Skip to content

Latest commit

 

History

History
49 lines (31 loc) · 2.14 KB

File metadata and controls

49 lines (31 loc) · 2.14 KB

CICERON

Classification of bIoaCtive pEptides fRom micrObial fermeNtation

CICERON is a script for the functional classification of bioactive peptides specifically trained on BPs obtained from microbial fermentation. Starting from peptide sequences, nine binary classifiers assign a functional prediction to the bioactive peptide. The functional classes that are predicted are the following: Antidiabetic, Antihypertensive, Antimicrobial, Antioxidant, Cardiovascular, Celiac disease, Immunomodulatory, Neuropeptides and Opiopid. For more information see the following paper: Classification of bioactive peptides: A systematic benchmark of models and encodings, Comput Struct Biotechnol J. 2024 May 23; 23:2442–2452. doi: 10.1016/j.csbj.2024.05.040

Installation

Create a conda environment with Python 3.9 and packages:

conda create -n ciceron python=3.9 pandas biopython scikit-learn=1.1.2 numpy tensorflow=2.15 pip install tensorflow-addons

To enter the environment:

conda activate ciceron

Clone this git repository

git clone https://github.com/BizzoTL/CICERON/

Usage

Fasta protein files of interest must be present in the input folder. From the cloned git folder, launch the following command:

python3 model_prediction.py -i input_folder -o output_folder -s suffix

The results will be displayed in the output folder with the same name as the input file and the suffix chosen(facultative).

Example:

Fasta file "Peptides.fasta" in the folder peptide_inputs:

>Peptide_1
ATGIQPP
>Peptide_2
MNIPYPYP

To run CICERON to predict the functional classification of these peptides use the following command line from the CICERON folder:

python3 model_prediction.py -i path/to/peptide_inputs -o path/to/predicted_peptides -s predicted

This will return a file called "Peptides_predicted.csv" inside the folder "predicted_peptides".

Example of output table:

Screenshot from 2023-08-03 15-03-29

Data availability

All the datasets are included in the repository, as the scripts needed to generate and train the models used in this work.