---
license: apache-2.0
tags:
- biology
size_categories:
- 10M<n<100M
---
# Mouse-Genecorpus-20M

Write: Keita Ito

## Abstract
This repository contains the datasets of mouse-Genecorpus-20M and evaluation datasets to analysis. Mouse-Genecorpus-20M is large-scale pretraining corpus comprised of ~21 million mosue single cell transcriptomes from a broad range of tissues from publicly available data. This corpus is used for pretraining mouse-Geneformer, a pretrained transformer model that enables context-aware predictions in settings with limited data in network biology.

<!-- PLOS Genetics の URL を載せる-->
<!-- [[PLOS Genetics](hhttps://journals.plos.org/plosgenetics/...)]-->

<!-- bioRxiv の URL を記載-->
[[bioRxiv](https://www.biorxiv.org/content/10.1101/2024.09.09.611960v1)]

## Citation
If you find this repository is useful. Please cite the following references.

<!-- PLOS Genetics の bibtex を載せる -->

<!--bioRxiv の bibtex を載せる -->
```bibtex
@article{Ito2024.09.09.611960,
   author = {Ito, Keita and Hirakawa, Tsubasa and Shigenobu, Shuji and Fujiyoshi, Hironobu and Yamashita, Takayoshi},
   title = {Mouse-Geneformer: A Deep Leaning Model for Mouse Single-Cell Transcriptome and Its Cross-Species Utility},
   journal = {bioRxiv},
   year = {2024},
   URL = {https://www.biorxiv.org/content/early/2024/09/13/2024.09.09.611960}
}
```

## How to use
<!-- Python program to use this dataset (Python version:3.8.10) : -->
<!-- ```bash
# How to use mosue-Genecorpus-20M dataset
from datasets import load_dataset
dataset = load_dataset("MPRG/Mouse-Genecorpus-20M/MLM-re_All_mouse_tokenize_dataset.dataset")
```
or -->
<!-- ```bash
$ git clone https://huggingface.co/datasets/MPRG/Mouse-Genecorpus-20M
``` -->
You need to download datasets on your disk you want to use.
After doing, you program in python as follow to use its :
```bash
from datasets inport load_from_disk
dataset = load_from_disk("/path/to/mouse-Genecorpus-20M/MLM-re_All_mouse_tokenize_dataset.dataset/")
```