Skip to content

ShirleyWISiu/xDeep-AcPEP

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

xDeep-AcPEP

Deep Learning Method for Anticancer Peptide Activity Prediction through Convolutional Neural Network and Multi-Task Learning

In this repository:

  1. We provide python scripts for reproducing the experiments of 5-folds cross-validation model comparison.
  2. We provide our final production models for peptide activity prediction.

Requirements

  • Anaconda 4.7.0
  • Python 3.6.9
  • Scikit-learn 0.21.3
  • Pytorch 1.2.0 with CUDA 10.0
  • Scipy 1.4.1
  • Pandas 1.0.2
  • Numpy 1.18.1
  • Openpyxl 3.0.6

Model and Data

  1. Models and data used for reproducing experiments are available at: Here
  2. Final production models for peptide activity prediction are available at: ./prediction/model/

Run the script

1. Reproducing Experiments

The script is located in model_comparison_CV folder

python reproduce.py -mo <model_folder_path> -da <data_folder_path> -o <output_folder_path>

Example:

python reproduce.py -mo ./model/ -da ./data/ -o ./

2. Final production model prediction

The script is located in prediction folder

python prediction.py -t <tissue_type> -m <model_folder_path> -d <fasta_file_path> -o <output_folder_path>

where:
<tissue_type> could be selected from breast, cervix, colon, lung, prostate and skin.

Example:

python prediction.py -t breast -m ./model/ -d ./test_breast.fasta -o ./result/

Note: The input peptide data must in the form of the following FASTA format.

>AmphiArc1
KWVKKVHNWLRRWIKVFEALFG
>AmphiArc2
KIFKKFKTIIKKVWRIFGRF
>Gradient2
AWLKRIKKFLKALFWVWVW 
>AmphiArc3
AFRHSVKEELNYIRRRLERFPNRL

References

  1. We used iFeature to extract all peptide features. (Github, Paper)

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%