- Notifications
You must be signed in to change notification settings - Fork1
The GAN model for designing AMP
NotificationsYou must be signed in to change notification settings
lsbnb/amp_gan
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
The GAN model to generate AMPs/ AVPs and other peptides.
To clone the package and install dependency packages:
git clone https://github.com/lsbnb/amp_gan.gitcd amp_ganpip3 install -r requirements.txt- Call the library:
#The path of input filefromamp_gan.trainimportmainastrain_ganfasta_path=""#The folder to save resultoutout_root=""train_gan(fasta_path,outout_root,batch_size=8,step=10,epoch=100)
- Call the script:
python3 amp_gan/train.py --f$fasta_path -o$outout_root --b 8 --s 10 --e 100
The detail settings for using train.py would be shown by using the follwoing codes
python3 amp_gan/train.py -h
There are two example to be used.
- Jupyter-notebook example is located in example/example.ipynb.
- Script example is located in example/example.sh.
If you find AMP-GAN useful, please consider citing:Discovering Novel Antimicrobial Peptides in Generative Adversarial Network
@article {Lin2021.11.22.469634,author = {Lin, Tzu-Tang and Yang, Li-Yen and Wang, Ching-Tien and Lai, Ga-Wen and Ko, Chi-Fong and Shih, Yang-Hsin and Chen, Shu-Hwa and Lin, Chung-Yen},title = {Discovering Novel Antimicrobial Peptides in Generative Adversarial Network},elocation-id = {2021.11.22.469634},year = {2021},doi = {10.1101/2021.11.22.469634},publisher = {Cold Spring Harbor Laboratory},URL = {https://www.biorxiv.org/content/early/2021/11/23/2021.11.22.469634},eprint = {https://www.biorxiv.org/content/early/2021/11/23/2021.11.22.469634.full.pdf},journal = {bioRxiv}}