.synapse.org/#!Synapse:syn52559388/files/).
4. For more detailed information, run the tutorial [BoneMarrowB_Liver.ipynb](BoneMarrowB_Liver.ipynb) for how to do data preprocessing and training.
### Tutorial 4: Multi-level cell type annotation and unknown cell type identification
1. Install the required environment according to [Installation](#Installation).
2. Create a `data` folder in the same directory as the 'SANGO' folder and download datasets from [BCC_TIL_atlas.h5ad, BCC_samples.zip, HHLA_atlas.h5ad](https://www.synapse.org/#!Synapse:syn52559388/files/).
3. Create a `genome` folder in the same directory as the 'SANGO' folder and download [GRCh38.primary_assembly.genome.fa.h5](https://www.synapse.org/#!Synapse:syn52559388/files/).
4. For more detailed information, run the tutorial [tumor_example.ipynb](tumor_example.ipynb) for how to do data preprocessing and training.
## Citation
If you find our codes useful, please consider citing our work:
~~~bibtex
@article{zengSANGO,
title={Deciphering Cell Types by Integrating scATAC-seq Data with Genome Sequences},
author={Yuansong Zeng, Mai Luo, Ningyuan Shangguan, Peiyu Shi, Junxi Feng, Jin Xu, Weijiang Yu, and Yuedong Yang},
journal={},
year={2023},
}
~~~