README.md

metadata

title: ARGnet UI
emoji: 🧬
colorFrom: indigo
colorTo: pink
sdk: gradio
sdk_version: 4.14.0
app_file: app.py
pinned: false
license: mit

ARGNet

A deep nueral network for robust identification and annotation of antibiotic resistance genes.

The input can be long amino acid sequences(full length/contigs), long nucleotide sequences, short amino acid reads (30-50aa), short nucleotide reads (100-150nt) in fasta format. If your input is short reads you should assign 'argnet-s' model, or if your input is full-length/contigs you should assign 'argnet-l' to make the predict.

Installation

clone the program to your local machine:
git clone https://github.com/patience111/ARGNet

1. Setting up environment

1.1 Installation with conda

1.1.1 For CPU inference, you could install the program with conda YAML file in the installation directory with the following commands:

cd ./installation 
conda env create -f ARGNet-CPU.yml -n ARGNet-cpu
conda activate ARGNet-cpu

(This was tested on Ubuntu 16.04, 20.04; Windows 10, macOS(14.1.1))

1.1.2 For GPU inference, you could install the program with conda YAML file in the installation directory with the following commands:

cd ./installation
conda env create -f ARGNet-GPU.yml -n ARGNet-gpu
conda activate ARGNet-gpu

(This was tested on Ubuntu 16.04, cuda 10.1, Driver Version: 430.64)

1.2 Or, if you prefer installing dependencies manually, you might find this information useful:
The program was tested with the following package version, you can install exactly the same version or other compatible versions.

Biopython:  1.79
tensorflow:  2.2.0 
cuda: 10.2 (for GPU using)
cudnn: 7.6.5.32 (for GPU using)
numpy: 1.18.5
scikit-learn: 0.24.1
tqdm: 4.56.0

2. Getting trained models
cd ./model bash get-models.sh Quickstart Guide

for long sequences

python argnet.py --input input_path_data  --type aa/nt --model argnet-l  --outname output_file_name

for short reads

python argnet.py --input input_path_data --type aa/nt --model argnet-s --outname output_file_name

general options:
--input/-i                 the test file as input
--type/-t                  molecular type of your test data (aa for amino acid, nt for nucleotide)
--model/-m             the model you assign to make the prediction (argnet-l for long sequences, argnet-s for short reads)
--outname/-on        the output file name

optional arguments:
-h, --help show this help message and exit


-i INPUT, --input INPUT
the test data as input

-t {aa,nt}, --type {aa,nt}
molecular type of your input file

-m {argnet-s,argnet-l}, --model {argnet-s,argnet-l}
the model to make the prediction

-on OUTNAME, --outname OUTNAME
the name of results output


Example

if we predict the long amino acid contigs by using ARGNet-L model, we could use command line (if you are in ARGNet dirctory):

python3 ./scripts/argnet.py -i ./tests/aa/long/arg100p.fasta -t aa -m argnet-l -on argnet_lsaa_test.txt

output will be like and saved in the results folder:

the first column test_id is the sequence label of the test sequnece.
the second column ARG_prediction is the "ARG" or "non-ARG" prediction of the input sequence.
the third column resistance_category is the classifition of the 36 antibiotics categories of the input sequence resisting to.
the last column probability is the classifition probability of the antibiotic predition of the input sequence by the model.

Contribute

If you'd like to contribute to ARGNet, check out https://github.com/patience111/argnet
Hope you enjoy ARGNet journey, any problem please contact [email protected]