scMamba: A Pre-Trained Model for Single-Nucleus RNA Sequencing Analysis in Neurodegenerative Disorders

This repository is the official pytorch implementation of "scMamba: A Pre-Trained Model for Single-Nucleus RNA Sequencing Analysis in Neurodegenerative Disorders".
The code was modified from [this repo], and some parts were modified from [this repo]

"scMamba: A Pre-Trained Model for Single-Nucleus RNA Sequencing Analysis in Neurodegenerative Disorders",
Gyutaek Oh, Baekgyu Choi, Seyoung Jin, Inkyung Jung, and Jong Chul Ye,
arXiv preprint [Paper]

Requirements

The code is implented in Python 3.8 with below packages.

torch               1.13.0
pytorch-lightning   1.8.6
numpy               1.24.4
scipy               1.10.1
pandas              2.0.3
scanpy              1.9.3
anndata             0.8.0

Checkpoint

If you want to use pre-trained model checkpoint, please contact [okt0711@gmail.com].

Usage

You can pre-train your model from scratch by running with:

python -m train wandb=null experiment=pretrain

You can fine-tune pre-trained model for downstream tasks by running with:

python -m train wandb=null experiment=[downstream task]

[downstream task] can be celltype, subtype, subcluster, doublet, and imputation.