AtSubP 2.0: Deep neural network based Arabidopsis thaliana Subcellular Localization Prediction

AtSubP 2.0 predicts the deep learning-based subcellular localization(s) of Arabidopsis thaliana. Currently it performs a multi-level prediction. First it differentiate between single and dual location proteins. Then it differentiate between 12 different single localizations: 'Vacuole', 'Cytoplasm', 'Peroxisome', 'Golgi', 'Endoplasmic', 'Memebrane', 'Secreted', 'Cell', 'Mitochondria', 'Endosome', 'Plastid','Nucleus'. Additionally, it differentiate predicted Membrane loclization into 2 different categories: Single-Pass, Multi-Pass. Similarly dual localizations are classified into 9 different localizations: 'Cell-memebrane_Membrane', 'Cytoplasm_Membrane','ER_Membrane','Golgi_Membrane','Membrane_Mitochondria','Membrane_Plastid','Plastid_Mitochondria','Nucleus_Chromosome','Nucleus_Cytoplasm'.

Input

The AtSubP 2.0 server requires amino acid sequence(s) in fasta format, and does not support nucleic acid sequences.

Prediction approach

The AtSubP2 can be excuted using two different approaches using different models: Sensitive and Fast

The 'Sensitive' approach uses Tripeptide amino acid composition. This model provides a more accurate prediction at the cost of longer computation time. This approach is useful in annotationg small number of proteins with high-quality prediction.

Development Environment and Prerequisite

This source code was developed in Linux, and has been tested on Linux and OS X. The only prerequisite is to have Python 3.7 or above installed.

Installation

The installation of AtSubP 2.0 can be done in two ways:

1. Create a dedicated miniconda3 environment

    Download AtSubP 2.0 from:
        
        https://bioinfo.usu.edu/AtSubP2/download/AtSubP-2.0.tar.gz

    Download the Miniconda installer: 
        
        (https://docs.conda.io/en/latest/miniconda.html#linux-installers)

    Extract the downloaded file:

        tar -xvzf AtSubP-2.0.tar.gz

    cd AtSubP-2.0
    
    Create and activate a conda environment

        conda env create -f environment.yml

        conda activate AtSubP2

        pip3 install .

2. Intall using system Python3

    Download AtSubP 2.0 from: 
        https://bioinfo.usu.edu/AtSubP2/download/AtSubP-2.0.tar.gz
        
    Extract the downloaded file:

        tar -xvzf AtSubP-2.0.tar.gz

    cd AtSubP-2.0

        pip3 install .

Example

Run AtSubP2

python AtSubP2.py -i ./example/test.fasta -o output -m fast

Queries and Contact

Written by Naveen Duhan (naveen.duhan@usu.edu),

Kaundal Bioinformatics Lab, Utah State University,

Released under the terms of GNU General Public Licence v3

In case of technical problems (bugs etc.) please contact Naveen Duhan naveen.duhan@usu.edu.

For any Questions on the scientific aspects of the AtSubP 2.0 method please contact:

Rakesh Kaundal, rkaundal@usu.edu

Naveen Duhan, naveen.duhan@usu.edu