deeparg

name	deeparg
description	Deep-learning prediction of antibiotic resistance genes from metagenomic data. Use this skill when working with deeparg: installing, choosing parameters, running commands, interpreting outputs, or troubleshooting. Upstream: https://github.com/gaarangoa/DeepARG. For routing across mining tasks use microbial-mining. For stage order use metagenomics-workflow.
license	MIT
category	analysis-tools
tags	["AMR","ARG","genome-mining"]
upstream	https://github.com/gaarangoa/DeepARG
stage	mining

DeepARG

Upstream: DeepARG

Citation

Arango-Argoty, G. et al. DeepARG: a deep learning approach for predicting antibiotic resistance genes from metagenomic data. Microbiome 6, 23 (2018). https://doi.org/10.1186/s40168-018-0401-z

See also docs/references.md.

Analytical thinking

Complement to RGI/CARD for ARG mining from reads or genes. Model-based predictions need careful thresholds; always cross-check critical ARGs with RGI and report both DB/model versions.

How to run

conda install -c bioconda deeparg  # if packaged; else follow upstream docs

# follow upstream install; example gene mode:
deepARG --align --type prot --genes genes.faa --model LS --out temp/deeparg/${id}

Key parameters

| mode | reads vs genes | | | model | LS/SS | See docs |

Decision notes

• Run mining on QC-filtered genomes/MAGs (or contigs for MGE/virus screens)
• Computational hits are hypotheses — state validation limits in Methods
• Pin software + database releases

Related skills

rgi · genomad · humann · microbial-mining · tool-selection · metagenomics-workflow